Kinetics and mechanism of the chromic oxidation of 3-O-methyl-d-glucopyranose

Article abstract of DOI:10.1016/j.poly.2009.01.003

The oxidation of 3-O-methyl-d-glucopyranose (Glc3Me) by Cr^VI in acid medium yields Cr^III, formic acid and 2-O-methyl-d-arabinose as final products when a 50-times or higher excess of Glc3Me over Cr^VI is used. The redox rea

Full text of DOI:10.1016/j.poly.2009.01.003

Polyhedron 28 (2009) 1049–1056
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Kinetics and mechanism of the chromic oxidation of 3-O-methyl-D-glucopyranose
a,
María Inés Frascaroli a, Juan Manuel Salas-Peregrin b, Luis F. Sala a, , Sandra Signorella
*
*
a Instituto de Química Rosario – CONICET, Universidad Nacional de Rosario, Suipacha 531, S2002LRK Rosario, Argentina
b Departamento de Química Inorgánica, Facultad de Ciencias, Universidad de Granada, Fuentenueva s/n, 18071 Granada, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The oxidation of 3-O-methyl-
D
-glucopyranose (Glc3Me) by CrVI in acid medium yields CrIII, formic acid
and 2-O-methyl-D
-arabinose as final products when a 50-times or higher excess of Glc3Me over CrVI is
Received 26 September 2008
Accepted 7 January 2009
Available online 21 March 2009
used. The redox reaction takes place through the combination of CrVI ? CrIV ? CrII and CrVI ? CrIV ? CrIII
pathways. Intermediacy of free radicals and CrII in the reaction was demonstrated by the observation of
2+
induced polymerization of acrylamide and detection of CrO2 formed by reaction of CrII with O2. Inter-
mediate oxo-CrV–Glc3Me species were detected by EPR spectroscopy. In 0.3–0.5 mol/L HClO4, intermedi-
ate CrV rapidly decompose to the reaction products, while, at pH 5.5–7.5, where the redox processes are
very slow, five-coordinate CrV bis-chelates of the pyranose and furanose forms of Glc3Me remain more
than 15 h in solution. The C1–C2 bond cleavage of Glc3Me upon reaction with CrVI distinguishes this
derivative from glucose, which is oxidized to gluconic acid.
Keywords:
Chromate
3-O-methylglucose
Redox chemistry
Kinetics
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
study the reaction of CrVI with 3-O-methyl-
D-glucopyranose
(Glc3Me), and show that replacement of C3–OH by C3–OMe favors
the C1–C2 bond cleavage over the formation of the aldonic acid,
thus evidencing the ‘‘noninnocent” role of the OMe group at C3
of the sugar in the reaction with CrVI.
Compounds of CrVI are dangerous pollutants used and gener-
ated by industries [1]. Due to the carcinogenic and toxic effects
of CrVI [2,3], and the fact that intermediate oxidation states gener-
ated during the reduction of CrVI to CrIII may be implied in the
mechanism of Cr–induced cancers [4], studies on the kinetics and
mechanism of the oxidation of biologically relevant reducing
agents by CrVI is of interest to both biochemists and inorganic
chemists [4–6]. Compounds such as polyalcohols and hydroxy-
carboxylic acids, are effective as nonenzymatic reductants (at low
pH) and can stabilize the labile oxidation states of chromium
[4,7–9]. Therefore, polyoxygenated compounds may play an
important role in chromate toxicity. Because of their potential bio-
logical and ecological relevance there is an increasing interest in
studying the stabilization/reduction of hypervalent chromium by
saccharides. Previous studies on the oxidation of aldoses [10,11],
2-deoxy-aldoses [11–13] and 6-deoxy-aldoses [10,14] by CrVI
showed that the C1–OH hemiacetalic function reacts faster than
the primary or any of the secondary alcoholic groups, to yield the
corresponding aldonic acid as the only reaction product. These
works provided information on the role of C2–OH and, in lesser ex-
tent, C6–OH, to act as chelating sites of CrVI to form chromate ester
intermediates that retard the oxidation of C1–OH, and established
that an equatorial 3-OH group has none or negligible influence on
the reaction kinetics and mechanism. In the present work, we
2. Experimental
2.1. Materials
3-O-methyl-D-glucopyranose (Sigma, 98%), potassium dichro-
mate (Mallinckrodt), sodium chromate (Merck), perchloric acid
(A.C.S. Baker), phosphoric acid (Anedra P.A.), sulfuric acid (Merck
P.A.), glutathione (Sigma, 98–100%, reduced form), acrylamide
(Sigma, >99%), N,N-dimethylformamide (Fisher Scientific), sodium
hydroxide (Cicarelli, P.A.) were used without further purification.
Aqueous solutions were prepared in milliQ-deionized water
(HPLC quality). For experiments performed at constant ionic
strength (I) and different [H+], mixtures of NaClO4 and HClO4 solu-
tions were prepared. The concentration of stock solutions of HClO4
was determined by titration using standard analytical methods. 4-
(2-Hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer
was used for experiments at pH 7.5. Acetate buffer was used for
experiments at pH 5.5. Na[CrVO(ehba)2] ꢀ H2O and Na2[CrIVO(eh-
ba)2] were synthesized from 2-ethyl-2-hydroxybutanoic acid
(ehba, Aldrich, 99%) according to literature methods [15,16].
Caution: CrVI compounds are human carcinogens, and CrV com-
plexes are mutagenic and potential carcinogens [17]. Contact with
skin and inhalation must be avoided. Acrylamide is a carcinogen
and must be handled in a well-ventilated fume hood [18].
* Corresponding authors. Tel./fax: +54 341 4350214.
E-mail addresses: sala@iquir-conicet.gov.ar (L.F. Sala), signorel@infovia.com.ar
(S. Signorella).
