Kinetics of oxidation of nitrosodisulfonate anion radical with a metallo-superoxide

Article abstract of DOI:10.1039/c2dt12019d

The metal bound superoxide in μ-superoxo-bis[pentaamminecobalt(iii)] ⁵⁺ (1) oxidizes the nitrosodisulfonate anion radical (NDS ^2-) by two electrons. Oxidized NDS^2- quickly decomposes to SO₄^2- and NO. 1 is itself reduced to the corresponding hydroperoxo complex which also decomposes fast to Co(ii), NH₄ ⁺ ions and oxygen. 1.5 moles of volatile products formed per mole of 1 mixed with excess NDS^2-. In the absence of superoxide in a bridged complex, e.g. the μ-amido-bis[pentaamminecobalt(iii)]⁵⁺ complex fails to oxidize the nitroxyl radicals, NDS^2-, TEMPO and 4-oxo TEMPO. With excess NDS^2- over 1, the reaction is first-order with respect to [1], [NDS^2-] and inverse first order in [H⁺]. The activation entropy, ΔS^≠, is largely negative, increased ionic strength decreased the rate and a Bronsted plot is fairly linear with a negative slope. Oxidant μ-superoxo-bis[(ethylenediamine) (diethylenetriamine)cobalt(iii)]⁵⁺ has ligands sterically more crowded though more basic than ammonia in 1. It oxidizes NDS^2- much more slowly. No solvent kinetic isotope effect (kH₂O/D₂O ≈ 1) could be seen; a spin-adduct formation by the conjugate base of 1 followed by electron transfer is postulated. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2dt12019d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 2714
www.rsc.org/dalton
PAPER
Kinetics of oxidation of nitrosodisulfonate anion radical with a
metallo-superoxide†
Kaustab Mandal* and Rupendranath Banerjee
Received 24th October 2011, Accepted 21st November 2011
DOI: 10.1039/c2dt12019d
The metal bound superoxide in μ-superoxo-bis[pentaamminecobalt(III)]5+ (1) oxidizes the
nitrosodisulfonate anion radical (NDS2−) by two electrons. Oxidized NDS2− quickly decomposes to
SO42− and NO. 1 is itself reduced to the corresponding hydroperoxo complex which also decomposes fast
to Co(II), NH4+ ions and oxygen. 1.5 moles of volatile products formed per mole of 1 mixed with excess
NDS2−. In the absence of superoxide in a bridged complex, e.g. the μ-amido-bis
[pentaamminecobalt(III)]5+ complex fails to oxidize the nitroxyl radicals, NDS2−, TEMPO and 4-oxo
TEMPO. With excess NDS2− over 1, the reaction is first-order with respect to [1], [NDS2−] and inverse
first order in [H+]. The activation entropy, ΔS≠, is largely negative, increased ionic strength decreased the
rate and a Brønsted plot is fairly linear with a negative slope. Oxidant μ-superoxo-bis[(ethylenediamine)
(diethylenetriamine)cobalt(III)]5+ has ligands sterically more crowded though more basic than ammonia in
1. It oxidizes NDS2− much more slowly. No solvent kinetic isotope effect (kH O/D2O ≈ 1) could be seen;
2
a spin-adduct formation by the conjugate base of 1 followed by electron transfer is postulated.
important probe for reactive oxygen species (ROS) in biology.12
Introduction
The superoxo moiety of the complex, μ-superoxo-bis
[pentaamminecobalt(III)]5+ (1) oxidizes the nitroxyl radical
NDS2− at a measurable rate, and here we report its kinetics and
plausible mechanism.
The primary objective of this work is to see how the metal-
coordinated superoxide oxidizes NDS2−, and to what products?
We wish thus to add further details to the oxidation chemistry of
NDS2− with a metal-bound radical. It can be here mentioned that
electron-transfer reactions involving radicals and radical ions
have attracted much attention.13
The persistent odd-electron species, nitrosodisulfonate anion
radical, ˙ON(SO3)2
(NDS2−) forms by virtually complete
2−
homolysis of Fremy’s salt (potassium nitrosodisulfonate dimer)
in polar solvent media.1 Many groups utilized mild oxidizing
capacity of this comparatively stable, water-soluble nitroxyl
radical in kinetic studies2 and preparation3 related to a wide
range of natural products.4 Some closed ring nitroxyl radicals
like TEMPO and its derivatives have also been oxidized to
oxoammonium salts, and in some cases, to their decomposition
products. The kinetics5 and catalytic activities6 of these reactions
are well studied. In comparison, the reducing action of NDS2− is
little known.
