Kinetics and mechanism of the cobalt phthalocyanine catalyzed reduction of nitrite and nitrate by dithionite in aqueous solution

Article abstract of DOI:10.1021/ic020290f

The reaction of sodium nitrite with sodium dithionite was studied in the presence of cobalt(II) tetrasulfophthalocyanine, Co^II(TSPc)^4-, in aqueous alkaline solution. The overall mechanism comprises the reduction of Cor^II(TSPc)^4- by dithionite, followed by the formation of an intermediate complex between Co^I(TSPc)^5- and nitrite, which undergoes two parallel subsequent reactions with and without nitrite as a reagent. Kinetic parameters for the different reaction steps of the catalytic process were determined. The final product of the reaction was found to be ammonia. Contrary to those found for the catalytic reduction of nitrite, the products of the catalytic reduction of nitrate were found to be dinitrogen and nitrous oxide. The possible catalytic reduction of nitrous oxide was confirmed by independent experiments. The striking differences in the reduction products of nitrite and nitrate are explained in terms of different structures of the intermediate complex between Co^I(TSPc)^5- and substrate, in which nitrite and nitrate are suggested to coordinate via nitrogen and oxygen, respectively.

Full text of DOI:10.1021/ic020290f

Inorg. Chem. 2003, 42, 618−624
Kinetics and Mechanism of the Cobalt Phthalocyanine Catalyzed
Reduction of Nitrite and Nitrate by Dithionite in Aqueous Solution
,†
Evgeny V. Kudrik,*,†,‡ Sergei V. Makarov,†,‡ Achim Zahl,† and Rudi van Eldik*
Institute for Inorganic Chemistry, UniVersity of Erlangen-Nu¨rnberg, Egerlandstrasse 1,
91058 Erlangen, Germany, and IVanoVo State UniVersity of Chemistry and Technology,
aV. F. Engels 7, 153460 IVanoVo, Russia
Received April 22, 2002
The reaction of sodium nitrite with sodium dithionite was studied in the presence of cobalt(II) tetrasulfophthalocyanine,
II
II
Co (TSPc)4-, in aqueous alkaline solution. The overall mechanism comprises the reduction of Co (TSPc)4- by
dithionite, followed by the formation of an intermediate complex between Co(TSPc)5- and nitrite, which undergoes
I
two parallel subsequent reactions with and without nitrite as a reagent. Kinetic parameters for the different reaction
steps of the catalytic process were determined. The final product of the reaction was found to be ammonia. Contrary
to those found for the catalytic reduction of nitrite, the products of the catalytic reduction of nitrate were found to
be dinitrogen and nitrous oxide. The possible catalytic reduction of nitrous oxide was confirmed by independent
experiments. The striking differences in the reduction products of nitrite and nitrate are explained in terms of
I
different structures of the intermediate complex between Co(TSPc)5- and substrate, in which nitrite and nitrate are
suggested to coordinate via nitrogen and oxygen, respectively.
Introduction
branching point in assimilatory nitrate reduction,6 this process
has received substantial attention in recent years. In addition
to extensive studies of the biological denitrification process,6
the kinetics of the chemical reduction of nitrite by metallic
iron,7 FeII(edta),8 amine-borane,9 and hydrazine10 has been
studied in detail. Despite this progress, considerable con-
troversy exists concerning the mechanisms of nitrite reduc-
tion. One of the unsettled questions is the role of nitric oxide.6
Is it an intermediate in the reduction of nitrite to N2O, or is
the formation of nitrous oxide a result of the reaction between
nitrite and a product of its one-electron reduction reaction?
Further interest in the redox reactions of N2O was stimulated
by reports of its greenhouse gas properties.11 N2O can account
for as much as 7% of the projected atmospheric warming.12
Denitrification plays a key role in the biogeochemical
nitrogen cycle.1 Nitrate serves as the substrate for this process
in which it is reduced via nitrite, nitric oxide, and nitrous
oxide to dinitrogen. The first step, reduction of nitrate to
nitrite, is catalyzed in organisms by molybdenum-containing
enzymes, viz., nitrate reductases.2 Since nitrate is a potentially
harmful compound to human health,3 processes for nitrate
removal from drinking water have regained interest in recent
years.4 One of the most promising methods is catalytic nitrate
reduction using bimetallic palladium catalysts.4
Reduction of nitrite is catalyzed in organisms by enzymes
of two types. The first type is heme enzymes in which FeII
is the active center in the porphyrin macrocycle. The second
type involves enzymes containing CuI, viz., nitrite and nitric
oxide reductases.5 Since reduction of nitrite represents the
(6) Averill, B. A. Chem. ReV. 1996, 96, 2951.