0277-5387/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2009.01.003

1050
M.I. Frascaroli et al. / Polyhedron 28 (2009) 1049–1056
2.2. Methods
2.2.4. Spectrophotometric measurements
Kinetic measurements were performed by monitoring absor-
bance changes using a Jasco V-550 UV/VIS spectrophotometer with
a fully thermostated cell compartment ( 0.2 °C). The reactions
were followed under pseudo-first-order conditions, at 33 °C, using
at least a 50-fold molar excess of Glc3Me over CrVI and different
[HClO4]. Reactant solutions were previously thermostated at
33 °C and transferred into a 1-cm pathlength cell immediately after
mixing.
2.2.1. Product analysis
HPLC was employed to detect the reaction products under the
conditions used in the kinetic measurements (excess of Glc3Me
over CrVI). The chromatograms were obtained on a KNK-500 A
chromatograph provided with a 7125 HPLC pump. The separation
was carried out on
a
Supelcogel C-610H HPLC column
(300 ꢁ 7.8 mm, Bio-Rad Laboratories) using 0.1% P/V H3PO4 as elu-
ent and a flow rate of 0.5 mL/min, at 30 °C. The effluent was mon-
itored with refractive index (ERC-7522, ERMA INC) and UV (115 UV
Gilson, k = 220 nm) detectors. The [H+] of the reaction mixtures
(Glc3Me:CrVI ratios of 25:1, 50:1 and 100:1) was adjusted to
0.2 mol/L by addition of 2 mol/L HClO4 and then the samples were
The disappearance of CrVI was followed spectrophotometrically
at 350 nm until at least 80% conversion. In the kinetic measure-
ments, the initial concentration of CrVI and I were kept constant
at 8.0 ꢁ 10ꢂ4 mol/L and 1.02 mol/L, respectively, while [HClO4]
was varied from 0.15 to 1.02 mol/L, at various [Glc3Me].
The experimental pseudo-first-order rate constants (kexp),
determined from the absorbance versus time curves at 350 nm,
were deduced from multiple determinations and were within
5% of each other. The first-order dependence of the rate upon
[CrVI] was verified in a set of experiments where the [CrVI]0 was
varied between 4.0 ꢁ 10ꢂ4 and 8.0 ꢁ 10ꢂ4 mol/L but T, [Glc3Me]0,
[H+] and I were kept constant.
filtered through a 0.2
chromatographic system.
Standard samples of 2-methyl-
methyl- -gluconic acid (Glc3MeA) – prepared following pub-
lm membrane prior to injection into the
D-arabinose (Ara2Me), 3-
D
lished procedures [19,20] – and formic acid were prepared indi-
vidually and the chromatographic retention times determined
separately.
Chromatograms of the reaction mixtures showed two peaks
with retention times (tR) of 130 4400 and 170 1100, in addition to
the peak of unreacted Glc3Me at tR = 110 5500. The retention times
of the new peaks are coincident with those of the standard solu-
tions of Ara2Me and formic acid, respectively. No peak correspond-
ing to Glc3MeA was detected in the reaction mixture. Furthermore,
co-chromatography of a CrVI/Glc3Me reaction mixture and Ara2Me
or formic acid resulted in the increase of peaks at 130 4400 or 170
1100, respectively. Additionally, the presence of HCO2H acid was
confirmed through the HgCl2 reaction [21].
The formation of CrIII was monitored by following changes in
the 570 nm absorption band. In these experiments the [CrVI]o
was kept constant at 8.0 ꢁ 10ꢂ3 mol/L, (Glc3Me:CrVI ratio of 50)
and the [H+] was varied between 0.40 and 1.02 mol/L. Rate con-
stants obtained at this wavelength are in agreement with those
calculated from data at 350 nm for the same experimental
conditions.
Cr(VI) esters were investigated by UV–Vis spectrophotometry in
the 350–500 nm region in which chromate esters show character-
istic absorption bands. Reactions were performed at pH 3.52,
where the redox reaction is slow enough to enable the observation
of the ester formation. The instrument was zeroed to an arrange-
ment of the reference and sample beams passing through matched
cuvettes, both containing 6 ꢁ 10ꢂ4 mol/L CrVI at pH 3.52. The solu-
tion in the sample cell was replaced with the reaction solution con-
taining 6 ꢁ 10ꢂ4 mol/L CrVI and 0.30–0.60 mol/L Glc3Me at pH
3.52, I = 1.02 M, and T = 33 °C. Spectra obtained within 40 min after
mixing revealed a distinctive absorption at 373 nm.
For the Glc3Me to CrVI ratios used in the kinetic studies (higher
than 50:1), carbon dioxide was never detected as a reaction
product.
2.2.2. Glc3Me stability
The stability of Glc3Me under conditions used in the kinetic
studies was checked by monitoring the optical rotation changes
on a JENA polarimeter with thermostated 20-cm tubes using the
sodium D line. Reactant solutions were previously thermostated
and transferred into the cell immediately after mixing. The opti-
cal rotations were recorded at different times after the prepara-
tion of the solutions and were identical to that of freshly
The possible formation of CrII was examined by UV–Vis spectro-
photometry in the 200–500 nm region in which CrO22+ show char-
acteristic absorption bands. This experiment was performed at
22 °C, by periodic scanning of solutions containing 0.21 mol/L
Glc3Me, 0.049 mmol/L CrVI and 1.26 mmol/L O2, in 0.84 mol/L
HClO4. Under these conditions, the CrVI absorption band at
350 nm decreased in intensity, while new peaks at 293 nm and
prepared Glc3Me solutions. The specific rotation of Glc3Me
33
was found to be [a] D = +54.1° 0.5° (c = 3.88 g/100 mL, 1.0 M
HClO4, t = 10 min to 30 h), a value similar to that measured
20
2+
for Glc3Me in aqueous solution: [
a
]
D = +55.7° 0.5° (c = 4.51
247 nm, characteristic of CrO2 [22], grew in.
g/100 mL).
The stability of the Glc3Me at 33 °C was also confirmed by HPLC
2.2.5. EPR measurements
after incubation of the solutions at fixed [H+].