In this context, it is interesting to note that the superoxo
complex 1oxidizes hydroxylamine disulfonate, HON(SO3K)2·
2H2O (HDS), faster than NDS2− under comparable reaction con-
ditions. Its preparative implications have also been discussed.
Reactive oxygen species (ROS)7 are a class of radical or non-
radical oxygen-containing molecules and display high reactivity
with lipids, proteins and nucleic acids.8 Depending on concen-
tration, location and context, ROS can be either “friends” or
“foes”.9 Accumulative and systemic ROS damage underlies cell
senescence and aging. However, increasing evidence indicates
that homeostatic and physiological levels of ROS are indispensa-
ble in regulating diverse cellular processes.10
Experimental
Materials
2,2,6,6-Tetramethylpiperidine-N-oxyl radical (TEMPO) and
Superoxide is one of the ROS. Even its metal complexes have
substantial unpaired spin density on the oxygen, which makes it
quite reactive towards radicals.5 Nevertheless, its reactivity can
be controlled by metal coordination.11 Again, NDS2− is an
2,2,6,6-tetramethyl-4-oxopiperidine-N-oxyl
TEMPO) were of analytical grade (Aldrich) and were used as
received.
radical
(4-oxo
Fremy’s salt was prepared by the reported procedure14 and
stored below 5 °C in an ammonia environment. Freshly prepared
samples were pure (ε545 = 20.8 M−1 cm−1 on the basis15 of
NDS2−) but slowly decomposed on storage. Therefore, fresh
samples were prepared regularly and samples with ε ≥ 19.5 were
only used.
Department of Chemistry, Jadavpur University, Kolkata, 700 032, India.
E-mail: kaustab2008@gmail.com
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c2dt12019d
2714 | Dalton Trans., 2012, 41, 2714–2719
This journal is © The Royal Society of Chemistry 2012

View Article Online
The pentachloride salt of μ-superoxo-bis[pentaam-
minecobalt(III)]5+ (1) was synthesized following a literature
process.16 The chloride salt was converted to the corresponding
perchlorate salt17 and re-crystallized from 10% HClO4 (ε, M−1
cm−1 at 670 nm = 835; lit. value: 83818).
amount of sulfate captured as barium salt did not change in the
presence of NH4Cl; it eliminated any chance15 for accumulation
of a detectable amount of HDS formed by the acid degradation
of NDS2−. These observations suggest the following stoichio-
metric equation (eqn (1)).16
μ-Amido-bis[pentaamminecobalt(III)] pentaperchlorate mono-
hydrate was prepared as described in the literature19 (ε, M−1
cm−1 at 360 nm = 705, at 505 nm = 426; lit. value: at 360 nm =
708, at 505 nm = 428). The compound is stable against auto-
decomposition.20
2 1 þ NDS2ꢀ ! NO þ 2 SO42ꢀ þ 2 O2 þ other product ð1Þ
The reduced form of 1 has been previously suggested to be 3,
the hydroperoxo derivative18 of 1 which rapidly decomposes26
+
to CoII, NH4 ion and O2. A blue color appeared by boiling the
product mixture with HCl that confirms the presence of Co2+
ions in the product solution. Co2+ ions, as with O2, had no effect
on the kinetics. Simple Co(II) species (hexaaqua) are d7 ions and
radicals on their own, which may react quickly with excess
NDS2−. Probably Co2+ and NDS2− experience ionic interaction
only and do not take part in kinetics which were not first-order
as observed if Co2+ takes part in kinetics (kind suggestion of one
referee).
The superoxo complex, μ-superoxo-bis[(ethylenediamine)
(diethylenetriamime)cobalt(III)] pentaperchlorate was synthesized
following a literature procedure21 and re-crystallized from 0.3 M
HClO4. Its purity was checked by measuring its absorbance at
708 nm (ε, M−1 cm−1 at 708 nm = 1200; lit. value: 1210).