(7) Hu, H.-Y.; Goto, N.; Fujie, K. Water Res. 2001, 35, 2789.
(8) Zang, V.; Kotowski, M.; van Eldik, R. Inorg. Chem. 1988, 27, 3279.
(9) Bell, K. E.; Kelly, H. C. Inorg. Chem. 1996, 35, 7225.
(10) Chevalier, A. A.; Gentil, L. A.; Amorebieta, V. T.; Gutierrez, M. M.;
Olabe, J. A. J. Am. Chem. Soc. 2000, 122, 11238.
* Authors to whom correspondence should be addressed. E-mail: ttos@
isuct.ru (E.V.K.); vaneldik@chemie.uni-erlangen.de (R.v.E.).
† University of Erlangen-Nu¨rnberg.
(11) (a) Rasmussen, R. A.; Khalil, M. A. K. Science 1986, 232, 1623. (b)
IPCC. In Climate Change 1995, the Science of Climate Change;
Houghton, J. T., Meira Filho, L. G., Callander, B. A., Harris, N.,
Kattenberg, A, Maskell, K., Eds.; Cambridge University Press:
Cambridge, U.K., 1996; pp 9-50. (c) Cicerone, R. J. Science 1987,
237, 35.
‡ Ivanovo State University of Chemistry and Technology.
(1) Richardson, D. J. Cell. Mol. Life Sci. 2001, 58, 163.
(2) Campbell, W. H. Annu. ReV. Plant Physiol. Plant. Mol. Biol. 1999,
50, 277.
(3) Haghighi, B.; Tavassoli, A. Fresenius’ J. Anal. Chem. 2001, 371, 1113.
(4) Prusse, U.; Vorlop, K.-D. J. Mol. Catal., A: Chem. 2001, 173, 313.
(5) Wang, Y.; Averill, B. A. J. Am. Chem. Soc. 1996, 118, 3972.
(12) Smart, D. R.; Bloom A. J. Proc. Natl. Acad. Sci. U.S.A. 2001, 98,
7875.
618 Inorganic Chemistry, Vol. 42, No. 2, 2003
10.1021/ic020290f CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/28/2002

Reduction of Nitrite and Nitrate by Dithionite
In addition, photolysis of N2O in the upper atmosphere leads
to ozone destruction.12
Two- and three-electron reduction processes of N2O have
been reported.13,14 In both reactions nitrous oxide reacts with
low-valence metal complexes. For this purpose the low-
valence metal complex should be relatively stable. A
promising candidate is cobalt phthalocyanine, Co(Pc). Ac-
cording to DFT calculations,15 the electron received during
reduction of CoII(Pc) is localized on the metal center and
not on the ligand as in the case of the phthalocyanine
complexes of Cu, Ni, and Zn. This illustrates the relative
stability of CoI(Pc). Another advantage of cobalt phthalo-
cyanine is that complexes of CoI(Pc), soluble in water and
in organic solvents, may be synthesized from commercially
available CoII(Pc) derivatives. For this purpose electrochemi-
cal16 and pulse radiolysis17 techniques, or reduction with
sodium in THF,17 have been used. Cobalt(I) phthalocyanine
is shown to be stable in alkaline solutions for at least several
hours.17
Figure 1. Visible absorbance spectra of different forms of Co(TSPc) at
pH 10: CoI(TSPc)4- (7.8 × 10-5 M, solid line), CoI(TSPc)5- after reaction
-
with NO2 (7.8 × 10-5 M, dashed line), CoIII(TSPc)4- prepared from a
reaction of CoII(TSPc)4- with sulfite in the presence of oxygen (4.4 × 10-5
M, dotted line).
(96% 15N grade) were used for 15N NMR and MS analyses. Oxygen-
free argon was used to deoxygenate solutions.
In this study, we prepared cobalt(I) tetrasulfophthalocya-
nine, CoI(TSPc),5- via reduction of CoII(TSPc)4- with sodium
dithionite. Subsequently, we used CoI(TSPc)5- as a catalyst
for the reduction of nitrite and nitrate in alkaline solution.
The kinetics and mechanisms of these reactions were studied
in detail.