The EPR spectra were obtained on a Bruker EMX spectrometer
operating at X-band frequencies (9–10 GHz). The microwave fre-
quency was generated with a Bruker 04 ER and measured with a
Bruker EMX 048T frequency meter. The magnetic field was mea-
sured with a Bruker EMX 035M NMR-probe gaussmeter. Spectra
were recorded as first derivatives of the microwave absorption in
1024 points at ambient temperature (20 1 °C). g-Values were
determined by reference to diphenylpicryhydrazyl radical
(g = 2.0036) as an external standard. Reactions were carried out
by mixing Glc3Me and K2Cr2O7 (ratio 49:1, [CrVI] = 0.025 mol/L)
in 0.3 and 0.5 mol/L HClO4; and by addition of Na[CrVO(ehba)2]
(1.0 mmol/L), or Na2CrO4 (1.0 mmol/L) + glutathione (1.0 mmol/
L), to solutions of Glc3Me (100 mmol/L) in acetate buffer
(0.1 mol/L, pH 5.5) or HEPES buffer (0.1 mol/L, pH 7.5). Stock solu-
tion of Na[CrVO(ehba)2] (0.1 mol/L) was prepared in N,N-dimethyl-
formamide. All of the EPR spectra were simulated using the
program PEST WINSIM [23], assuming 100% Lorentzian line shapes.
2.2.3. Polymerization test
Polymerization of acrylamide was investigated during the reac-
tion of Glc3Me with CrVI as a test for free radical generation. Acryl-
amide (0.6 g) was added to a reaction mixture containing K2Cr2O7
(3 ꢁ 10ꢂ3 mmol) and Glc3Me (0.3 mmol) in 2 mL of 0.5 mol/L
HClO4, at 33 °C. When the color of CrVI disappeared, 1 mL of the
mixture was diluted with 1 mL of methanol and a white polymer
precipitated. Control experiments showed that no polymerization
of acrylamide takes place under the experimental conditions with
either HCrJ4ꢂ or the Glc3Me alone. The possible reaction of CrV or
CrIV with acrylamide was tested with Na[CrVO(ehba)2] and Na2[CrI-
VO(ehba)2]. No precipitation occurred on mixing the CrV or CrIV
complexes with acrylamide under the same conditions as those
used in the CrVI + Glc3Me reaction.

M.I. Frascaroli et al. / Polyhedron 28 (2009) 1049–1056
1051
The spectral parameters for each CrV species were consistent with-
0.12
0.10
0.08
0.06
0.04
0.02
0.00
in all simulation, with maximum deviations in the giso values being
0.0001 units. In the simulations, values for 1H superhyperfine
coupling constant aH were included only where the aH value was
greater than the LW (line width) of the CrV species, since the signal
is not significantly affected where the aH value is 6LW.
3. Results and discussion
3.1. Products of the reaction of Glc3Me with CrVI
(b)
(a)
Under conditions used in the kinetic studies, excess of Glc3Me
over CrVI in acid medium, HCO2H and Ar2Me were identified by
HPLC as the oxidation products of Glc3Me, while EPR and UV–Vis
spectroscopies showed that aqueous CrIII was the final Cr species
in the reaction mixture. Thus, the reaction of Glc3Me and CrVI
can be written as in Scheme 1.
This result implies that, in this reaction, C1–C2 bond break pre-
vails over C1–H cleavage. A fact that differentiates the reactivity of
the 3-O-methyl derivative towards CrVI from glucose that yields
the corresponding aldonic acid as the only reaction product [10].
Aosrbance
300
350
400
450
500
550
600
λ (nm)
3.2. Detection of an intermediate CrVI-ester
Fig. 1. UV–Vis difference spectra of Glc3Me/CrVI solutions at pH 3.52, showing the
increasing band at 373 nm with increasing [Glc3Me]: (a) 0.30 and (b) 0.60 mol/L.
[CrVI] = 6.0 ꢁ 10ꢂ4 mol/L, I = 1.02 mol/L, T = 33 °C. Spectra taken 30 min after prep-
aration of the reaction mixture.
At pH 3.52, the redox reaction of CrVI with Glc3Me proceeds
very slowly, with negligible reduction of CrVI in the first 40 min.
Thus, at this pH, the ester formation can be distinguished clearly
from the electron transfer reaction. Differential UV–Vis spectra of
mixtures of CrVI and Glc3Me exhibit, immediately after mixing, a
distinctive absorption band with kmax = 373 nm consistent with
that ascribed to CrVI oxo-esters [24] that can be assigned to
Glc3Me–CrVI species. Continued scanning over 40 min showed no
further change in the spectra. The absorbance of this band in-
creased with the increasing concentration of Glc3Me when the
[Glc3Me] was varied at pH 3.5 (Fig. 1), probably as a result of a shift
toward the ester in the esterification equilibrium.
CrVI
0.10
3.3. Intermediacy of CrII
6'
Involvement of CrII in the mechanism of the oxidation of a num-
ber of alcohols, saccharides and hydroxyacids by CrIV and CrVI in
HClO4 has been demonstrated by conversion into CrO2 upon
15'
0.05
2+
30'
reaction with molecular oxygen [22,25–27]. At high [O2] and low
[CrVI] the reaction of CrII with O2 can compete successfully with
the reaction of CrII with CrVI and the autocatalytic consumption
of CrO22+ by CrII, and if formed, CrII should yield the CrO22+ product
[25,26,28,29]. We examined the presence of intermediate CrII in
the reaction of Glc3Me with CrVI, by monitoring the formation of
CrO22+, using [CrVI]0 low enough to avoid the CrVI + CrII competitive
Aosrbance
45'
75'
2 h
3 h
CrO22+
0.00
reaction.