HDS was prepared by a reported procedure.22 Its purity was
checked by H, N microanalyses (Anal. Calc. for HON(SO3K)2·
2H2O: H, 1.63; N, 4.50. Found: H, 1.60; N 4.41%).
Physical measurements, kinetics and stoichiometry
Kinetics and mechanism
In the range pH 4.85–5.85, individually neither 1 nor NDS2−
suffer any bleaching, even in 3 h, but on mixing them the reac-
tion is completed in less than 1.5 h. Murib et al.15 reported that
at pH 4.96 the decomposition time of NDS2− is 19 h. If we con-
sider the possibility that very small amounts of decomposition
products of NDS2− may act as chain propagators, one would
then expect complex auto-catalytic kinetics instead of the first-
order kinetics observed here. Hence this possibility is not further
taken up. The complex ion μ-amido-bis[pentaamminecobalt
The pH values of the reaction solutions were measured with a
Toshniwal pH meter (CL-54), having electrodes calibrated23
with standard pH 4, 7 and 9 buffers. In D2O media, the relation
pD = pH + 0.4 was used.24,25 H, N microanalyses were done on
a Perkin-Elmer 240C elemental analyzer. Absorbance, UV-Vis
spectra and kinetics were recorded with a Jasco (V-1700) spec-
trophotometer using 1.00 cm quartz cells. The kinetics was mon-
itored at 670 nm, observing the decay of 1 in aqueous acetate
buffer media (TOAc = 0.10 M) in the range pH, 4.85–5.85 at 25
( 0.1) °C in an electrically controlled thermostated cell-housing
of the instrument. Ionic strength (I) of the reaction media was
maintained at 1.5 M. A large excess of NDS2− over 1 was used
to maintain the pseudo first-order conditions. NDS2− has a small
contribution to the total absorbance at 670 nm. It was subtracted
from the observed total absorbance, assuming essential con-
stancy of [NDS2−]. The absorbance vs. time (t) data fitted a first-
order decay equation (ORIGIN 7.0). The reported first-order rate
constants (kobs) are the averages of 2–3 independent measure-
ments. It was verified that purging the reaction media with Ar
gas had practically no effect on kinetics and, therefore avoided.
With [NDS2−] in excess over [1], the amount of sulfate
formed was measured gravimetrically. The volume of gas(es)
evolved during the reaction was collected by downward displace-
ment of water and corrected to NTP as usual. NO gas was quali-
tatively tested in the ensuing gas by passing it through a fresh
FeSO4 solution acidified with sulfuric acid. The consumption
ratio, Δ[NDS2−]/Δ[1], was measured at 670 nm with [1] in
excess over [NDS2−]. Doubly distilled water was used all
throughout the reaction.
(III)]5+ has no superoxo moiety and it does not react with NDS2−
,
TEMPO or 4-oxo-TEMPO. This observation strongly suggests
that reaction of 1 with NDS2− is actually the reaction of the
superoxo group in 1. Other researchers17,27 also presumed the
same.
Spectral changes after mixing excess NDS2− with 1 are shown
in Fig. 1. The reactions were found to be first-order in [1]
(Fig. 2), first-order in [NDS2−] and inverse first order in [H+]
Results and discussion
The consumption ratio, Δ[1]/Δ[NDS2−] ≈ 2 and NDS2− is a two
electron donor (see ESI, Table S1a†) in contrast to some other
nitroxyl radicals like TEMPO and its derivatives which are
known to be one electron donors.5 With excess NDS2−, each
mole of 1 produces 1.5 moles of gaseous products (see ESI,
Fig. 1 Spectra at different times of 0.20 mM of 1 reacting with 4 mM
of NDS2−. pH = 5.60. TOAc = 0.10 M, I = 1.5 (NaCl) M and T = 25.0
( = 0.1) °C. Spectra (a–h) are reaction mixtures at 0, 60, 120, 180, 240,
300, 360 and 720 s, respectively.
