Experimental study of decomposition of aqueous nitrosyl thiocyanate

Article abstract of DOI:10.1021/ic102445d

This study has examined the kinetics of the decomposition of nitrosyl thiocyanate (ONSCN) by stopped flow UV-vis spectrophotometry, with the reaction products identified and quantified by infrared spectroscopy, membrane inlet mass spectrometry, ion chromatography, and CN^- ion selective electrode. The reaction results in the formation of nitric oxide and thiocyanogen, the latter decomposing to sulfate and hydrogen cyanide in aqueous solution. The rate of consumption of ONSCN depends strongly on the concentration of SCN ^- ions and is inhibited by nitric oxide. We have developed a reaction mechanism that comprises three parallel pathways for the decomposition of ONSCN. At high thiocyanate concentrations, two reaction pathways operate including a second order reaction to generate NO and (SCN)₂ and a reversible reaction between ONSCN and SCN^- producing NO and (SCN)₂^-, with the rate limiting step corresponding to the consumption of (SCN)₂^- by reaction with ONSCN. The third reaction pathway, which becomes significant at low thiocyanate concentrations, involves formation of a previously unreported species, ONOSCN, via a reaction between ONSCN and HOSCN, the latter constituting an intermediate in the hydrolysis of (SCN)₂. ONOSCN contributes to the formation of NO via homolysis of the O-NO bond and subsequent dimerization and hydrolysis of OSCN. Fitting the chemical reactions of the model to the experimental measurements, which covered a wide range of reactant concentrations, afforded estimation of all relevant kinetic parameters and provided an excellent match. The reaction mechanism developed in this contribution may be applied to predict the rates of NO formation from ONSCN during the synthesis of azo dyes, the gassing of explosive emulsions, or nitrosation reactions occurring in the human body.

Full text of DOI:10.1021/ic102445d

ARTICLE
pubs.acs.org/IC
Experimental Study of Decomposition of Aqueous
Nitrosyl Thiocyanate
Mark S. Rayson, John C. Mackie, Eric M. Kennedy, and Bogdan Z. Dlugogorski*
Process Safety and Environment Protection Research Group, School of Engineering, The University of Newcastle,
Callaghan, NSW 2308, Australia
S Supporting Information
b
ABSTRACT: This study has examined the kinetics of the decomposition of nitrosyl thiocyanate
(ONSCN) by stopped flow UVꢀvis spectrophotometry, with the reaction products identified and
quantified by infrared spectroscopy, membrane inlet mass spectrometry, ion chromatography, and
CNꢀ ion selective electrode. The reaction results in the formation of nitric oxide and thiocyanogen,
the latter decomposing to sulfate and hydrogen cyanide in aqueous solution. The rate of
consumption of ONSCN depends strongly on the concentration of SCNꢀ ions and is inhibited by nitric oxide. We have developed
a reaction mechanism that comprises three parallel pathways for the decomposition of ONSCN. At high thiocyanate concentrations,
two reaction pathways operate including a second order reaction to generate NO and (SCN)2 and a reversible reaction between
ONSCN and SCNꢀ producing NO and (SCN)2ꢀ, with the rate limiting step corresponding to the consumption of (SCN)2ꢀ by
reaction with ONSCN. The third reaction pathway, which becomes significant at low thiocyanate concentrations, involves
formation of a previously unreported species, ONOSCN, via a reaction between ONSCN and HOSCN, the latter constituting an
intermediate in the hydrolysis of (SCN)2. ONOSCN contributes to the formation of NO via homolysis of the OꢀNO bond and
subsequent dimerization and hydrolysis of OSCN. Fitting the chemical reactions of the model to the experimental measurements,
which covered a wide range of reactant concentrations, afforded estimation of all relevant kinetic parameters and provided an
excellent match. The reaction mechanism developed in this contribution may be applied to predict the rates of NO formation from
ONSCN during the synthesis of azo dyes, the gassing of explosive emulsions, or nitrosation reactions occurring in the human body.
’ INTRODUCTION
wide range of nitrosation reactions, with elevated reaction rates
induced by an increase in the concentration of nitrosating agents
in the system. Thiocyanate ions, in particular, represent powerful
catalysts owing to the large equilibrium constant for the forma-
tion of nitrosyl thiocyanate, ONSCN, and in many cases the
catalyzed rate of nitrosation can be many orders of magnitude
faster than the uncatalyzed rate.11ꢀ15 For a wide variety of
substrates, the formation of ONSCN is sufficiently fast for this
species to reach its equilibrium concentration as shown below:16
Nitrosation reactions, in which the nitroso (NO+) group
transfers from a reactant to a substrate, are important in a
diverse range of industrial, environmental, and biological
applications. Industrially, nitrosation reactions play important
roles in the synthesis of azo dyes,1 the production of hydro-
xylamine,2 the sensitization of emulsion explosives,3 and have
potential to assist in removal of wax deposits in oil pipelines.4
In biology, nitrosation reactions have received great attention
owing to the importance of S-nitrosothiols as a source of nitric
oxide in the human body,5,6 and the formation of carcinogenic
N-nitrosamines, which can occur under acidic conditions in the
human body or in the environment.7ꢀ9 Under acidic condi-
tions, nitrosation reactions proceed via the nitrosating agents
Hþ þ HNO2 þ SCN- h ONSCN þ H2O
ð3Þ
½ONSCNꢁ
KONSCN
¼
ð4Þ
½Hþꢁ½HNO2ꢁ½SCNꢀꢁ
N2O3 and H2NO2 , derived from nitrous acid.10 N2O3 serves
+
The evidence for the existence of ONSCN comes from
spectroscopic and kinetic studies, which both indicate the pre-
sence of a species formed according to eq 3. Acidic solutions of
nitrous acid containing thiocyanate ions display a characteristic
red color, the intensity of which depends on the concentrations
of H+, SCNꢀ, and HNO2. The red colored species exhibits a
broad absorption band with a maximum at 460 nm (ε460 = 100
as the primary nitrosating agent under mildly acidic condi-
tions (pH > 1), while H2NO2+ dominates under highly acidic
conditions.