Kinetic Measurements and Instrumentation. Conventional
kinetic experiments were performed on a Cary 1 or Cary 5 UV-
vis spectrophotometer under anaerobic conditions. No salt was
added to control the ionic strength. The data were analyzed using
Origin 6.1 and SPECFIT software. A thermostated ((0.1 °C)
Applied Photophysics SX 18MV stopped-flow spectrophotometer
with an observation path length of 1.0 cm was used to follow the
faster reactions. NMR measurements were performed using a Bruker
Avance DRX400 WB spectrometer equipped with a superconduct-
ing BS-94/89 magnet system, at 40.56 MHz for 15N and 400.13
MHz for 1H. 15N chemical shifts were referenced externally to neat
nitromethane. D2O (99%) was used for all measurements. Mass
spectra were recorded on a JEOL Mstation 700 (EI, 70 eV)
spectrometer. EPR spectra were recorded on a Bruker ESP 300 E
spectrometer (9.96 GHz) in H2O at 177 K.
Results and Discussion
Reaction of Sodium Dithionite with Cobalt(II) Tetra-
sulfophthalocyanine. Reduction of a CoII(TSPc)4- solution
by excess sodium dithionite is accompanied by a color
change from blue to brown. An intensive absorption maxi-
mum appears at 450 nm in the UV-vis spectrum, and
simultaneously the Q-band is shifted to the red region (Figure
1).
This spectrum differs considerably from that for CoIII-
(TSPc)3- and is essentially identical to that reported for
CoI(TSPc).5-17 The rate of reduction of CoII(TSPc)4- was
found to be independent of the CoII(TSPc)4- concentration.
Rate constants determined from the increase in the absor-
bance maximum at 450 nm (viz., 2.2 ( 0.1 s-1 at 25 °C)
are in close agreement with that reported for the dissociation
of dithionite in reaction 1, viz., 2.5 s-1 at 25 °C.19 This
suggests that reaction 1 is clearly the rate-determining step,
which is followed by the rapid reduction of CoII(TSPc)4-
by the sulfur dioxide anion radical in reaction 2.
Experimental Section
Materials. Sodium nitrate, sodium nitrite, and sodium dithionite
(93% grade) were obtained from Aldrich and used as received.
Nitrous oxide was supplied by Linde. CoII(TSPc)4- was prepared
and purified using a literature method.18 Boric acid/potassium
chloride/sodium hydroxide, TRICINE, CHES, and CAPS buffers
were used to control the pH. Na15NO2 (95% 15N grade) and K15NO3
(13) Groves, J. T.; Roman, J. S. J. Am. Chem. Soc. 1995, 117, 5594.
(14) Cherry, J.-P. F.; Johnson, A. R.; Baraldo, L. M.; Tsai, Y.-C.; Cummins,
C. C.; Kryatov, S. V.; Rybak-Akimova, E. V.; Capps, K. B.; Hoff, C.
D.; Haar, C. M.; Nolan, S. P. J. Am. Chem. Soc. 2001, 7271.
(15) Liao, M. S.; Scheiner S. J. Chem. Phys. 2001, 114, 9780.
(16) Meshisuka, S.; Ichikawa, M.; Tamaru, K. J. Chem. Soc., Chem.
Commun. 1975, 9, 360.
(17) Grodkowski, J.; Dhanasekaran, T.; Neta, P.; Hambright, P.; Brunschwig
B. S.; Shinozaki, K.; Fujita, E. J. Phys. Chem. A 2000, 104, 11332.
(18) (a) Weber, J. N.; Busch, D. H. Inorg. Chem. 1965, 4, 469. (b) Yang,
Y.-C.; Ward, R. J.; Seiders, R. P. Inorg. Chem. 1985, 24, 1765.
(19) Creutz, C.; Sutin, N. Inorg. Chem. 1974, 13, 4041.
Inorganic Chemistry, Vol. 42, No. 2, 2003 619

Kudrik et al.
Reaction of CoI(TSPc)5- with Nitrite. A kinetic study
was undertaken under anaerobic conditions at pH 10 and 25
°C. The ESR spectrum of the products of the reaction
-
between CoI(TSPc)5- and an excess of NO2 (Figure S2,
Supporting Information) is similar to that reported for
CoII(TSPc)4-.22 This spectrum is in agreement with that
2
expected for a single, unpaired electron localized in the dz
orbital of a metal complex, which is essentially characteristic
for low-spin octahedral CoII(Pc) complexes.
The reaction was studied under pseudo-first-order condi-
tions, i.e., with NaNO2 in excess. It should be noted that an
excess of NaNO2 was maintained with respect to both
CoI(TSPc)5- and dithionite, since a direct reaction between
nitrite and CoI(TSPc)5- may occur when the dithionite is
used up. The change in absorbance at 450 nm was used for
the kinetic measurements. In this spectral region the phtha-
locyanine complexes of CoI have a much more intensive
absorption than those of CoII. The shape of the absorbance/
time plots at 450 nm depends on the [NaNO2]:[CoI(TSPc)5-]
ratio with nitrite in excess as shown in Figure 3.