A periodic scanning of the O2-saturated solution
(1.26 ꢁ 10ꢂ3 mol/L O2) of a CrVI + Glc3Me reaction mixture in
0.84 mol/L HClO4 over a period of 3 h, showed that the band at
350 nm, characteristic of CrVI, decreased in intensity while two
absorption bands at kmax = 247 and 293 nm appeared (Fig. 2). These
two bands are characteristic of CrO22+ formed as a long-lived inter-
mediate [22] that then slowly transforms into the final CrIII. Since
200
250
300
350
λ (nm)
400
450
500
2+
Fig. 2. Formation of CrO2 (kmax 293, 247) from the reaction between 0.21 mol/L
Glc3Me, 1.26 mmol/L O2 and 4.9 ꢁ 10ꢂ2 mmol/L CrVI, in 0.84 mol/L HClO4. T = 22 °C.
OH
HOH2C
HOH2C
HO
O
O
+ 3 HCO2H + 2 CrIII
+ 2 CrVI
OH
3
OMe
3
MeO
OH
HO
Scheme 1.

1052
M.I. Frascaroli et al. / Polyhedron 28 (2009) 1049–1056
CrO22+ can be exclusively formed by reaction of CrII with O2, and, in
turn, CrII had previously been demonstrated to form exclusively
through two-electron reduction of CrIV [26,28], our spectroscopic
results provide evidence that CrII and CrIV are involved in the redox
mechanism of the reaction between CrVI and Glc3Me.
the kinetic measurements, CrV–Glc3Me redox processes are fast
and, consequently, high modulation amplitude is required to ob-
serve CrV. With a modulation amplitude of 4 G, the EPR spectra
of solutions from the reaction between CrVI and 50-times excess
of Glc3Me in 0.3–0.5 mol/L HClO4, consisted of two signals: a major
one centered at giso1 1.9788, and an additional weak signal at giso2
1.9712 (Fig. 3) with relative intensity 610% of the main signal. The
high modulation amplitude used to observe CrV species at this [H+],
disabled the resolution of the signal superhyperfine (shf) pattern
(the 1H shf splitting (aH) for CrV–alcoholato species is usually <1
G [4,9]). Given that the g values of CrV species are very sensitive
to the co-ordination number and the nature of the donor groups
bound to CrV [4,9,30], an estimation of the intermediate CrV species
formed in the reaction of CrVI with Glc3Me was made based on giso
values of the EPR signals. The giso value of 1.9788 is typical of five-
coordinate oxochromate(V) complexes formed with O-donor li-
gands, while the value of 1.9712 is typical of a six-co-ordinate
oxo-CrV complex with two alcoholato donor sites and three water
molecules, and can be assigned to [Cr(O)(O1,O2-Glc3Me)(H2O)3]+ (I,
Fig. 4) [4,9,30]. The positive charge of this species is also consistent
with its appearance at high [H+] and it is possibly responsible for
the higher rates observed for the CrV–Glc3Me redox reactions at
pH < 1.
3.4. Detection of intermediate CrV
The most common means of detecting CrV complexes in solu-
tion is EPR spectroscopy, where strong isotropic signals are ob-
served at room temperature in X-band spectra. The reduction of
CrV by Glc3Me is [H+] dependent, and at the high [H+] used in
70
60
50
40
30
20
At pH > 3 the redox reaction between CrV with Glc3Me becomes
slow, CrV species remain in solution for longer periods of time and
the shf pattern of the CrV EPR signals can be resolved using a mod-
ulation amplitude lower than the 1H shf splitting. With this in
mind, CrV–Glc3Me complexes formed upon addition of 100-times
molar excess of Glc3Me to CrV generated in either the one-electron
reduction of CrVI by glutathione or the ligand-exchange reaction of
[CrVO(ehba)2] were investigated in the 5.5–7.5 pH range by EPR
spectroscopy. The CrV EPR signals are more intense at pH 5.5 than
at pH 7.5, probably because CrV disproportionation is slower at the
lower pH [31]. In this pH range, the EPR spectra taken at the begin-
ning of the reaction were dominated by a triplet at giso1 1.9792
(aH = 1.0 ꢁ 10ꢂ4 cmꢂ1), with a minor component at giso2 1.9794
(quintet, aH = 0.9 ꢁ 10ꢂ4 cmꢂ1). Fifteen minutes after mixing, the
two components at giso1 and giso2 were present in 71% and 29%,
Po-pakeight
10
3500
3520
50
3540
3560
G
0
0
100
150
200
250
(min)
t
Fig. 3. Peak-to-peak heights of CrV EPR signals vs. time. [CrVI] = 0.025 mol/L,
[Glc3Me] = 1.225 mol/L, [H+] = 0.3 mol/L, I = 1.02 mol/L, T = 20 °C. Inset: time evo-
lution of X-band EPR spectra from the reaction mixture in 0.3 mol/L HClO4.
m
= 9.766887, mod. ampl. = 4 G.
-
+
O
HOH2C
HO
O
O
CH2OH
OH
O
Cr
w
HOH2C
OH
O
O
O
O
O
w
w
O
O
Cr
OMe
OMe
OMe
II
I
-
O
O
Cr
O
HOH2C
HO
O
O
O
O
OH
OMe
MeO
OH
III
-
O
O
O
O
Cr
O
O
OH
O
HO
HO
MeO
OMe
OH
IV
Fig. 4. Structures of CrV complexes formed with Glc3Me.

M.I. Frascaroli et al. / Polyhedron 28 (2009) 1049–1056
1053
the proportion of the components varied and a third component
appeared. The EPR spectrum taken 1 h after mixing, shown in
Fig. 5b, could be deconvoluted into a triplet at giso1 1.9792
a
b
(aH = 1.0 ꢁ 10ꢂ4 cmꢂ1),
a
quintet at giso2 1.9794 (aH = 0.9 ꢁ
10ꢂ4 cmꢂ1) and a septuplet at giso3 1.9793 (aH = 1.0 ꢁ 10ꢂ4 cmꢂ1),
in 6:3:1 ratio. Other minor species were also present, but they
were not included in the simulations to avoid over-parameteriza-
tion. Based on the shf coupling, the three components could corre-
spond to [Cr(O)(cis-O1,O2Glc3Me)2] (II), [Cr(O)(cis-O1,O2-
Glc3Me)(O5,O6-Glc3Me-furanose)] (III) and [Cr(O)(O5,O6-Glc3Me-
furanose)2] (IV) (Fig. 4). At longer times, the proportion of the spe-
cies at giso1 decreased and the spectra were dominated by species
at giso2 and giso3. These five-coordinate CrV bis-chelates were still
observed 15 h after mixing. Thus, initially, CrV bis-chelate is
formed with CrV bound to the 1,2-cis-diolato moiety of the pyra-
nose form of Glc3Me (kinetic control); but, with time, it transforms
into bis-chelates with CrV bound to the 5,6-vic-diolato moiety of
the furanose form of the ligand.