2−
Table S1b†) and one mole of SO4 (see ESI, Table S1c†). The
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 2714–2719 | 2715

View Article Online
Table 1 Representative rate constants for the oxidation of NDS2− with
1 (0.20 mM) at different pH. TOAc = 0.10 M, I = 1.5 (NaCl) M and T =
25.0 ( 0.1) °C
103kobs (s−−1) at pH
[NDS2−]/mM
4.85
5.10
5.35
5.60
5.85
2.0
3.0
4.0
5.0
6.0
0.46
0.60
0.82
1.15
1.44
0.71
0.95
1.41
1.71
1.97
1.35
1.71
2.55
3.02
3.45
2.05
3.15
4.09
5.01
6.05
3.74
4.87
7.75
8.81
10.41
solvent kinetic isotope effect (see ESI, Table S2†) supports this
proposition. Therefore rate-enhancement with pH is ascribed to
the dissociation of an H+ ion from one of the coordinated NH3
molecules in 1, thus forming the conjugate base11
[(NH3)5Co(O2)Co(NH3)4(NH2)]4+ (2) as the kinetically reactive
species (eqn (2)). Decomposition of 1 by base hydrolysis16 sup-
ports the formation of 2 and inverse first order in [H+]. Kineti-
cally reactive conjugate bases like 2 have been sufficiently
demonstrated to participate30 in the base-catalysed hydrolysis
reactions of acidopentaamminecobalt(III) complexes.31
Fig. 2 Kinetic profile at 670 nm for the oxidation of NDS2− (4.0 mM)
with 1 (0.2 mM). pH = 5.6, TOAc = 0.10 M, I = 1.5 (NaCl) M and T =
25.0 ( 0.1) °C. The solid line is the first-order fit of the experimental
values (shown in solid circles). Absorbance of NDS2− was subtracted
from the observed total absorbance, assuming essential constancy of
[NDS2−].
ð2Þ
Other mechanistic possibilities for [OH−] dependence may
arise by attack of OH− ion at the superoxide bridge or by an
equilibrium attack of OH− ion on NDS2− (suggestion of a
referee). But attack by OH− ion on the superoxo-bridge in 1
needs catenation of three oxygen atoms and seems improbable.
Also, attack of OH− on NDS2− requires28,29 much higher pH of
the solution. We prefer conjugate base formation by 1 because of
the reasons stated above and because (a) several CoIII-amines
deprotonate31b,c at acidic pH to show base catalysis and (b) for-
mation of the conjugate base appears easier for a dinuclear
complex than for its mononuclear analogue.30
Fig. 3 Plot kobs vs. 1/[H+] at different concentrations of NDS2−. [1] =
0.20 mM, TOAc = 0.10 M, I = 1.5 (NaCl) M and T = 25.0 ( 0.1) °C.
(1)–(4) are at [NDS2−] = 2.0, 3.0 4.0 and 5.0 mM respectively. Corre-
The reduction potential of NDS2− is much lower (see Table 2)
than the reduction potential of 1 under similar conditions. So
sponding values of K1K2k1 are fairly constant = 5.1 ( 0.3) × 10−6 s−1
.
Previously11 for H2O2 the value was found to be 5.26 ( 0.2) × 10−8
s−1. A comparison for the reduction of 1 by different reducing agents is
given later in Table 3.
Table 2 A comparison of the reduction potentials
Reduction potential
(Volt vs. NHE)
Reagent
pH
4.0
Source
(Fig. 3). Some representative values of the rate constant (kobs
)
μ-Superoxo-bis
[pentaamminecobalt(III)]5+
NDS2−
1.00
Ref.
27c
are summarized in Table 1. kobs of the reaction decreases with
increasing ionic strength of the media.
5.9
4.9
4.0
1.0
1.0
1.0
0.70
0.76
0.81
0.98
0.91
0.73
Ref. 37
Ref. 37
Ref. 37
Ref. 37
Ref. 5
Ref. 5
NDS2−
The superoxo moiety in 1 has no acidic proton and cannot
explain the inverse proton dependence of kobs. NDS2−, in basic
media may add to water28 or a nucleophile29 and thus can act as
a source for the inverse proton dependence. In the acetate buffer
media, such a function for NDS2− is unexpected. Absence of
NDS2−
NDS2−
4-Oxo-TEMPO
TEMPO
2716 | Dalton Trans., 2012, 41, 2714–2719
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 1
NDS2− should act as a reductant for 1 according to the
thermodynamics.