HNO2 þ NO2 þ Hþ h N2O3 þ H2O
ð1Þ
ꢀ
M
ꢀ1 cmꢀ1), and an equilibrium constant of 32 Mꢀ1 at 25 °C.17
HNO2 þ Hþ h H2NO2
ð2Þ
þ
First-order dependencies of H+, SCNꢀ, and HNO2 concentra-
tions on the rate of catalytic reaction demonstrate that the
In the presence of added nucleophiles, such as Clꢀ, Brꢀ, Iꢀ,
and SCNꢀ, a new nitrosation pathway operates as a consequence
of the formation of nitrosyl halides, denoted ONX (where X = Cl,
Brꢀ, Iꢀ, or SCNꢀ). The formation of nitrosyl halides catalyzes a
Received: December 7, 2010
Published: July 20, 2011
r
2011 American Chemical Society
7440
dx.doi.org/10.1021/ic102445d Inorg. Chem. 2011, 50, 7440–7452
|

Inorganic Chemistry
ARTICLE
abundance of ONSCN accounts for both the visible absorption
and the catalysis of nitrosation reactions by thiocyanate ions.16
Our interest in nitrosyl thiocyanate stems from the potential
for this species to produce nitric oxide as a side product during
thiocyanate-catalyzed nitrosation reactions, in particular, the
sensitization of emulsion explosives, wherein thiocyanate ions
catalyze the nitrosation of ammonia (from ammonium nitrate)
forming small bubbles of nitrogen gas within the explosive.
These bubbles undergo adiabatic compression during explosive
detonation, acting as hot spots to propagate the explosion front
through the bulk explosive.18 Whereas the sensitization reac-
tion produces primarily N2, a fraction of the gas generated
consists of NO and NO2, which can pose a safety hazard to
miners prior to explosive detonation, particularly when used in
confined underground mines.
(1.2 ꢂ 10ꢀ3 M atmꢀ1). Further details of this procedure are contained
in the Supporting Information.
Instrumental Methods and Procedures. Kinetics: Stopped
Flow UVꢀvis. An RX-2000 rapid mixing accessory from Applied
Photophysics was coupled to a Varian Cary50 UVꢀvisible spectrometer
to study the decomposition kinetics of ONSCN. A constant-tempera-
ture circulating bath supplied the water to surround the drive syringes of
the RX-2000 and maintain them at a preset temperature. A thermo-
statted cell holder, regulated by a Varian Cary single-cell Peltier
temperature controller, housed the observation cell. All experiments
were performed at 298 K. Deoxygenation of reactant solutions was
achieved by bubbling with nitrogen gas for at least 15 min, with longer
degassing times yielding identical results. Prior to each experiment, we
flushed the stopped flow apparatus with deoxygenated water and
acquired a baseline absorbance reading, storing it in the Varian software
to be automatically subtracted from subsequent measurements.
ONSCN was produced in situ, by reacting a dissolved NaNO2 and
NaSCN solution with a solution of perchloric acid. The absorbance at
460 nm (ε460 = 100 Mꢀ1 cmꢀ1) served to determine the concentration
of ONSCN.
ONSCN, known to be unstable in pure form, decomposes
readily even at temperatures approaching ꢀ60 °C to form nitric
oxide and thiocyanogen.19
2ONSCN f 2NO þ ðSCNÞ2
ð5Þ
Analysis of Gaseous Products: FTIR and MIMS. Fourier transform
infrared spectroscopy (FTIR) of product gases was performed on a
Varian IR-600 with 0.5 cmꢀ1 resolution and 10 m path length cell
(Infrared Analysis Inc.). We generated gas samples in a sealed 50-mL
glass reactor initially containing a deoxygenated solution of SCNꢀ and
NO2ꢀ at pH > 7. Reaction was initiated by addition of a small volume of
2 M HClO4 through a septum side port in the reactor, forming ONSCN
in situ. The product gases were purged from the reactor with high-purity
nitrogen gas, dried by passage over a column of silica gel and collected in
a 3-L gas transfer bag for analysis. Separate experiments established that
silica gel did not adsorb NO. Prior to recording a background spectrum,
the gas cell was subjected to three cycles of evacuation and filling with
high-purity nitrogen. The cell was then loaded with the sample for
scanning, with the background automatically subtracted from the sample
spectrum by the Varian software. Gases were quantified using the
QASoft program (Infrared Analysis Inc.) which applies a region
integration and subtraction (RIAS) method to determine the concen-
trations based on standard spectra from the QASoft database.
The membrane inlet mass spectroscopy (MIMS)26 experiments
employed a Pfeifer Thermostar quadrupole mass spectrometer to detect
the gases diffusing from the aqueous solution. A 1-cm length of
semipermeable silastic tubing housed the end of the MS capillary,
allowing direct measurement of dissolved gases. In these experiments,
the solution was degassed with high purity argon to remove oxygen and
nitrogen present from air. The purge gas was turned off at the start of the
experiment.
Dilute aqueous solutions of ONSCN are relatively stable,
however, the red color of the solutions fades over time with
evolution of nitric oxide.20 Despite its widespread applications
as a catalyst for nitrosation reactions, to our knowledge, the
kinetics and mechanism of ONSCN decomposition have not
been studied in detail. Thus, we are interested to determine the
kinetics and mechanism of ONSCN decomposition, allowing
us to assess the contribution of ONSCN to the formation of
NOx during explosive sensitization. In addition to its industrial
applicability, ONSCN has the potential to be a source of nitric
oxide in living systems. Knowledge of the mechanism and rate
of ONSCN decomposition will enable an assessment of its
contribution to NO formation in biological systems. Further to
its role as a catalyst for nitrosation reactions, ONSCN has been
detected during the oxidation of thiocyanate ions by nitric
acid,21,22 however, the role of ONSCN in the reaction mechan-
ism is unclear. Knowledge of the rate and mechanism of
ONSCN decomposition will also assist in elucidating this
reaction pathway.
’ EXPERIMENTAL SECTION
Reagents and Solutions. Sodium nitrite (>99.5%), sodium
thiocyanate (>98%), and sodium perchlorate (ACS reagent, 98%)
were obtained from Sigma Aldrich, whereas perchloric acid (analytical
reagent grade) was purchased from Ajax. Reactant solutions were
prepared by dissolving a known mass of solid in distilled deionized
water. In the case of NaSCN, the solid was dried in an oven at 150 °C to
constant weight before use.23 Sodium perchlorate was stored in a
desiccator to prevent moisture absorption. Solutions of perchloric acid
of the desired concentration were produced by diluting 70% HClO4 in
distilled deionized water and were standardized by titrating against
Na2CO3. The ionic strength of all reactant solutions was adjusted to
1 M using NaClO4, except for the initial rates study in which the ionic
strength was 2 M.