3495
3500
3505
G
3510
3515
3.5. Rate studies
The kinetics of the reaction of Glc3Me with CrVI was examined
by monitoring the [CrVI] consumption under conditions of excess of
Glc3Me over CrVI, in the 0.15–1.02 mol/L HClO4 range. UV–Vis
spectra of the reaction mixtures showed an absorbance band at
350 nm and a shoulder at 420–500 nm, characteristic of CrVI in
acidic medium. At 350 nm, the absorbance versus time curves
exhibited a monotonic decrease and the rate constants were calcu-
lated from the absorbance changes at this wavelength. As stated
before, CrV species are formed in the chromic oxidation of Glc3Me,
and it is known that these species absorb at 350 nm and may
superimpose CrVI absorbance [35]. However, if CrV reacts faster
than CrVI and exists in solution in a sufficiently small concentra-
tion, changes in absorbance at 350 nm essentially reflect changes
in CrVI concentration. The relative values of the CrVI versus CrV
reduction rate constants obtained from the CrV EPR signal intensity
variation during the reaction course demonstrated this is the case
for the reaction studied here (discussed below).
The kinetic profiles at 350 nm could be adequately described by
a single exponential decay from which CrVI pseudo-first order rate
constants (kexp) were calculated. Table 1 summarizes values of kexp
for various concentrations of Glc3Me and HClO4. Plots of kexp ver-
sus [Glc3Me] at constant [H+] gave good straight lines (Fig. 6) from
which values of kh were determined (Table 1, right). Values of kh
showed quadratic dependence on [H+] (Fig. 6, inset), and the rate
constant k6, calculated by non-linear least-square fit of {kh, [H+]}
data pairs, was found to be (6.35 0.02) ꢁ 10ꢂ4 molꢂ3 L3 sꢂ1. Con-
sequently, the complete rate law for the CrVI consumption can be
expressed as in Eq. (1):
3500
3505
3510
3515
3520
3525
G
Fig. 5. Experimental (
) and simulated (
) X-band EPR spectra from mixtures
of 1 mmol/L glutathione + 1mmol/L CrVI and 100 mmol/L Glc3Me, pH 5.5; taken (a)
15 min (
m = 9.705912 GHz) and (b) 1 h (m = 9.728655) after mixing. Mod. ampl. = 0.2
G. T = 20 °C.
respectively (Fig. 5a). It is known that the EPR spectrum of a CrV-
diolato species of six-membered ring cis-diols yields a doublet,
since only one proton is in the plane of the unpaired electron den-
sity of the CrV ion. Consequently, the EPR spectral multiplicity of
bis-chelate CrV-diolato2 species formed between CrV and pyranosic
cis-diols exhibit a triplet with two (one from each chelate ring) car-
binolic protons coupled to the CrV electronic spin. In the present
case, the major component of the EPR signal can be attributed to
the bis-chelate [Cr(O)(cis-O1,O2-Glc3Me)2] (II, Fig. 4). This result
is in line with the reported higher ability of the cis- versus the
trans-diolato for binding CrV [4,9,32]. The minor component at giso2
1.9794 can be attributed to the bis-chelate formed with the pyra-
nose and furanose forms of Glc3Me: [Cr(O)(cis-O1,O2-
Glc3Me)(O5,O6-Glc3Me-furanose)] (III, Fig. 4), with four protons
coupled to the electronic spin of CrV. The assignment of the minor
component to [Cr(O)(cis-O1,O2-Glc3Me-furanose)2] was disre-
garded based on the aH values ꢂ aH values expected for CrV-diola-
to2 species of five-membered ring cis-diols are lower than found
here [33,34]. With time, the EPR spectral pattern slowly changed,
2
ꢂd½CrVIꢃ=dt ¼ k6½Hþꢃ ½Glc3Meꢃ½CrVIꢃ:
ð1Þ
Table 1
a
Experimental pseudo-first-order rate constants (kexp
)
for different concentrations of Glc3Me and HClO4.
[HClO4] (mol/L)
105 kexp (sꢂ1) for [Glc3Me] (mol/L)
0.04
0.08
0.12
0.16
0.20
104 kh (molꢂ1 L sꢂ1
)
0.15
0.31
0.46
0.61
0.77
0.92
1.02
1.94(1)
6.19(3)
12.4(1)
20.1(1)
29.9(3)
44.5(1)
53.2(1)
2.58(3)
8.23(9)
17.7(1)
31.0(2)
44.4(6)
63.6(9)
78.5(4)
2.94(3)
10.4(3)
22.0(4)
40.2(1)
59.2(1)
84.0(7)
104(1)
4.09(8)
14.4(1)
28.4(1)
49.9(9)
72.4(2)
106(1)
135(1)
2.0(1)
7.0(2)
3.12(1)
6.51(4)
10.9(2)
15.8(4)
22.8(3)
27.0(4)
14.3(3)
25.2(2)
36.7(3)
53.1(5)
66.4(6)
a
Mean values from multiple determinations. [CrVI
]
0 = 6 ꢁ 10ꢂ4 mol/L, T = 33 °C; I = 1.02 M, k350
.