Based on the kinetic and stoichiometric observations, we
propose the Scheme 1 for the oxidation of NDS2−
.
Two moles of 1 are consumed by one mole of NDS2−, so it is
a non-complementary reaction and stoichiometric factor is two.
Assuming K1K2 ≪ 1 the proposed scheme leads to kobs
=
0.5K1K2k1[NDS2−]/[H+] (see ESI†), in accord with the obser-
vations. Some values of K1K2k1 are summarized in Fig. 3. From
this figure it is clear that composite K1K2k1 regenerated kobs
values reasonably well.
Previously Goldstein et al.32 suggested adduct formation
before electron transfer between the reaction of NO2˙ and
nitroxyl radical. Also, Bakac5 showed that LMOO2+ undergoes
adduct formation with nitroxyl radicals. Such adducts are inner-
sphere spin adducts and involve distinct chemical bonding
between the reactants.33 We found that NDS2− reacts 104 times
slower with μ-superoxo-bis[(ethylenediamine)(diethylenetri-
amine)cobalt(III)]5+ ion than with 1. The pKa of ammonia is 9.3
while that of ethylenediamine34 and diethylenetriamine35 are
respectively 10.0 and 9.9. So K1, and k1 are expected to be
larger in the μ-superoxo-bis[(ethylenediamine)(diethylenetri-
amine)cobalt(III)]5+ complex but the reverse has been found.
Possibly, steric factors dominate over pKa values and steric
congestion in the μ-superoxo-bis[(ethylenediamine)(diethylene-
triamine)cobalt(III)]5+ complex slows down the adduct formation
equilibrium prior to the electron transfer.
Since K1, K2 and k1 are inseparable, Erying plot with the third
order rate constants, kobs[H+]/[NDS2−], has been constructed
(Fig. 4). This gives an activation enthalpy ΔH≠ = 83.8 kJ mol−1
and the activation entropy ΔS≠ = −54.3 JK−1 mol−1. Such a
highly negative value of ΔS≠ indicates close approach of the
reagents, and supports the K2 step in the Scheme 1.
Ionic strength (I) plays an important role in reactions between
charged species. We investigated the effect of ionic strength on
the reaction by varying NaCl concentration in the reaction
media. The Brønsted plot of log(kobs) vs. I1/2 (at pH = 5.35 and
[NDS2−] = 4 mM) is fairly linear with a negative slope (see ESI,
Fig. 4 Eyring plot at pH = 5.85, [1] = 0.20 mM, [NDS2−] = 10.0 mM,
TOAc = 0.10 M and I = 1.5 (NaCl) M. Third-order rate constants (k) =
k
obs[H+]/[NDS2−].
Fig. S1†) probably due the interaction of opposite charges in the
K2 step.
1 reacts with TEMPO almost immediately. In comparison, its
reaction with 4-oxo-TEMPO is slow (Table 3) but faster than
that of NDS2−. From the reduction potential data shown in
Table 2, we conclude that thermodynamically weaker reducing
agents react at a slower rate. So electron transfer appears to be
through outer sphere, after the spin-adduct formation.
As expected, no solvent kinetic isotope effect is observed for
the oxidation of NDS2− with 1 and kH O/D2O ≈ 1. Therefore
2
this accounts to describe a pathway in which concerted electron–
proton transfer occurs, but involves more than one site, i.e. elec-
tron transfer from NDS2− and proton transfer from the solvent at
different time.
All the observations support the proposed Scheme 1, where 2,
the conjugate base of 1 accepts electron from NDS2−; μ-super-
oxo-bis[(ethylenediamine)(diethylenetriamine)cobalt(III)]5+ ion is
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 2714–2719 | 2717

View Article Online
Table 3 Kinetic data for the reduction of 1 in aqueous solution at T =
25.0 °C
Thus we see that a nitroxyl radical can either donate or accept
an electron. Although other nitroxyl radicals like TEMPO, 4-
oxo-TEMPO, HO-TEMPO also act in the same way, they are
one electron agents,5 but NDS2− is a two electron species, an
important probe12 for reactive oxygen species (ROS) and an
efficient catalyst for the dismutation of superoxide.36 Such a
nitroxyl radical (NDS2−) may act as reducing agent as well as an
oxidizing agent.