Nitric oxide (NO) solutions were prepared by first degassing the
target solution with N2 to remove oxygen, prior to bubbling with NO
generated via the reduction of nitrous acid by ascorbate.24 Dissolved
NO concentrations were determined by UV absorbance (as nitrite)
subsequent to reaction of a small aliquot of NO solution with water
containing dissolved oxygen. Results coincided with literature values25
for NO saturation in water (1.9 ꢂ 10ꢀ3 M atmꢀ1), and in 1 M NaClO4
Analysis of Aqueous Products: Ion Chromatography and Ion
Selective Electrode. Analysis of ionic products was performed on a
Dionex DX-100 ion chromatograph with an eluent containing 8 mM
Na2CO3 and 1 mM NaHCO3, an Ionpac AS14A analytical column and
AG14A guard column, with suppressed conductivity detection. We
employed the same reactor setup for these experiments as was used
for generating gases for the FTIR analysis. In these experiments, HCl
functioned to acidify the solution rather than HClO4 due to the
ꢀ
incompatibility of ClO4 with the IC column. Samples were drawn
from the reactor using a gastight syringe via a septum side port and
quenched with 0.1 M NaOH to cease the reaction, before being diluted
ꢀ
between 20 and 50 times for analysis. Concentrations of NO2ꢀ, NO3
,
and SO42- were determined by comparing the peak area and retention
time of the sample solution to a calibration plot generated using known
standards. A Nico2000 cyanide ion selective electrode and double
junction reference electrode, connected to a Hanna 213 pH/mV meter
via a Nico2000 dual electrode head, determined the cyanide ion concen-
tration after the sample was diluted in 0.1 M NaOH. The electrode
7441
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
system was calibrated with cyanide solutions prepared by dissolving solid
NaCN in 0.1 M NaOH and diluting to the appropriate concentration.
Data Analysis. We fitted the measurements for the first 5 min of
experiments to potential reaction mechanisms using the DynaFit
program.27 The mechanism was written into a Dynafit script file, and
rate constants were included for all known reactions. In the case of
protonation equilibria, we followed the same approach as da Silva et al.28
whereby the association reaction between the hydrogen ion and a
to that of reaction 10. We therefore attempted to analyze both
the aqueous and gaseous reaction products to determine if this is
indeed the correct stoichiometry.
3ðSCNÞ2 þ 4H2O f 5SCNꢀ þ SO4 þ HCN þ 7Hþ
2ꢀ
ð9Þ
ꢀ
2ꢀ
negatively charged base (i.e., reaction between NO2 and H+) was
6ONSCN þ 4H2O f 6NO þ 5SCNꢀ þ SO4
assumed to occur with a rate of 1010 Mꢀ1 ꢀ1, with the dissociation
s
þ HCN þ 7Hþ
ð10Þ
reaction rate constant determined from the equilibrium constant. The
DynaFit program automatically generates a system of first-order ordin-
ary differential equations describing the reaction mechanism, and fits the
rate constants in these equations to minimize the error between the
model and the experimental data using a LevenbergꢀMarquardt algo-
rithm. For each potential mechanism, several initial guesses were trialed
to ensure the convergence of the program to a global minimum.
Under the experimental conditions of this study, conversion of
HNO2 to ONSCN is incomplete, and absorbance readings must be
corrected for this when determining the overall rate of nitrous acid
consumption (or NO generation). Equation 6 relates the concentration
of ONSCN to the total nitrous acid concentration:
Analysis of Gaseous Reaction Products Using FTIR and MIMS.
As quantified by FTIR, nitric oxide accounted for 87% (245
ppm) of the detected gases. A considerable amount of NO2 also
formed, corresponding to 10% (27 ppm) of the products as a
result of NO oxidation by a small volume of air contaminating the
sample bag. Thus, in complete absence of oxygen, NO would
constitute 97% of the gaseous reaction products.
Unexpectedly, we observed a small amount of N2O, corre-
sponding to 2.5% (7 ppm) of the gaseous products. This result
indicates a side reaction leading to N2O rather than NO, as
outlined in the Discussion. A small amount of CO2 also appeared
in the analyzed mixture, corresponding to less than 1% (1.5 ppm)
of the gases detected. The very low concentration of CO2 made it
impossible to determine whether CO2 arises in a side reaction, or
KONSCN½SCNꢀꢁ½Hþꢁ
½ONSCNꢁ ¼
½HNO2ꢁTotal
ð6Þ
1 þ KONSCN½SCNꢀꢁ½Hþꢁ
Under the conditions of the present study, with high H+ and relatively
low nitrous acid concentrations, the concentrations of nitrite ion and
dinitrogen trioxide are negligible, and the total nitrous acid concentra-
tion is defined according to eq 7.
ꢀ
whether it results from the liberation of dissolved CO2/HCO3
upon acidification of the reactant solution. The yields of CO2 and
N2O are sufficiently small that the reactions generating these
species have a negligible impact on the overall kinetics of
ONSCN decomposition.
The gases were also analyzed by MIMS to determine if any N2
formed as a product of ONSCN decomposition. The reaction
solution was degassed with argon prior to generation of ONSCN
via the addition of a small volume of HClO4 to a sealed reactor
housing the MIMS probe. The results showed a rapid increase in
the ion current at m/z = 30 corresponding to the production of
NO. The signal at m/z = 28 remained constant throughout the
experiment, indicating that no N2 formed as a result of ONSCN
decomposition.
Analysis of Aqueous Products. Ion chromatography afforded
the analysis of the ionic products, with the exception of CNꢀ, as
quantified with an ion selective electrode. The nitrite concentra-
tion decreased with increasing decomposition time, owing to the
decomposition of ONSCN. Nitrate was detected at concentra-
tions corresponding to less than 1% of the initial nitrite added,
with the nitrate concentration initially increasing with time
before decreasing slowly. This small amount of nitrate formed
from the decomposition of nitrous acid. This process is reversible
under the experimental conditions and explains why the con-
centration of nitrate displays nonmonotonic concentration dur-
ing experiments.30
The concentration of sulfate increased with time, with the
final yield enclosed between 75 and 90% of the stoichiometric
conversion according to reaction 10, for acid concentrations
ranging from 0.18 to 0.33 M. Under the same conditions, the
final concentration of cyanide varied from 35 to 50% of
that expected for complete conversion according to reaction
10. The results appear very similar to those of a prior study
on the oxidation of SCNꢀ by HNO3, which also proceeds via
an (SCN)2 intermediate.31 In that study, Stedman and Win-
cup reported lower than expected yields of sulfate and HCN,
findings consistent with a competing reaction between (SCN)2
½HNO2ꢁTotal ¼ ½HNO2ꢁ þ ½ONSCNꢁ
’ RESULTS
ð7Þ
1. Equilibrium Constant for the Formation of ONSCN.
Applying the molar absorptivity of 100 Mꢀ1 for ONSCN
reported by Stedman and Whincup,17 we determined the equili-
brium constant for the formation of nitrosyl thiocyanate at an
ionic strength of 1 M as 22.1 ( 0.7 Mꢀ2. This value agrees with
those of Jones et al.21 (KONSCN = 22 Mꢀ2) and Bazsa and
Epstein22 (KONSCN = 20.4 Mꢀ2), measured in 1 M perchloric
and nitric acids, respectively. Note that KONSCN constitutes an
apparent equilibrium constant as it involves no correction for the
effect of nonunity activity coefficients. The apparent molar
absorptivity of nitrosyl thiocyanate increases at high thiocyanate
ion concentrations. Doherty et al. investigated this increase in
absorbance and found it to be consistent with formation of an
adduct between SCNꢀ and ONSCN.