Figure 3 demonstrates that the oxidation of CoI(TSPc)5-
by nitrite exhibits an induction period, which depends on
the nitrite concentration, i.e., on the rate of the redox reaction.
In fact, the induction period can be related to the redox
cycling of CoI(TSPc)5-, during which CoI(TSPc)5- is oxi-
dized by nitrite and the produced CoII complex is reduced
by the excess of dithionite present in solution. The induction
period depends linearly on the selected dithionite concentra-
tion at a fixed nitrite concentration as shown in Figure 4.
Once the dithionite is used up, the oxidation of CoI(TSPc)5-
to CoII(TSPc)4- by nitrite can be followed as a pseudo-first-
Figure 2. Time dependence of [Na2S2O4] during the reaction of NaNO2
with Na2S2O4 in the presence of CoII(TSPc)4-, [NaNO2] ) 3.5 × 10-2 M,
[CoII(TSPc)4-] ) 7.8 × 10-5 (a), 5.2 × 10-5 (b), and 2.6 × 10-5 (c) M,
pH 10, 25 °C.
-
S2O42- h 2SO2
(1)
CoII(TSPc)4- + SO2- + OH- f CoI(TSPc)5- + HSO3
-
(2)
Since CoI(TSPc)5- is diamagnetic, 1H NMR spectroscopy
was used to study its structure in solution. Chemical shifts
in the range 7.82-7.39 ppm (see Figure S1, Supporting
Information) correspond to the 12 protons of the isoindole
fragments. The 1H NMR spectrum indicates that the
CoI(TSPc)5- complex is a mixture of four regioisomers. It
should, however, be noted that the mentioned signals are
upfield shifted in comparison to those reported for other
tetrasubstituted phthalocyanines,20 probably due to distortion
of the planar structure of the macrocycle as a result of an
increase in the covalent radius of the central metal cation.
Reaction of Sodium Dithionite with Sodium Nitrite in
the Presence of Cobalt Tetrasulfophthalocyanine. Pre-
liminary experiments showed that the direct reaction between
dithionite and nitrite did not proceed with measurable rates
under conditions employed in this study (i.e., pH 10, 15-
50 °C). During the investigation of reactions catalyzed by
cobalt(II) tetrasulfophthalocyanine, we found that the ab-
sorbance observed at 315 nm did not depend on the oxidation
state of the cobalt complex. Although nitrite also absorbs at
315 nm, its molar absorbance coefficient (11 M-1 cm-1) is
much less than for dithionite21 (8043 M-1 cm-1). Since
NaNO2 was used in excess, changes in its absorbance at 315
nm during the reaction will be negligible. Therefore, the
observed decrease in absorbance at 315 nm is associated with
a decrease in the dithionite concentration (Figure 2). Interest-
ingly, the rate of the reaction does not depend on the
dithionite concentration, indicating that dithionite does not
participate in the rate-determining step of the catalytic
process. It follows that more attention must be given to
the other part of the process, viz., the reaction between
CoI(TSPc)5- and nitrite, since the rate does depend on the
CoI(TSPc)5- concentration as seen in Figure 2.
order reaction with the rate constant kobsd
.
The kinetics of the oxidation of CoI(TSPc)5- by nitrite
depends on the selected nitrite concentration as shown in
Figure 5. For nitrite concentrations up to 0.03 M, the
observed rate constant first increases linearly with increasing
nitrite concentration and then almost reaches a limiting value
as shown by the double reciprocal plot in Figure 5a. At higher
concentrations of nitrite, kobsd increases further with increas-
ing nitrite concentration as seen in Figure 5b. On the basis
of these observations, it is suggested that the reaction under
investigation proceeds via an inner-sphere electron-transfer
reaction which involves complex formation between CoI-
-
(TSPc)5- and NO2 , presented in reactions 3 and 4. Although
preliminary experiments showed that the band at 450 nm
did not depend on the concentration of nitrite, i.e., was
insensitive to the structure of the CoI(TSPc)5- complex, the
changes in absorbance at 450 nm are attributed to a decrease
in the concentration of the intermediate complex. The
limiting rate constant reached at a nitrite concentration of
approximately 0.03 M represents the electron-transfer rate
constant (k4) of the intermediate species. The increase in kobsd
observed at higher nitrite concentrations is ascribed to a redox
path that involves a further nitrite ion as reaction partner as
indicated in reactions 5 and 6.