Concentration
(mM)
Reductant
pH
kobs, s−1
Source
Ascorbic acida
Ascorbic acida
Ascorbic acida
Hydrazineb
200
200
200
150
150
100
5.17
4.94
4.65
6.10
5.80
6.10
5.01
5.30
5.01
5.01
5.48
5.01
4.39
4.39
4.39
24.0
20.1
18.8
Ref. 17
Ref. 17
Ref. 17
26.0 × 10−4 Ref. 38
16.0 × 10−4 Ref. 38
18.0 × 10−4 Ref. 38
2.7 × 10−3 Ref. 11
5.3 × 10−3 Ref. 11
1.7 × 10−3 Ref. 11
8.1 × 10−3 This work
25.0 × 10−3 This work
14.2 × 10−3 This work
Hydrazineb
Hydrazineb
Hydrogen peroxidec 50
Hydrogen peroxidec 50
Hydrogen peroxidec 30
Acknowledgements
HDSd
2
2
5
5
5
5
The work has been carried out with the financial assistance from
DST (Project No. SR/S1/IC-40/2007) and UGC-CAS to the
Department of Chemistry, Jadavpur University. The award of an
SRF (CSIR, New Delhi) to K.M. is gratefully acknowledged.
HDSd
HDSd
TEMPOe
4-Oxo-TEMPOe
NDS2−e
Instant
This work
32.5 × 10−3 This work
0.5 × 10−3 This work
a At [1] = 0.01 mM, TOAc = 0.20 M, I = 0.20 M (NaClO4). b At [1] =
0.50 mM, [dipicolinic acid] = 0.10 mM, TOAc = 0.20 M, I = 1.00 M
(NaClO4). c At [1] = 0.20 mM, TOAc = 0.20 M, I = 0.50 M (NaClO4).
d At [1] = 0.20 mM, TOAc = 0.10 M, I = 1.50 M (NaCl). e At [1] =
0.20 mM, TOAc = 0.10 M, I = 1.00 M (NaCl).
References
1 (a) R. W. Asmussen, Z. Anorg. Allg. Chem., 1933, 212, 317;
(b) B. J. Wilson, J. M. Hayes and J. A. Durbin, Inorg. Chem., 1976, 15,
1703.
2 (a) Z. Yang and E. S. Gould, Res. Chem. Intermed., 2003, 29, 181;
(b) G. Palmisano, B. Danieli, G. Lesma, F. Trupiano and T. Pilati,, J. Org.
Chem., 1988, 53, 1056; (c) B. Giethlen and J. M. Schaus,, Tetrahedron
Lett., 1997, 38, 8483; (d) J. Zielonka, H. Zhao, Y. Xu and
B. Kalyanaraman, Free Radical Biol. Med., 2005, 39, 853.
3 (a) S. C. Jaehrig, S. Rohn, L. W. Kroh, L. G. Fleischer and T. Kurz, J.
Agric. Food Chem., 2007, 55, 4710; (b) J. M. Saa, C. Marti and A.
G. Raso, J. Org. Chem., 1992, 57, 589; (c) S. Kumar, J. Org. Chem.,
2002, 67, 8842; (d) G. A. Kraus and N. Selvakumar, Tetrahedron Lett.,
1999, 40, 2039; (e) R. T. LaLonde and M. Zhang, J. Nat. Prod., 2004,
67, 697.
4 (a) T. P. Holler and P. B. Hopkins, Biochemistry, 1990, 29, 1953;
(b) Z. L. Liu, Z. X. Han, P. Chen and Y. C. Liu, Chem. Phys. Lipids,
1990, 56, 73; (c) H. Zimmer, D. C. Lankin and S. W. Horgan, Chem.
Rev., 1971, 71, 229.