ꢀ
ONSCN þ SCNꢀ a ONðSCNÞ2
ð8Þ
ꢀ
The absorbance of ON(SCN)2 must be taken into account
when determining the concentration of ONSCN, especially
at higher SCNꢀ concentrations where reliance on the
100 Mꢀ1 cmꢀ1 extinction coefficient would result in consider-
able error. Supporting Information provides further details of
these experiments.
2. Reaction Products. The decomposition of pure nitrosyl
thiocyanate produces NO and (SCN)2.19 Aqueous solutions of
(SCN)2 are unstable,29 decaying to form SCNꢀ, SO42-, and
HCN (reaction 9). Therefore, if ONSCN decomposition in
aqueous solution proceeds with the stoichiometry of reaction 5,
the overall stoichiometry of ONSCN disappearance corresponds
7442
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
Figure 1. Spectrum of (SCN)2 generated from 0.1 M SCNꢀ, 1 M HClO4, and 0.9 mM NaClO. The spectrum shown was calculated by subtracting the
spectrum after 1 h (at which time (SCN)2 had been almost completely hydrolyzed) from the initial spectrum.
and CNꢀ forming sulfur dicyanide, S(CN)2.
ðSCNÞ2 þ CNꢀ f SðCNÞ2 þ SCNꢀ
During the initial stages of reaction, when the concentration of
HCN is low, the formation of HCN and sulfate advances in
accordance with reactions 9 and 10, which strongly suggests the
intermediacy of (SCN)2.
3. Spectrum of Intermediate Species. The formation of
sulfate and cyanide ions suggests that ONSCN decomposition
proceeds through an (SCN)2 intermediate. (SCN)2 has been
identified as a product of the oxidation of SCNꢀ by a variety
of oxidizing species, including H2O2,32 HNO3,31 ClO2,33 and
HOCl,34 and has a UV spectrum with an absorption maxi-
mum near 290 nm. By following the procedure outlined by
Barnett et al.,29 we recorded the spectrum of (SCN)2 to confirm
that it did not extend into the visible region where it might
interfere with the absorbance of ONSCN, and to aid in identify-
ing (SCN)2 as a reaction intermediate. This involved the
oxidation of a NaSCN solution by acidified NaOCl to generate
(SCN)2 as shown below:
for the initial absorbance. In the absence of additional absorbing
species, the decomposition of ONSCN should therefore result in
a consistent decrease in absorbance across all wavelengths. This
was the case only for wavelengths above 400 nm, with absorbance
at lower wavelengths decreasing at a slower rate. Assuming that
ONSCN was solely responsible for the absorption at 460 nm, the
additional absorbance caused by the reaction intermediate at
other wavelengths was calculated and compared to that of
(SCN)2 in Figure 2, after scaling the spectrum to a common
absorbance. The spectrum of the intermediate concurs with the
spectrum of (SCN)2 generated in the independent experiment,
indicating that (SCN)2 indeed represents the observed inter-
mediate. The absorbance of the intermediate below 310 nm
exceeds that of (SCN)2, suggesting the presence of an additional
species absorbing in this region. However, the data at these
wavelengths display considerable scatter, induced by high absor-
bance of ONSCN, which confined the measurements to wave-
lengths longer than 300 nm.
4. Effect of Initial Dissolved Oxygen on Reaction Kinetics.
Initial kinetic experiments with solutions that had not been
degassed exhibited an induction period of slow decomposition
during the early stages. Bubbling nitrogen gas through the
reactant solutions for 15 min prior to starting the reaction
removed the induction period, as demonstrated in Figure 3.
The induction period arises as a consequence of dissolved
oxygen, initially present in the solutions from the atmosphere,
reacting with nitric oxide to form NO2. NO2 combines with NO
and water to regenerate nitrous acid, and as nitrous acid exists in
equilibrium with ONSCN, the concentration of ONSCN
remains unchanged, as illustrated in Scheme 1.
ð11Þ
HOCl þ 2SCNꢀ þ Hþ f ðSCNÞ2 þ Clꢀ þ H2O
ðor equivalently Cl2 þ 2SCNꢀ f ðSCNÞ2 þ 2ClꢀÞ
ð12Þ
The reaction was performed with 1 M HClO4, 0.9 mM NaOCl,
and SCNꢀ concentrations of 0.1 and 0.2 M, conditions leading to
reasonably stable solutions of (SCN)2 and reliable measure-
ments of its spectrum. Figure 1 confirms a negligible absor-
bance of (SCN)2 above 420 nm that does not interfere with the
measurements acquired at 460 nm, the absorbance maximum
of ONSCN.
Nonuniform changes to the spectra of decomposing ONSCN
solutions indicated the formation of a UV absorbing species
not present at the start of the experiments. To isolate the spec-
trum of the intermediate, we subtracted the spectrum of
SCNꢀ and assumed that species in equilibrium with nitrous acid
(e.g., HNO2, ONSCN, and ON(SCN)2ꢀ) accounted exclusively
Therefore, to avoid complications owing to dissolved oxygen,
we degassed all reactant solutions by bubbling high-purity nitro-
gen for at least 15 min. This was sufficient to remove the indu-
ction period, however, some experiments showed a miniscule but
noticeable rise in the curve suggesting a residual contamination
of the solutions by a minute amount of oxygen. This rise, which
7443
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
Figure 2. Comparison of spectrum of intermediate species and (SCN)2; intermediate spectrum from experiment with 3 mM NaNO2, 0.2 M HClO4,
and 0.2 M NaSCN.