1054
M.I. Frascaroli et al. / Polyhedron 28 (2009) 1049–1056
[H+]=1.02 mol/L
0.025
0.006
0.004
0.002
0.000
0.0014
0.0012
0.0010
0.0008
0.0006
0.0004
0.0002
0.0000
0.020
h
k
0.0 0.3 0.6 0.9 1.2
[H+] (mol/L)
CrV
CrVI
CrIII
0.015
[H+]=0.92 mol/L
[H+]=0.77 mol/L
0.010
k
[H+]=0.61 mol/L
0.005
[H+]=0.46 mol/L
0.000
[H+]=0.31 mol/L
[H+]=0.15 mol/L
0
50
100
150
200
250
t (min)
Fig. 7. Simulated kinetic profiles for Cr species. [Cr] calculated using
ka = 0.00496 minꢂ1 and kb = 0.1224 minꢂ1 obtained from EPR data in 0.3 mol/L
HClO4. [Cr]T = 0.025 mol/L.
0.00
0.03
0.06
0.09
0.12
0.15
0.18
0.21
[Glc3Me] (mol/L)
Fig. 6. Effect of [Glc3Me] on kexp at 33 °C, I = 1.02 mol/L, k = 350 nm and different
[H+]. Inset: dependence of kh on [H+].
pected for the free Cr3+ and then slowly decayed to the value cor-
responding to [Cr(H2O)6]3+ (409 nm,
e
= 17.7 molꢂ1 L cmꢂ1 and
572 nm,
e
= 15.2 molꢂ1 L cmꢂ1) [37], as shown in the inset of
The kinetic results were independently confirmed by following
the redox reaction by EPR spectroscopy. Fig. 3 shows a typical
curve for the growth and decay of the CrV EPR signal peak-to-peak
height as a function of time. The kinetic profiles could be ade-
quately fitted with the rate expression derived from a two-step
reaction sequence involving two consecutive first-order reactions
(Scheme 2 and Eq. (2)) [11,36]:
Fig. 8. This behavior suggests the formation of an intermediate
CrIII–Glc3Me complex, which then hydrolyzes to the final product
[30,38]. EPR spectra taken at the end of the reaction confirm that
Cr3+ is the final Cr species in the reaction of CrVI with excess
Glc3Me in acid medium.
In order to determine the rate of CrIII formation relative to that
of the CrVI consumption, the formation of CrIII was followed at 570
nm, in the presence of excess of Glc3Me in the 0.40–1.02 mol/L
HClO4 concentration range. Under these conditions, the first order
dependence of the rate upon [CrVI] was verified. For
[Glc3Me]0 = 0.4 mol/L, values of kexp calculated at this wavelength
EPRarea ¼ Akafexpðꢂ2katÞ ꢂ expðꢂkbtÞg=ðkb ꢂ 2kaÞ;
ð2Þ
where A depends on the spectrometer settings, and ka and kb refer
to the rate of disappearance of CrVI and CrV, respectively, and were
evaluated from a non-linear least-squares fit of Eq. (2). In the 0.3 to
0.5 mol/L HClO4 range, the calculated values of ka were more than
ten-times lower than kb. For conditions in Fig. 3, ka = 0.00496 minꢂ1
and kb = 0.1224 minꢂ1. The fact that kb > ka implies that the slow re-
dox step involves the reduction of CrVI. The simulation of the kinetic
profiles employing ka and kb values obtained from EPR data in
0.3 mol/L HClO4 (Fig. 7), shows that the maximal [CrV] represents
3% of the total Cr in solution. These results confirm that [CrV] is
low throughout the reaction and should not interfere absorbance
at 350 nm, which essentially reflects changes in [CrVI]. Rate con-
stants calculated by this technique were consistent with those ob-
tained from the spectrophotometric measurements at 350 nm,
under the same experimental conditions. For 1.225 mol/L Glc3Me
and 0.3 mol/L HClO4, the value determined spectrophotometrically
0.16
0.12
1.5
1.0
0.5
0.0
570
0.08
0.04
Asb
0
6000
t (min)
12000
for kexp was 1.90 ꢁ 10ꢂ4 sꢂ1
,
that compares well with
Aosrbance
2ka = 1.65 ꢁ 10ꢂ4 sꢂ1 obtained from EPR measurements.
Sequential absorption spectra of the reaction mixture showed
the appearance and growth of two d–d bands at kmax = 409 and
572 nm (Fig. 8). Kinetics traces at 570 nm showed that at this
wavelength, the absorbance grew to intensities higher than ex-
k
k
b
a
300
400
500
600
700
III
Cr
V
VI
Cr
2 Cr
λ (nm)
+
Cr
III
Fig. 8. Time evolution of the UV–Vis spectra of a mixture of 0.4 mol/L Glc3Me and
0.01 mol/L CrVI over a period of 16 h. [HClO4] = 1.02 mol/L, T = 33 °C. Inset: growth
and slow decay of Abs570 over a period of 9 days.
Scheme 2.

M.I. Frascaroli et al. / Polyhedron 28 (2009) 1049–1056
1055
fast
0.0030
0.0025
0.0020
0.0015
0.0010
0.0005
0.0000
VI
VI
(3)
(4)
Glc3Me + Cr
Glc3Me-Cr
IV
k6, 2 H+
VI
Glc3Me-Cr
Cr + HCO H + Ara2Me
2
fast
fast
III
(5)
(6)
Cr + HCO H + Glc3Me
2
IV
Cr + Glc3Me
II
Cr + HCO H + Ara2Me
2
k
fast
fast
V
III
II
(7)
(8)
VI
Cr + Cr
Cr + Cr
V
VI
Cr + Ara2Me + HCO H
2
Cr + Glc3Me
0.0
0.2
0.4
0.6
0.8
1.0
fast (k )
b
[H+] (mol/L)
Glc3Me + Cr
III
V
(9)
Cr + HCO H + Ara2Me
2
Fig. 9. Effect of acidity on kexp calculated from absorbance data at 570 nm.