5 M. J. Vasbinder and A. Bakac, Inorg. Chem., 2007, 46, 2322.
6 (a) P. Ferreira, E. Phillips, D. Rippon and S. C. Tsang, Appl. Catal., B,
2005, 61, 206; (b) D. Brunel, F. Fajula, J. B. Nagy, B. Deroide, M.
J. Verhoef, L. Veum, J. A. Peters and H. van Bekkum, Appl. Catal., A,
2001, 213, 73.
sterically congested and forms weaker adduct with NDS2−; so
K2 value is smaller leading to a smaller kobs
.
No attempt was taken to detect any intermediate such as 2 or
the “spin adduct”, since no saturation kinetics was observed,
which tells us that no intermediate accumulates to a detectable
extent.
1, under similar conditions, reacts with hydroxylamine disul-
fonate (HDS) almost ten times faster than NDS2− (see ESI,
Table S3†). HDS upon oxidation gives the radical NDS2−, which
itself is a mild oxidant. So strong oxidants “kill” NDS2− as
observed here and the yield of NDS2− thus reduces.
7 (a) V. J. Thannickal and B. L. Fanburg, Am. J. Physiol. Lung Cell. Mol.
Physiol., 2000, 279, L1005; (b) W. Droge, Physiol. Rev., 2002, 82, 47.
8 (a) F. Haber and J. Weiss, Proc. R. Soc. London, Ser. A, 1934, 147, 332;
(b) C. Beauchamp and I. Fridovich, J. Biol. Chem., 1970, 245, 4641.
9 Y. Luo, Y. Tian and Q. Rui, Chem. Commun., 2009, 3014.
10 (a) S. M. Hahn, F. J. Sullivan, A. M. DeLuca, M. Sprague, V.
A. Hampshire, M. C. Krishna, A. Russo and J. B. Mitchell,
Int. J. Oncol., 1997, 10, 119; (b) J. B. Mitchell, M. C. Krishna,
A. Samuni, A. Russo and S. M. Hahn, in Reactive Oxygen Species in Bio-
logical Systems: An Interdisciplinary Approach, D. L. Gilbert and
C. A. Colton, Plenum ed., New York, 1999, pp. 293–313; (c) R. Zhang,
E. Shohami, E. Beit-Yannai, R. Bass, V. Trembovler and A. Samuni, Free
Radical Biol. Med., 1998, 24, 332; (d) G. Hu, B. G. Lyeth, X. Zhao, J.
B. Mitchell and J. C. Watson, J. Neurosurg., 2003, 98, 393.
11 R. Mishra, S. Mukhopadhyay and R. Banerjee, Dalton Trans., 2009,
5469.
12 (a) L. Valgimigli, G. F. Pedulli and M. Paolini, Free Radical Biol. Med.,
2001, 31, 708; (b) J. Skolimowski, A. Kochman, L. Gebicka and
D. Metodiewa, Bioorg. Med. Chem., 2003, 11, 3529.
13 S. Swavey, M. C. Ghosh, V. Manivannan and E. S. Gould, Inorg. Chim.
Acta, 2000, 306, 65.
14 S. Yamada and R. Tsuchida, Bull. Chem. Soc. Jpn., 1959, 32, 721.
15 J. H. Murib and D. M. Ritter, J. Am. Chem. Soc., 1952, 74, 3394.
16 R. Davies,M. Mori,A. G.SykesandJ. A.Weil, Inorg. Synth., 1972,12, 197.
17 S. K. Saha, M. C. Ghosh and E. S. Gould, Inorg. Chem., 1992, 31, 5439.
18 A. B. Hoffman and H. Taube, Inorg. Chem., 1968, 7, 1971.
19 R. Davies, M. Mori, A. G. Sykes and J. A. Weil, Inorg. Synth., 1972, 12,
212.
Conclusions
NDS2− radical is generally used as a mild oxidant, but it supplies
−
electron to the O2 part of the conjugate base of μ-superoxo-bis
[pentaamminecobalt(III)]5+ (1), i.e [(NH3)5CoO2Co(NH2)
(NH3)4]4+ (2). NDS2− first forms an intimate spin adduct with
superoxo part of 2 and then supplies two electrons to two mol-
ecules of 2. NDS2− thus converts 1 ultimately to its hydroperoxo
form [(NH3)5Co(HO2)Co(NH3)5]5+ (3) which quickly decom-
+
poses to Co(II), O2, NH4 and other products. No solvent kinetic
effect (kH O/D2O ≈ 1) could be seen and the reaction appears to
2
be non-electroprotic, i.e. necessary proton transfer and electron
transfer occurs to different orbitals at different time. NDS2− after
2−
electron transfer decomposes to NO and SO4
.