Figure 3. Effect of oxygen on ONSCN decomposition, λ = 460 nm. Both experiments started with 1 mM NaNO2, 1 M HClO4, and 0.1 M NaSCN.
Scheme 1. RegenerationofONSCNviaOxidationofNObyO2
effect on the overall kinetics and was neglected in the kinetic
analysis.
5. Kinetics: Initial Rates and Effect of Thiocyanate Ion
Concentration. ONSCN was generated in situ via the reaction
of HClO4, SCNꢀ, and NO2ꢀ ions. To assist in the develop-
ment of the mechanism, we studied the initial rates of ONSCN
decomposition as a function of the concentrations of nitrous
acid and thiocyanate ions. A number of issues complicated this
part of our investigation, limiting the quality and usefulness of
the initial rates measurements to identifying the early decom-
position mechanism and the inhibitory effect of NO on the
reaction (described in Section 6). For this reason, the mea-
surements of initial rates did not serve to fit the kinetic
was not reproducible and was not removed by degassing for
longer time periods, arises because of diffusion of a small amount
of O2 through the fluorinated ethylene propylene (FEP) tubing
in the stopped flow device. This phenomenon appears to have no
7444
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
Figure 4. Effect of nitric oxide on ONSCN decomposition, λ = 460 nm. Both experiments contain 3 mM NaNO2, 0.2 M HClO4, and 0.1 M NaSCN.
parameters of the model. The first problem involved the
absorbance increasing slightly over the first 200 ms of reac-
tion, as opposed to an expected decrease. This increase was
not reproducible, suggesting inefficient mixing in the hand-
driven stopped flow device. The second problem comprised
the rate remaining constant only for a very short period of time
(<1 s) before commencing to decrease rapidly, leading to a
small change in absorbance. The resulting initial rate measure-
ments displayed a significant uncertainty, owing to the large
noise in the measurements relative to the change in absor-
bance, and to the irreproducible nature of the early increase in
absorbance.
In the first set of experiments (see Supporting Information),
the nitrous acid concentration was varied from 1 to 4 mM in
increments of 1 mM, with concentrations of 1 and 0.1 M
HClO4 and NaSCN, respectively. Under these conditions,
with a large excess of SCNꢀ and H+, the concentration of
ONSCN becomes directly proportional to the concentration
of nitrous acid. Using higher concentrations of nitrous acid
engendered interference from the decomposition of nitrous
acid, which is initially very rapid. Plotting the log of the initial
rate versus the log of the HNO2 concentration indicated a
reaction order with respect to nitrous acid of 1.07 ( 0.09. In
the second series of experiments, the concentration of SCNꢀ
was varied from 0.1 to 0.25 M in increments of 0.05 M, with
constant concentrations of HNO2 and HClO4 of 2 mM and 1
M, respectively. Under these conditions, the fraction of initial
nitrous acid converted into ONSCN ranges from 0.77 to 0.9.
The order with respect to SCNꢀ amounted to 1.2, however,
the slight elevation in ONSCN concentration induced a
portion of the rate increase. The initial rate was corrected
for the effect of the increased ONSCN concentration, assum-
ing that the initial rate was first-order in ONSCN, resulting in
an order in SCNꢀ of 1.06 ( 0.16. Thus, a process first-order
in nitrous acid and first-order in SCNꢀ dominates the initial
rate of decomposition of ONSCN, as described by eq 13,
assuming that ONSCN accounts for the term first-order in
nitrous acid, where k = 0.28 Mꢀ1 sꢀ1 gives the best fit to the
measurements.
d½HNO2ꢁTotal
ꢀ
¼ k½ONSCNꢁ½SCNꢀꢁ
ð13Þ
dt
6. Kinetics: Effect of Nitric Oxide. The rate of ONSCN
decomposition decreases rapidly with time, despite only rela-
tively small changes in ONSCN concentration. This decrease
resulted from the inhibition of the reaction by one of the
reaction products, either (SCN)2 or NO. It was impossible to
test directly for inhibition of ONSCN decomposition by
(SCN)2, owing to the instability of (SCN)2 in aqueous solution.
However, we readily generated stable NO solutions by passing
NO gas through the solution of interest and ensuring that
oxygen was strictly excluded during generation and transfer of
the solution. Our approach involved examining the effect of NO
on the decomposition of ONSCN, by first saturating the
reactant solution containing SCNꢀ and NO2ꢀ with NO before
reacting it with HClO4 in the stopped flow apparatus. Figure 4
illustrates the results of a typical experiment, confirming that
the solution initially saturated with NO exhibits significantly
slower decomposition of ONSCN. Whereas the solution with-
out NO exhibited a rapid decrease in the rate over time, the
solution with NO decomposed at a much steadier and con-
siderably slower rate. This result appears qualitatively very
similar to decomposition of nitrosothiourea,35,36 which is also
inhibited by NO.
The fact that NO suppresses the initially rapid decomposition
of ONSCN, yet the slope decreases at a steady rate suggests a
combination of reversible and irreversible reaction steps in the
mechanism. The presence of NO suppresses the reversible step,
responsible for the initial rapid decrease in ONSCN concentra-
tion, while similar slopes of both curves at 300 s indicate other
pathways, or subsequent reaction steps that remain unaffected by
NO. Figure 5 illustrates plots of 1/[HNO2]Total versus t, with
their linearity after the first 350 s revealing a second-order
7445
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
Figure 5. Second-order plots of ONSCN decomposition. All experiments initially contain 0.01 M HNO2 and 0.05 M NaSCN. HClO4 concentrations
are 0.4, 0.3, and 0.2 M for experiments 1, 2, and 3, respectively.