[CrVI] = 8 ꢁ 10ꢂ3 mol/L, [Glc3Me] = 0.4 mol/L, I = 1.02 mol/L, T = 33 °C.
fast
fast
2+
III
(10)
(11)
II
CrO
Cr
2
Cr + O
2
were plotted against [H+] (Fig. 9). The rate constant calculated by
non-linear least-square fit of {kexp, [H+]} data pairs was found to
be 25.4 ꢁ 10ꢂ4 molꢂ2 L2 sꢂ1, which is the same as that calculated
from data at 350 nm. This indicates that the rate of formation of
CrIII equals the rate of consumption of CrVI and that the slow redox
path effectively implies the reduction of CrVI, which should react
slower than CrV does. This is also consistent with the observation
of an isosbestic point at 527 nm (Fig. 8) and implies a low CrV con-
centration throughout the reaction in the [H+] range used in the
kinetics studies.
Glc3Me
ox
O + Glc3Me
2
Scheme 3.
yield CrIV, HCO2H and Ara2Me (Eq. 4). The initial two-electron
reduction of CrVI by Glc3Me is in agreement with previous reports
on a number of oxygenated compounds that were selectively oxi-
dized by CrVI to the lower homolog [30,35]. In the mechanism, we
have included two competitive one- and two-electron reductions
of CrIV by Glc3Me. Thus, CrIV is proposed to react with excess
Glc3Me to yield CrIII, HCO2H and Glc3Meꢀ, or CrII, HCO2H and
Ara2Me, through two alternate fast steps (Eqs. 5 and 6). The first
is supported by the observed polymerization of acrylamide when
it is added to the CrVI/Glc3Me reaction mixture, while the second,
3.6. Mechanism of the Glc3Me + CrVI reaction
In the range of substrate and acid concentration used in the
kinetics measurements, the oxidation of Glc3Me by CrVI is a com-
plex reaction that yields CrIII, HCO2H and Ara2Me as final redox
products. The fact that CrO22+ is detected in the reaction of Glc3Me
with CrVI together with the observation of CrV species and the suc-
cessful trap of organic radicals using acrylamide indicates that the
reaction occurs through both one- and two-electron pathways
involving CrIV and CrV intermediate species. However, under condi-
tions used in the kinetic measurements, CrIV and CrV react with
Glc3Me faster than CrVI and do not accumulate in the reaction mix-
ture. Therefore, CrIV and CrV, although formed in the CrVI + Glc3Me,
should be involved in fast steps of the reaction pathway. Thus, the
absorbance-time profiles reflect the [CrVI] monotonic decay (at
350 nm) or the [CrIII] monotonic growth (at 570 nm) without inter-
ference of [CrIV] or [CrV]. In Scheme 3, we propose a mechanism
that combines CrVI ? CrIV ? CrII and CrVI ? CrIV ? CrIII pathways,
and takes into account: (a) kinetic results, (b) the polymerization
of acrylamide added to the reaction mixture, (c) detection of inter-
mediate of CrVI esters and oxochromate(V) species, (d) observation
2+
by the observation of CrO2 (the product of the reaction of CrII
with O2). CrV can form by fast reaction of CrII with CrVI (Eq. 7)
and, alternatively, by rapid reaction of the Glc3Meꢀ with CrVI (Eq.
8). CrV can further oxidize Glc3Me to yield CrIII, HCO2H and Ara2Me
as final redox products (Eq. 9). In the mechanism, one half of the
CrVI reaches CrIII through the CrV intermediate, in accordance with
Scheme 2 used to fit the time evolution of the CrV EPR signal.
In O2-saturated solutions (1.26 mmol/L) and [CrVI]0 < 0.1 mmol/
L, reactions 7 and 8 can be neglected because Glc3Meꢀ and CrII
intermediates formed in reactions 5 and 6 should be rapidly
trapped by O2 (reactions 10 and 11) [39]. The proposed mechanism
is in accordance with the observation that O2 has not kinetic effect
on this reaction, because when [CrVI]0 P 0.5 mmol/L (as employed
in the kinetic measurements), both CrII and Glc3Meꢀ react with CrVI
faster than they do with O2, and reactions 10 and 11 can be ne-
glected [27,40,41].
2+
of CrO2 and (e) the reaction products.
The first step of the mechanism proposed in Scheme 3 involves
the formation of a Glc3Me–CrVI mono-chelate (Eq. 3), in agreement
with the observation of the absorption band at 373 nm character-
istic of chromate oxo-ester, immediately after mixing Glc3Me and
CrVI under conditions where the redox reaction is extremely slow.
Therefore, such a chromate ester should be formed rapidly prior to
the redox steps. The formation of the chromate ester is followed by
the slow redox step where the C–C bond cleavage is proposed to
occur through an acid catalyzed two-electron redox process to
4. Conclusion
The reaction of Glc3Me with CrVI strongly depends on pH. In
acid medium, redox reaction occurs and reactive CrV, CrII and CrIV
intermediate species are generated in the redox process, together
with free radicals, HCO2H and Ara2Me. At pH > 3, Glc3Me oxidation
by CrVI or CrV is very slow and long-lived oxo-CrVI or oxo-CrV–
Glc3Me species form. EPR spectroscopy shows that at pH 5.5ꢂ7.5
Glc3Me is able to trap CrV through the 1,2-diolato moiety to yield

1056
M.I. Frascaroli et al. / Polyhedron 28 (2009) 1049–1056
a CrV bis-chelate that then slowly converts into species with CrV
bound to the 5,6-diolato moiety of its furanose form. The selective
C1–C2 bond cleavage of Glc3Me upon reaction with high valent
chromium in acid medium, distinguishes the 3-OMe derivative
from glucose that, under the same experimental conditions, is oxi-
dized to the gluconic acid at comparable rate [10].
[15] M. Krumpolc, J. Rocek, Inorg. Synth. 20 (1980) 63.
[16] M.C. Ghosh, E.S. Gould, Inorg. Chem. 30 (1991) 491.
[17] International Agency for Research on Cancer (IARC), IARC Monogr. Eval.
Carcinog. Risk Chem. Hum. (Suppl. 7) (1987) 165.
[18] R.Feldam, in: Occupational and Environmental Neurotoxicology, Lippincott-
Raven Publishers, Philadelphia, 1999, p. 337.