Another important conclusion pertains to the synthesis of
Fremy’s salt. Oxidation of hydroxylamine disulfonate (HDS)
gives the radical NDS2−, which decomposes after further
oxidation. In these oxidations, 1 reacts with HDS faster than
NDS2−, which therefore appears to be a kinetically controlled
product. Hence, NDS2− should be taken out of the contact of the
oxidant, as soon as NDS2− is formed and the yield of Fremy’s
salt may be increased.
2718 | Dalton Trans., 2012, 41, 2714–2719
This journal is © The Royal Society of Chemistry 2012

View Article Online
20 R. D. Mast and A. G. Sykes, J. Chem. Soc., 1968, A, 1031.
21 D. L. Duffy, D. A. House and J. A. Weil, J. Inorg. Nucl. Chem., 1969,
31, 2053.
22 G. K. Rollefson and C. F. Oldershaw, J. Am. Chem. Soc., 1932, 54, 977.
23 S. Mukhopadhyay and R. Banerjee, J. Chem. Soc., Dalton Trans., 1993,
933.
24 (a) H. M. Irving, M. G. Miles and L. D. Petit, Anal. Chim. Acta, 1967,
38, 475; (b) P. K. Glasoe and F. A. Long, J. Phys. Chem., 1960, 64, 188.
25 (a) P. Salomaa, L. L. Schaleger and F. A. Long, J. Am. Chem. Soc., 1964,
86, 1; (b) A. K. Covington, R. A. Robinson and R. G. Bates, J. Phys.
Chem., 1966, 70, 3820.
26 S. P. Ghosh, S. K. Saha, R. N. Bose, J. W. Reed, M. C. Ghosh and E.
S. Gould, Inorg. Chem., 1993, 32, 2261.
27 (a) T. D. Hand, M. R. Hyde and A. G. Sykes, Inorg. Chem., 1975, 14,
1720; (b) S. K. Ghosh, S. K. Saha, M. C. Ghosh, R. N. Bose, J. W. Reed
and E. S. Gould, Inorg. Chem., 1992, 31, 3358; (c) G. McLendon and
W. F. Mooney, Inorg. Chem., 1980, 19, 12.
30 J. S. Anderson, H. V. A. Briscoe and N. F. Spoor, J. Chem. Soc., 1943,
361.
31 (a) F. Basolo and R. G. Pearson, Mechanisms of InorganicReactions,
Wiley Eastern Limited, New Delhi, 2nd edn, 1967, ch. 3;
(b) Y. Kurimura, Bull. Chem. Soc. Jpn., 1973, 46, 2093; (c) Y. Kurimura,
I. Sekine, E. Tsuchida and Y. Karino, Bull. Chem. Soc. Jpn., 1974, 47,
1823.
32 S. Goldstein, A. Samuni and A. Russo, J. Am. Chem. Soc., 2003, 125,
8364.
33 A. Bakac, J. Am. Chem. Soc., 2002, 124, 9136.
34 J. Bjerrum, Metal-amine Formation, P. Haase and Son, Copenhagen,
1941.
35 J. E. Prue and G. Schwarzenbach, Helv. Chim. Acta, 1950, 33, 985.
36 S. Goldstein, G. Merenyi, A. Russo and A. Samuni, J. Am. Chem. Soc.,
2003, 125, 789.
37 A. Morante, R. Forteza and V. Cerda,, Thermochim. Acta, 1987, 118,
215.
28 B. J. Wilson and D. M. Ritter, Inorg. Chem., 1963, 2, 974.
29 P. N. Balasubramanian and E. S. Gould, Inorg. Chem., 1983, 22, 1100.
38 A. Mondal and R. Banerjee, Indian Journal of Chemistry, 2009, 48A,
645.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 2714–2719 | 2719