Table 1. Reaction Mechanism for ONSCN Decomposition
reaction
equation number
kf
kb
6400 sꢀ1
530 sꢀ1
source
ref 28
HNO2 + H+ + NO2ꢀ a N2O3 + H2O
HNO2 + H+ + SCNꢀ a ONSCN + H2O
2ONSCN f 2NO + (SCN)2
1
32000 Mꢀ2sꢀ1
11900 Mꢀ2sꢀ1
0.57 Mꢀ1sꢀ1
3
ref 16
5
N/A
this study
this study, ref 37
this study, ref 37
this study
ref 23
SCNꢀ + ONSCN a NO + (SCN)2
14
15
18
20
21
22
23
24
25
26
27
28
29
30
31
2.3 Mꢀ1sꢀ1
4.3 ꢂ 109 Mꢀ1sꢀ1
N/A
ꢀ
(SCN)2ꢀ + ONSCN f NO + (SCN)2 + SCNꢀ
1.9 ꢂ 107 Mꢀ1sꢀ1
12000 Mꢀ2sꢀ1
19.8 sꢀ1
+ HNO + H+ f NO + (SCN) + H2O
ꢀ
(SCN)2
N/A
2
2
(SCN)2 + H2O a HOSCN + SCNꢀ + H+
ONSCN + HOSCN a ONOSCN + H+ + SCNꢀ
ONOSCN a NO + OSCN
5140 Mꢀ2sꢀ1
980 Mꢀ2sꢀ1
4.3 ꢂ 109 Mꢀ1sꢀ1
N/A
260 Mꢀ1sꢀ1
this study
this study
this study
ref 29a
1.3 ꢂ 106 Mꢀ1sꢀ1
1.7 ꢂ 107 Mꢀ1sꢀ1
1600 Mꢀ1sꢀ1
O f HOSCN + HO2SCN
2OSCN + H2
2HOSCN f HO2SCN + SCNꢀ + H+
2HO2SCN fHOSCN + HO3SCN
HO3SCN + H2O f SO42ꢀ + HCN + 2H+
2(SCN)2ꢀ f (SCN)2 + 2SCNꢀ
NO2ꢀ + H+ a HNO2
N/A
2 ꢂ 108 Mꢀ1sꢀ1
5 ꢂ 108 Mꢀ1sꢀ1
1.3 ꢂ 109 Mꢀ1sꢀ1
N/A
ref 33
N/A
ref 33
N/A
ref 39
1010
M
ꢀ1sꢀ1
6.76 ꢂ 106 sꢀ1
1.0 ꢂ 108 Mꢀ1sꢀ1
5 ꢂ 10ꢀ3 Mꢀ2sꢀ1
N/A
ref 28
refs 40 and 30b
N2O3 a NO + NO2
3941 sꢀ1
2NO2 + H2O f HNO2 + NO3ꢀ + H+
(SCN)2ꢀ + NO2ꢀ f 2SCNꢀ + NO2
8.4 ꢂ 107 Mꢀ1sꢀ1
ref 30
2.2 ꢂ 106 Mꢀ1sꢀ1
ref 41
2
a The value of K20 from ref 23 was used in conjunction with the value of K20 k24 from ref 29 to determine k24. b The value of k28 was determined from ref
40 in conjunction with the equilibrium constant for N2O3 from ref 28 kꢀ28 was taken as the average of values reported in refs 40 and 30.
reaction in total nitrous acid concentration, largely independent
of NO concentration.
the experiments covering a reasonably wide range of HNO2,
SCNꢀ, and H+ concentrations including some with solutions
initially saturated with nitric oxide. Thus, [HNO2]Total
ranged from 2 to 10 mM, [SCNꢀ] from 0.05 to 0.5 M and
[H+] from 0.02 to 0.5 M. This approach allowed the inclusion
of the decomposition of HNO2 in the model, by incorpo-
rating appropriate equations and rate constants from the
literature (Table 1, reactions 28ꢀ30). Table S3 in the Sup-
porting Information provides the initial conditions of each
experiment.
Reaction Mechanism: Global Kinetic Fit. The initial rates
study has identified the early decomposition of ONSCN to be
first order in both ONSCN and SCNꢀ and has demonstrated the
reaction to be inhibited by nitric oxide. As the starting rate
remained valid only for the first second or so of reaction, we
excluded the results of initial rates experiments when deriving
the kinetic parameters. Potential models were fitted to a set of
20 experiments simultaneously, each executed for 300 s, with
7446
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
Figure 6. Comparison of experimental data and kinetic model (reactions 14ꢀ16) for selected experiments with high thiocyanate ion concentrations.
Data points correspond to experimental measurements and solid curves correspond to the best fit of Pathways 1 and 2 to the data set of 10 experiments
with [SCNꢀ] g 0.2 M. Conditions are 4: [SCNꢀ] = 0.3 M, [H+] = 0.1 M, 5: [SCNꢀ] = 0.3 M, [H+] = 0.3 M, 6: [SCNꢀ] = 0.3 M, [H+] = 0.5 M, 7:
[SCNꢀ] = 0.3 M, [H+] = 0.3 M, 8: [SCNꢀ] = 0.3 M, [H+] = 0.5 M, 9: [SCNꢀ] = 0.5 M, [H+] = 0.5 M.
Kinetics at High SCNꢀ Concentration. At thiocyanate concentra-
tions of 200 mM or higher, the kinetics of ONSCN decomposition
comprises two parallel reaction pathways. Pathway 1, denoted
as the thiocyanate-dependent pathway, involves reaction
between nitrosyl thiocyanate and thiocyanate ions producing
nitric oxide and (SCN)2ꢀ radicals (eq 14), followed by rate-
limiting consumption of (SCN)2ꢀ forming NO and thiocya-
nogen, (SCN)2 (eq 15). Pathway 2 consists of a second-order
reaction in nitrosyl thiocyanate, which produces two mol-
ecules of NO and one of thiocyanogen.
and k15 provided that the product K14k15 remained constant. A value
of K14k15 of 1.15 ꢂ 10ꢀ2 Mꢀ1 sꢀ1 in conjunction with k16
=
0.5 Mꢀ1 sꢀ1 afforded a good match to the acquired measurements.
Figure 6 shows the best fit of reactions 5, 14, and 15 to the results of
selected experiments, characterized by initial concentrations of
SCNꢀ in excess of 200 mM. The Supporting Information provides
additional graphs for other high [SCNꢀ] experiments.
Kinetics at Low SCNꢀ Concentrations. At thiocyanate con-
centrations below 200 mM, the decomposition of ONSCN
displayed faster kinetics than would be expected from Pathways 1
and 2, indicating the presence of an additional reaction pathway
operating under these conditions. A variety of different reaction
schemes were considered, with three potential mechanisms for
ONSCN decomposition found to provide similarly good fits to the
data. All three mechanisms contain Pathways 1 and 2 in conjunc-
tion with an additional reaction channel. These additional pathways
include (i) the homolysis of ONSCN followed by rapid reaction of
Pathway 1.
ꢀ
ONSCN þ SCNꢀ h ðSCNÞ2 þ NO
ð14Þ
ð15Þ
ꢀ
ꢀ
ðSCNÞ2 þ ONSCN f NO þ ðSCNÞ2
ꢀ
SCN radicals to form (SCN)2ꢀ, (ii) the nitrosation of (SCN)2
by NO+/H2NO2+ formingNOand(SCN)2, or (iii) the nitrosation
of HOSCN (formed from (SCN)2 hydrolysis) to produce
ONOSCN, and subsequent decomposition of this species. These
pathways are outlined below.