(20) Stuzhin, P. A.; Khelevina, O. G. Coord. Chem. ReV. 1996, 147, 41.
(21) McKenna, C. E.; Gutheil, W. G.; Song, W. Biochim. Biophys. Acta
1991, 1075, 1091.
(22) (a) Rollmann L. D.; Iwamoto R. T. J. Am. Chem. Soc. 1968, 90, 1455.
(b) Rollmann L. D.; Chan S. I. Inorg. Chem. 1971, 10, 1978.
620 Inorganic Chemistry, Vol. 42, No. 2, 2003

Reduction of Nitrite and Nitrate by Dithionite
Figure 3. Time dependence of the absorbance of CoI(TSPc)5- at 450 nm during the reaction of NaNO2 with CoI(TSPc)5-, [NaNO2] ) 8.4 × 10-3 (a) and
3.0 × 10-2 (b) M, [CoI(TSPc)5-] ) 7.8 × 10-5 M, pH 10, 25 °C.
Figure 4. Dependence of the induction period on [Na2S2O4] at 25 °C and
pH 10, [CoI(TSPc)5-] ) 8.3 × 10-5 M, [NaNO2] ) 0.0081 M.
CoI(TSPc)5- + NO2- h CoI(TSPc)(NO2)6- (K3) (3)
CoI(TSPc)(NO2)6-
f
CoII(TSPc)4- + nitrite reduction products (k4) (4)
CoI(TSPc)(NO2)6- + NO2- h CoI(TSPc)(NO2)27- (K5) (5)
CoI(TSPc)(NO2)27- f
CoII(TSPc)(NO2)5- + nitrite reduction products (k6) (6)
This mechanism can qualitatively account for the complex
dependence of the reaction rate on the nitrite concentration
presented in Figure 5. In the low nitrite concentration range
(Figure 5a), reactions 3 and 4 account for the observed kinetic
behavior, and the observed first-order rate constant can be
expressed as given in eq 7. The nonlinear dependence of
kobsd on the nitrite concentration observed in Figure 5a
enables the estimation of K3 and k4, which have the values
102 ( 8 M-1 and 0.0121 ( 0.009 s-1 at 25 °C and pH 10,
respectively. At higher nitrite concentration, eq 7 can be
reduced to kobsd ) k4.
Figure 5. Dependence of the pseudo-first-order rate constant (kobsd
,
measured following the induction period in Figure 3) on [NaNO2] at 25 °C
and pH 10 for different nitrite concentration ranges: [CoI(TSPc)5-] )
7.8 × 10-5 M, [Na2S2O4] ) 8.1 × 10-4 M. For the lower concentration
range, the data are presented as a double reciprocal plot to demonstrate the
saturation kinetics almost reached under these conditions.
-
-
kobsd ) k4K3[NO2 ]/(1 + K3[NO2 ])
(7)
In the high nitrite concentration range (Figure 5b), reaction
4 competes with the nitrite-induced reaction path involving
Inorganic Chemistry, Vol. 42, No. 2, 2003 621

Kudrik et al.
Scheme 1. Proposed Mechanism for the Catalyzed Reduction of
-
NO2 in the 0-0.03 M Concentration Range
Figure 6. Dependence of k4 on [H+] at 25 °C, [CoI(TSPc)5-] ) 7.8 ×
10-5 M, [Na2S2O4] ) 6.3 × 10-4 M.
reactions 5 and 6, for which the rate expression is given in
eq 8. From the data reported in Figure 5b, a linear regression
fit resulted in k6K5 ) 0.188 ( 0.005 M-1 s-1, which is ca.
6 times smaller than the value of k4K3 at pH 10.
suggest that quite a similar situation could occur in the case
of natural CuI-containing nitrite reductases.
For more detailed mechanistic information, the thermal
activation parameters were determined at pH 10 from the
corresponding Eyring plots over the temperature range 15-
50 °C. For reaction 4, the following parameters were found
from the temperature dependence of k4 ()Kak7) calculated
from kobsd at [NaNO2] ) 0.03 M: ∆Hq ) 63 ( 4 kJ mol-1
and ∆Sq ) -89 ( 11 J K-1 mol-1. The complexity of the
system restricts quantitative interpretation of these activation
parameters. The large negative value of ∆Sq suggests a more
ordered transition state as compared to the intermediate
complex CoI(TSPc)(NO2)6-, presumably due to electron
transfer from the nonplanar CoI complex to the planar CoII
complex. 15N NMR spectra showed that, in comparison to
that for free 15NO2-, the signal for nitrite in CoII(TSPc)(NO2)5-
is shifted upfield by 175 ppm.