[19] L.F. Sala, A. Fernández Cirelli, R.M. Lederkremer, Anales Asoc. Química Arg. 66
(1978) 57.
[20] H.S. Isbell, J. Res. Nat. Bur. Stand. (US) 14 (1935) 305.
[21] A.G. Fadnis, S.K. Kulshrestha, Rev. Roumaine Chim. 28 (1983) 99.
[22] S.L. Scott, A. Bakac, J.H. Espenson, J. Am. Chem. Soc. 113 (1991) 7787.
[23] WINSIM EPR Calculations for MS-Windows NT, Version 0.96, National Institute of
Environmental Health Sciences, 1995.
[24] M. Mitewa, P. Bontchev, Coord. Chem. Rev. 61 (1985) 241.
[25] J.C. González, S. García, N. Mamana, L.F. Sala, S. Signorella, Inorg. Chem.
Commun. 9 (2006) 437.
[26] S.L. Scott, A. Bakac, J.H. Espenson, J. Am. Chem. Soc. 114 (1992) 4205.
[27] S. Signorella, M.I. Frascaroli, S. García, M. Santoro, J.C. González, C. Palopoli, N.
Casado, L.F. Sala, J. Chem. Soc., Dalton Trans. (2000) 1617.
[28] A. Al-Ajlouni, A. Bakac, J. Espenson, Inorg. Chem. 33 (1994) 1011.
[29] S. Signorella, M. Santoro, C. Palopoli, C. Brondino, J.M. Salas Peregrin, M. Quiroz,
L. Sala, Polyhedron 17 (1998) 2739.
Acknowledgments
We thank the National Research Council of Argentina (CONI-
CET), the National University of Rosario (UNR), the National Agency
for Sciences Promotion (FONCyT) and the ALFA Program (CROCON
Network, Project ALR/B7-30.11/94-7.0064.9) for financial support.
MIF thanks V. Andreu for helping with the kinetic measurements.
References
[30] G. Barr-David, M. Charara, R. Codd, R.P. Farrell, J.A. Irwin, P.A. Lay, R. Bramley, S.
Brumby, J.Y. Ji, G.R. Hanson, J. Chem. Soc., Faraday Trans. 91 (1995) 1207.
[31] A. Levina, P.A. Lay, Coord. Chem. Rev. 249 (2005) 281.
[32] M. Rizzotto, A. Levina, M. Santoro, S. García, M.I. Frascaroli, S. Signorella, L.F.
Sala, P. Lay, J. Chem. Soc., Dalton Trans. (2002) 3206.
[1] J. Barnhart, J. Soil Contam. 6 (1997) 561 (special issue).
[2] S.A. Katz, H. Salem, in: The Biological and Environmental Chemistry of
Chromium, VCH Publishers, Inc., New York, 1994.
[3] C.B. Klein, in: L.W. Chang (Ed.), Toxicology of Metals, CRC-Lewis Publishers,
New York, 1996, p. 205.
[33] S. Signorella, V. Daier, M. Santoro, S. García, C. Palopoli, L. Korecz, A.
Rockenbauer, L.F. Sala, Eur. J. Inorg. Chem. (2001) 1829.
[34] S. Signorella, S. García, M. Rizzotto, A. Levina, P. Lay, L.F. Sala, Poyhedron 24
(2005) 1079.
[35] G. Haight, G. Jwisich, M. Kelso, P. Merril, Inorg. Chem. 24 (1985) 2740.
[36] V. Roldán, M. Santoro, J.C. González, J.M. Salas Peregrin, S. Signorella, L.F. Sala, J.
Inorg. Biochem. 98 (2004) 347.
[37] A.B.P. Lever, in: Inorganic Electronic Spectroscopy, 2nd Ed., Elsevier,
Amsterdam, 1984, p. 419.
[38] S. Signorella, M. Santoro, A. Frutos, G. Escandar, J.M. Salas-Peregrin, V. Moreno,
M. González Sierra, L.F. Sala, J. Inorg. Biochem. 73 (1999) 93.
[39] A. Bakac, S.L. Scott, J.H. Espenson, K.R. Rodgers, J. Am. Chem. Soc. 117 (1995)
6483.
[40] O. Pestovsky, A. Bakac, J.H. Espenson, J. Am. Chem. Soc. 120 (1998) 13422.
[41] P.A. Lay, A. Levina, J. Am. Chem. Soc. 120 (1998) 6704.
[4] R. Codd, C.T. Dillon, A. Levina, P.A. Lay, Coord. Chem. Rev. 216–217 (2001) 537.
[5] R. Codd, J.A. Irwin, P. Lay, Curr. Opin. Chem. Biol. 7 (2003) 213.
[6] A.K. Das, Coord. Chem. Rev. 248 (2004) 81.
[7] M. Ciéslak-Golonka, M. Daszkiewicz, Coord. Chem. Rev. 249 (2005) 2391.
[8] B. Gyurcsik, L. Nagy, Coord. Chem. Rev. 203 (2000) 81.
[9] S. Signorella, J.C. González, L.F. Sala, J. Argentine Chem. Soc. 90 (2002) 1.
[10] S. Signorella, V. Daier, S. García, R. Cargnello, J.C. González, M. Rizzotto, L.F.
Sala, Carbohydr. Res. 316 (1999) 14.
[11] V. Daier, S. Signorella, M. Rizzotto, M. Frascaroli, C. Palopoli, C. Brondino, J.M.
Salas-Peregrin, Can. J. Chem. 77 (1999) 57.
[12] M. Rizzotto, S. Signorella, M.I. Frascaroli, V. Daier, L.F. Sala, J. Carbohyd. Chem.
14 (1995) 45.
[13] S. Signorella, M. Rizzotto, V. Daier, M.I. Frascaroli, C. Palopoli, D. Martino, A.
Boussecksou, L.F. Sala, J. Chem. Soc., Dalton Trans. (1996) 1607.
[14] M. Rizzotto, M. Frascaroli, S. Signorella, L. Sala, Polyhedron 15 (1996) 1517.