Pathway 2.
2ONSCN f 2NO þ ðSCNÞ2
ð5Þ
Provided that reaction 14 governs the early rate, this mecha-
nism is consistent with our initial rates study, which high-
lighted a reaction first order in both ONSCN and SCNꢀ.
Reaction 14 might be expected to be rate limiting initially,
when the concentration of NO remains low. Czapski et al.37
studied the reverse of reaction 14, the reaction between NO
and (SCN)2ꢀ, by monitoring the decay of (SCN)2ꢀ produced
via the pulse radiolysis of SCNꢀ solutions containing NO.
They reported k-14 as 4.3 ꢂ 109 Mꢀ1 sꢀ1, and we have adopted
this value in our simulations. We fitted the kinetic parameters
of reactions 5, 14, and 15 to the measurements at high
[SCNꢀ], in conjunction with relevant equations for formation
of ONSCN, and for the protonation and decomposition of
Pathway 3(i).
ONSCN h NO þ SCN
SCN þ SCNꢀ f ðSCNÞ2
ð16Þ
ð17Þ
ꢀ
Pathway 3(ii).
ꢀ
ðSCNÞ2 þ HNO2 þ Hþ f NO þ ðSCNÞ2 þ H2O
ð18Þ
Assuming that the rate limiting step in Pathways 3(i) and 3(ii)
is the consumption of (SCN)2ꢀ, the following simplified rate law
can be derived for ONSCN decomposition, where ka = k15k16/
k-14 for pathway 3(i) and k18K14/K3 for pathway 3(ii) (derivation
nitrous acid as outlined in Table 1. For values of 105 < k15
108 Mꢀ1 sꢀ1, the simulations were insensitive to changes in k14
<
7447
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
Figure 7. Comparison of experimental and modeled decomposition of ONSCN for 0.05 mM NaSCN, 0.01 M NaNO2 and increasing initial acid
concentrations, 0.05, 0.1, 0.2, 0.3, and 0.4 M HClO4 for experiments 10, 11, 3, 2, and 1 respectively. Experiment numbers correspond to those tabulated
in the Supporting Information.
provided in the Supporting Information):
6.5 ꢂ 105 Mꢀ2
s
ꢀ1, which is significantly larger than that
observed for other substrates. As a result, Pathway 3(ii), although
providing a good fit to the data, is unlikely to represent the
correct mechanism because the required rate constant for reac-
tion 18 is not feasible.
2
d½HNO2ꢁTotal
dt
ka½ONSCNꢁ
½NOꢁ
ꢀ
¼
2
k15K14½SCNꢀꢁ½ONSCNꢁ
Pathway 3(iii) involves formation of a previously unreported
species, ONOSCN resulting from the nitrosation of HOSCN.
HOSCN forms as a result of the hydrolysis of (SCN)2 (reaction
20),23 produced from Pathways 1ꢀ2. HOSCN can then either
disproportionate via a second-order reaction which results
eventually in sulfate and cyanide (reactions 24ꢀ26 in Table 1),29
or it can react with ONSCN, forming ONOSCN (reaction 21).
The decomposition of ONOSCN is proposed to form NO and
OSCN radicals via reaction 22. Two OSCN molecules then
combine to form a dimer, which hydrolyzes forming HOSCN
and HO2SCN. This reaction scheme, combined with Pathways 1
and 2 operating in parallel, provides an excellent fit to the data
over the complete range of reactant concentrations used in this
study, as outlined in Figures 7ꢀ9. Table 1 summarizes the
complete model and associated rate parameters.
þ
½NOꢁ
2
þ k5½ONSCNꢁ
ð19Þ
SCN radicals are known to react at or near the diffusion-
controlled limit with SCNꢀ ions forming (SCN)2 radicals,38
ꢀ
with an equilibrium constant of 2 ꢂ 105
M
ꢀ1. Thus, in
accordance with Hess’ Law, the equilibrium constants for reac-
tions 14 and 16 must satisfy the relationship K14/K16 = K17 = 2 ꢂ
105 Mꢀ1. The best fit of Pathway 3(i) (occurring in parallel with
Pathways 1 and 2) to the data resulted in K14/K16 = 10, a factor of
104 lower than required thermodynamically. Therefore, although
this reaction model provides an excellent fit to the data, the
required equilibrium parameters are not thermodynamically
feasible and alternate mechanisms must be considered. As noted
above, Pathway 3(ii) simplifies to the same rate law as Pathway
3(i), with the distinguishing feature of that mechanism being an
Pathway 3-(iii).
ðSCNÞ2 þ H2O a HOSCN þ SCNꢀ þ Hþ
ð20Þ
ꢀ
acid-catalyzed reaction between (SCN)2 (generated via reac-
ONSCN þ HOSCN a ONOSCN þ SCNꢀ þ Hþ
tion 14) and nitrous acid. Such a reaction would likely proceed
+
via a H2NO2 intermediate, making it possible to estimate an
ð21Þ
ð22Þ
ð23Þ
upper bound on this rate constant based on the observed rate of
other acid-catalyzed nitrosation reactions involving negatively
charged substrates. The rate constants of such reactions, includ-
ing the nitrosation of SCNꢀ appear to approach a limiting value
of 12 000 Mꢀ2 sꢀ1 with increasing substrate reactivity.16 Hence,
the rate of nitrosation of (SCN)2ꢀ is unlikely to exceed this value.
The fact that similar rate constants are observed for a range of
reactive substrates suggests that these reactions are encounter
ONOSCN a NO þ OSCN
2OSCN þ H2O f HOSCN þ HO2SCN
’ DISCUSSION
controlled, with H2NO2 in equilibrium with HNO2 and H+.
+
In contrast, the minimum value of k18 required to fit the data
Pathway 1 - Reactions 14 and 15. Pathway 1 involves a
reaction between nitrosyl thiocyanate and thiocyanate ions
for ONSCN decomposition at low SCNꢀ concentrations is
7448
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
Figure 8. Comparison of experimental and modeled decomposition of ONSCN at low NaNO2 concentrations. Concentrations are 9: [SCNꢀ] = 0.5,
[H+] = 0.5, 12: [SCNꢀ] = 0.1 M, [H+] = 0.2, [NO] = 0.62 mM, 13: [SCNꢀ] = 0.1 M, [H+] = 0.2, 14: [SCNꢀ] = 0.2 M, [H+] = 0.2, [NO] = 0.62 mM, 15:
[SCNꢀ] = 0.2 M, [H+] = 0.2 M.