-
kobsd ) k6K5[NO2 ]
(8)
If k4 is a “pure” inner-sphere electron-transfer rate constant,
its value should not depend on pH. However, experiments
similar to those reported in Figure 5a were performed as a
function of pH and showed that k4 (the limiting rate constant)
strongly depends on pH in an alkaline medium (see Figure
6).
Figure 6 demonstrates that, for solutions with pH < 8.6,
k4 reaches a limiting value of 0.029 ( 0.006 s-1. In more
alkaline solutions (pH > 8.6), k4 strongly decreases with
increasing pH. This can only be due to the fact that the
electron-transfer reaction is preceded by protonation of the
complex such that k4 should be expressed by eq 9, where Ka
On the basis of the kinetic data, the catalytic reduction of
nitrite, including formation of the CoIINO complex, may be
summarized as in Scheme 1.
In Scheme 1, coordination of nitrite to the CoI complex is
followed by protonation and cleavage of the N-O bond of
nitrite to produce a hydroxyl group and NO bound to
hexacoordinate CoII. This scheme is quite similar to that
mentioned above for the CuI-containing nitrite reductase
H255N.24
The composition of the reaction products at a high excess
of nitrite as compared to CoI(TSPc)5- ([NaNO2]:[CoI-
(TSPc)5-] ) 104; [NaNO2]:[Na2S2O4] ) 1:0.8) was studied
using 15N NMR and mass spectrometry. Analysis of the gas
phase showed no presence of 15N-enriched products. In the
15N NMR spectra (see Figures S3 and S4 in the Supporting
Information), three new signals were observed at 53, -283,
and -382 ppm, respectively. The first signal corresponds to
the CoII complex (see above), the second to coordinated
15NH3, and the third to free 15NH3.25 These data show that
on the route from NO2- to NH3 there is no evidence for the
Ka[H+]
k4 )
k7
(9)
+
(
)
1 + Ka[H ]
is the protonation constant of CoI(TSPc)(NO2)6- and k7 is
the electron-transfer rate constant. At pH < 8.6, eq 9 can be
simplified to k4 ≈ k7 and k7 has the value 0.029 ( 0.006
s-1. The value of Ka was determined from the initial slope
()Kak7) of the plot in Figure 6 to be (1.24 ( 0.06) × 108
M-1. This is an important result of this study, since it shows
that the role of the catalyst is to shift the reaction into a
more alkaline pH range, which even indicates the possibility
for its realization at physiological pH. The complex in this
case has the role of an electron donor to increase the electron
density on the coordinated nitrite fragment. The decomposi-
tion of the protonated complex CoI(TSPc)(NO2H)5- results
in the formation of CoII(TSPc)4- and the hydronitrite radical
NO22-, which is known to be a product of the reduction of
nitrite and rapidly decays to NO in aqueous solution.23 We
(23) Lymar, S. V.; Schwarz, H. A.; Czapski, G. J. J. Phys Chem. A 2002,
106, 7245.
(24) Boulanger, M. J.; Murphy M. E. P. Biochemistry 2001, 40, 9132.
622 Inorganic Chemistry, Vol. 42, No. 2, 2003

Reduction of Nitrite and Nitrate by Dithionite
reaction products. It is important to note that the analysis of
the products using 15N NMR and mass spectrometry gave
unexpected results. Contrary to those found for the reduction
of nitrite, the products of the reaction between CoI(TSPc)5-
and nitrate are N2 and N2O. Dinitrogen was detected in the
gas phase and nitrous oxide in solution. The nitrogen atoms
in 15N2O give two signals in the 15N NMR spectrum (see
Figure S5 in the Supporting Information) at -127.8 and
-237.0 ppm (in the gas phase they have signals at -147.3
and -237.0 ppm, respectively27). The observed upfield shift
may be accounted for by the participation of N2O in hydrogen
bond formation. It should be noted that 15NH3 was not
observed at all. On the basis of these data, the reduction of
nitrate can be represented as shown in reaction 11.
NO3- f NO2 f NO2- f NO f N2O f N2 (11)
Figure 7. Spectral changes observed during the reaction of CoI(TSPc)5-
with sodium nitrate, [CoI(TSPc)5-] ) 7.8 × 10-5 M, [NaNO3] ) 0.32 M,
induction period omitted for clarity, 35 °C. The spectra were recorded at
120 s intervals.
No simple kinetics was observed for this reaction (see
Figures S6 and S7 in the Supporting Information). The
dependence of the pseudo-zero-order rate constant (kobsd) on
[NaNO3] reported in Figure S7 differs significantly from that
observed for nitrite in Figure 5. It may be explained in terms
of different electrophilic properties of nitrite and nitrate, since
the latter is a significantly more powerful electrophile with
electron density being localized on the nitrogen atom.
Therefore, interaction with the electron-rich CoI center is
more favorable in the case of nitrate, the intermediate
complex CoI(TSPc)(NO3)6- being more stable and presum-
ably containing a CoI-O bond. This leads to a more
pronounced catalytic effect of CoI(TSPc)5- in the reduction
of nitrate since the formation of the intermediate complex
proceeds more readily and does not slow the overall process.
The different types of complexes formed, viz., a CoI-N bond
for nitrite and a CoI-O bond for nitrate, are suggested to
account for the different observed reaction products of the
catalytic redox process. The hypothesis of two reaction
centers in the N-O species has been confirmed recently by
results from pulse radiolysis studies of nitrite.23 It was shown
that the products obtained from the reduction of nitrite ion
by solvated electrons and by hydrogen atoms are not
connected through a rapid protic equilibrium as previously
believed. The authors assumed that the H atom quantitatively
formation of N2O and N2 under our experimental conditions.
The overall reaction sequence for nitrite reduction is given
by reaction 10.
NO2- f NO f NO- f HNO- f H2NO- f NH3 (10)
Accordingly, the second molecule of nitrite plays a
catalytic role by accelerating electron transfer and breakage
of the NO bond on the nitrite fragment of the intermediate
complex, accompanied by formation of NO. For less excess
of nitrite, the composition of the reaction products does not
change, indicating no presence of N2O and N2. These data
show that, on variation of the excess of nitrite employed,
the intermediate formation of nitric oxide can be suggested
since it is reduced to ammonia, but it is not formed as a
free compound since the stability constant of the [CoII-
(TSPc4-)(NO)] complex is 1.3 × 108 M-1.26
Reaction of CoI(TSPc)5- with Nitrate. The CoI(TSPc)5-
complex reacts with sodium nitrate at pH 10 with formation
of an EPR-active CoII(TSPc)4- species. The changes in the
UV-vis spectra during the reaction are shown in Figure 7.
Similar to the reaction with nitrite, oxidation of the CoI
complex by nitrate is accompanied by a decrease in absor-
bance at 450 nm. When nitrate is used in a significant excess,
the main part of the kinetic curve following the induction
period, where at least 50% of the CoI complex reacts, has a
pseudo-zero-order character with the following activation
parameters: ∆Hq ) 128 ( 5 kJ mol-1 and ∆Sq ) +77 (
17 J K-1 mol-1. These activation parameters cannot be
interpreted in a quantitative way, although the positive value
of ∆Sq may indicate that the rate-determining step involves
breakage of the NO bond followed by formation of the
-
reacts by addition to the unsaturated N atom of NO2 ,
whereas the NO22- radical is always protonated at its O atom.
Reaction of CoI(TSPc)5- with Nitrous Oxide. As re-
ported above, N2 and N2O were formed during the reduction
of nitrate. However, formation of N2 is only possible if
reduction of N2O occurs under similar conditions. To
investigate this further, a study was undertaken under
anaerobic conditions at pH 10 and 25-50 °C. N2O-saturated
buffer solutions were used in these experiments. Addition
of N2O to solutions of CoI(TSPc)5- in the presence of
dithionite (reaction 12) resulted in spectral changes reported
in Figure 8, which correspond to the stepwise oxidation of
CoI(TSPc)5-. A decrease in the peak at 450 nm and an
increase in the peak at 643 nm correspond to the formation
(25) (a) Martin, G. J.; Martin, M. L.; Gouesnard, J.-P. 15N NMR
Spectroscopy; Springer-Verlag: Berlin, Heidelberg, Germany, 1981;
382 pp. (b) Andersson, L.-O.; (Banus) Mason, J.; van Bronswijk, W.
J. Chem. Soc. A 1970, 296. (c) Chen, Y.; Lin; F.-T.; Shepherd, R. E.
Inorg. Chem. 1999, 38, 973.
(26) Zilbermann, I.; Hayon, J.; Katchalski, T.; Ygdar, R.; Rishpon, J.;
Shames, A. I.; Korin, E.; Bettelheim, A. Inorg. Chim. Acta 2000, 305,
53.
(27) Mastikhin, V. M.; Mudrakovsky, I. L.; Filimonova, S. V. Chem. Phys.
Lett. 1988, 149, 175.
Inorganic Chemistry, Vol. 42, No. 2, 2003 623

Kudrik et al.
Figure 8. Spectral changes observed during the reaction of CoI(TSPc)5- with nitrous oxide, [CoI(TSPc)5-] ) 8.3 × 10-5 M, [N2O] ) 1/3 saturated,
induction period omitted for clarity, 50 °C. The spectra were recorded at 300 s intervals.
of the ESR-active CoII complex with an ESR spectrum
similar to that reported in Figure S2.
tion is the fact that there are apparently no reliable methods
presently available to determine N2O in aqueous solutions.
Conclusions
CoI(TSPc)5- + N2O f CoII(TSPc)4- + N2O- (12)
Our study has clearly shown that cobalt tetrasulfophtha-
locyanine is an effective catalyst for the reduction of nitrite
and nitrate by dithionite. The catalytic cycle includes
reversible reduction of CoII T CoI and reduction of the
coordinated substrate. Surprisingly, reduction of nitrite and
nitrate leads to the formation of different products, although
NO2- is an intermediate of the nitrate reduction process. This
finding suggests that the structure of the complex between
nitrite or nitrate and CoI(TSPc)5- determines the composition
of the final reaction products. In our opinion, the most
plausible structure of the intermediate complex in the case
A subsequent decrease in the peak at 643 nm and an
increase in the peak at 673 nm (see Figure 8) correspond to
the formation of the ESR-silent CoIII complex. The possibility
of a one-electron reduction of N2O has been demonstrated
recently by our kinetic studies on the reaction between
sulfoxylate, SO22-, and nitrous oxide in alkaline solutions28
(thiourea dioxide served as the precursor of sulfoxylate). In
the course of this reaction we observed the formation of
dithionite, which can be explained by reaction 13 and the
reverse of reaction 1.
-
of nitrite is CoI(TSPc)5-fNO2 . Following protonation and
SO22- + N2O f SO2- + N2O-
(13)
inner-sphere electron transfer, the CoIIrNdO complex is
formed, which is followed by reduction of coordinated NO
and production of ammonia. Contrary to that of nitrite,
ligation of nitrate presumably leads to formation of a complex
N2O- is a very unstable species and decomposes im-
mediately to OH and N2, which are known to be products
of the reaction of N2O and hydrated electrons.29 Therefore,
the observed decrease in the peak at 643 nm and an increase
in the peak at 673 nm, corresponding to formation of the
ESR-silent CoIII complex, cannot be explained by the direct
interaction between CoII(TSPc)4- and N2O- (it should be
noted that N2O itself does not react with the CoII complex).
The nature of the species that oxidizes CoII(TSPc)4- is
presently unknown.
Figure S8 in the Supporting Information shows that the
duration of the induction period at 450 nm depends linearly
on the initial concentration of dithionite. These data indicate
that nitrous oxide can be reduced by dithionite in the presence
of CoI(TSPc)5-. Unfortunately, we could not obtain quantita-
tive kinetic data for this reaction since the kinetic curves
are extremely irreproducible (excluding induction periods)
and sensitive to the initial concentrations of dithionite and
N2O (Figure S9, Supporting Information). Another complica-
-
with a CoI-O bond: (TSPc)5-CoIfOdNO2 . Further re-
duction results in the formation of the complex (TSPc)4-CoIf
OdN, where NO is coordinated via oxygen. The reduction
of coordinated nitric oxide gives OdN-, which dimerizes
and produces N2O. Thus, N-coordination of the substrate
leads to formation of ammonia, whereas O-coordination leads
to the formation of nitrogen. In both processes nitric oxide
is a reaction intermediate. Nitrous oxide can also be reduced
under similar conditions, and this reaction may be relevant
for the purification of gases containing N2O.
Acknowledgment. We gratefully acknowledge financial
support from the Deutsche Forschungsgemeinschaft (SFB
583 “Redox-Active Metal Complexes: Control of Reactivity
via Molecular Architecture”) and DAAD for fellowships to
E.V.K. and S.V.M.
Supporting Information Available: Three 15N NMR spectra
1
and one H NMR spectrum recorded during the study, one ESR
spectrum, and kinetic data. This material is available free of charge
via the Internet at http://pubs.acs.org.
(28) Makarov, S. V.; Kudrik, E. V.; van Eldik, R.; Naidenko, E. V. J. Chem.
Soc., Dalton Trans. 2002, 4074.
(29) Huie, R. E.; Clifton, C. L.; Altstein, N. Radiat. Phys. Chem. 1989,
33, 361.
IC020290F
624 Inorganic Chemistry, Vol. 42, No. 2, 2003