Figure 9. Comparison of experimental and modeled decomposition of ONSCN for experiments 16: [H+] = 0.02 M,[SCNꢀ] = 0.2 M, and 17ꢀ19:
[H+] = 0.1 M, [SCNꢀ] = 0.1, 0.15, 0.2 M respectively.
generating (SCN)2ꢀ radicals. This reaction is analogous to those
reported for decomposition of other nitroso-species, including
nitroso-thioureas36 and nitrosyl iodide,42 which feature a reaction
between the nitroso compound and the relevant nucleophile.
The fact that similar reactions operate for other nitroso species
supports the proposed mechanism for ONSCN decomposition,
and may indicate a general mechanism applicable for this class of
compounds. With regard to the elementary mechanism of this
reaction, our measurements and the results of studies published
in the literature indicate that the reaction between ONSCN and
SCNꢀ likely proceeds through an ON(SCN)2ꢀ intermediate. A
species of this stoichiometry formed as a result of pulse radiolysis
of nitric oxide saturated thiocyanate solutions37 (i.e., the reverse
of reaction 14). Doherty at al. propoꢀsed a species of the same
stoichiometry, denoted ON(SCN)2 to explain the visible
spectrum of nitrosyl thiocyanate solutions containing very high
thiocyanate ion concentrations.20 These researchers claimed that
ꢀ
this species differs from the species denoted NO(SCN)2
observed in pulse radiolysis experiments, and that Czapski
et al.37 had observed a different form of ONSCN. This distinc-
tion results from the differences in the UV spectra of ONSCN
and NOSCN, and the rapid rate of decay of NO(SCN)2ꢀ, which,
according to Doherty et al., would cause the species to “have
completely decomposed long before our first measurements,
7449
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
even with stopped flow”. The latter point seems somewhat
tenuous considering that Doherty et al. measured an equilibrium
concentration of ON(ꢀSCN)2ꢀ, thus despite having a rapid rate
of decay, ON(SCN)2 would still be observable provided the
rate of its formation exhibited a comparable magnitude to that of
concentration of which is low (i.e., due to the low value
of K14) and as such the rate of this reaction remains insignifi-
cant compared with the rate of reaction 15.
Pathway 2 - Reaction 5. This pathway comprises a second
order reaction in ONSCN, directly generating NO and (SCN)2.
Evidence supporting this reaction includes the tendency of
ONSCN decomposition kinetics to approach a second order
reaction in the later stages (conditions whereby the accumula-
tion of NO in the solution suppresses reaction 14) and the fact
that inclusion of this reaction in kinetic simulations leads to a
substantially improved fit to the data. Other related systems,
its consumption. The small equilibrium constant (∼0.036 Mꢀ1
)
for formation of the ON(SCN)2ꢀ adduct would mean that under
the conditions of Czapski et al.’s experiments with 0.1 M SCNꢀ,
the equilibrium concentration of ON(SCN)2 would be very
ꢀ
low and its absorbance indistinguishable from ONSCN con-
sidering their similar UV spectra. As such, one cannot conclude
whether the species observed in these studies are indeed dif-
including the decomposition of nitroso-thiourea35,36 and the
ꢀ
42,45
ferent. Provided that ON(SCN)2 exists at equilibrium witꢀh
oxidation of iodide ions by HNO2
(which likely proceeds
ONSCN and SCNꢀ, the reaction proceeding via ON(SCN)2
remains kinetically indistinguishable from that producing NO
and (SCN)2ꢀ from ONSCN and SCNꢀ in a single step. Doherty
et al.20 showed that ON(SCN)2ꢀ formation reached equilibrium
within the dead time of their stopped flow, thus, we have defined
this rate constant in terms of ONSCN and SCNꢀ.
through an ONI intermediate) contain a term in the initial rate
that is second order in nitroso species. Hence, it stands to reason
that a similar mechanism operates for ONSCN.
Pathway 3 - (iii) Formation and decomposition of ONOSCN ꢀ
Reactions 21-23. Pathways 1 and 2 proceed too slowly to
account for the observed rate of decomposition at thiocyanate
concentrations below 0.2 M, indicating that an additional reac-
tion pathway must operate under these conditions. We propose
the formation and decomposition of a new species, ONOSCN, to
account for the kinetics of ONSCN decomposition under
conditions of relatively low thiocyanate ion concentration. In
the proposed mechanism, ONOSCN arises as a consequence of
the reaction of ONSCN with HOSCN, and decomposes via
homolysis forming nitric oxide and OSCN. The magnitude of
KONOSCN indicates that only a small fraction of HOSCN converts
into ONOSCN. The combination of OSCN and NO was
assumed to occur rapidly with a rate constant of 4.3 ꢂ 109
The simulations were insensitive to changes in k14 and k15
for 105 < k15 < 108 Mꢀ1sꢀ1, provided that the magnitude of
K14k15 remained constant. The constancy of this quantity can
be readily explained by the fact that the rate of reaction 14 is
ꢀ
significantly faster than reaction 15. As such, NO and (SCN)2
exist in equilibrium with ONSCN and SCNꢀ, and the rate limit-
ing reaction step is the consumption of (SCN)2ꢀ via reaction 15.
Therefore, under conditions where reaction 14 is at equilibrium,
it is not possible to isolate the parameters k14 and k15, only K14k15
can be determined.
A key step in the proposed reaction mechanism is reaction 15
between (SCN)2ꢀ and ONSCN, for which the rate constant, k15,
could range from 2 ꢂ 105 to 5 ꢂ 108 Mꢀ1sꢀ1 depending on the
magnitude of K14. Thisꢀreaction could be interpreted as the
S-nitrosation of (SCN)2 by ONSCN, wherein NO+ transfers
from ONSCN to one of the sulfur atoms of (SCN)2ꢀ, presum-
ably forming an unstable intermediate species ON(SCN)2 which
rapidly decomposes to NO and (SCN)2. Alternatively, thꢀe
M
ꢀ1sꢀ1, which falls in the range of rate constants reported in the
literature for reactions between NO and a variety of radicals,
including (SCN)2ꢀ, (Br)2ꢀ, CO3ꢀ 37 organic peroxyl radicals46
and thiyl radicals47 that all occur with rate in excess of 109
ꢀ1sꢀ1. Formation of OSCN has previously been proposed to
,
M
account for the complex kinetics associated with oxidation of
SCNꢀ by ClO2.33 That study assumed that OSCN undergoes
dimerization and hydrolysis with a rate constant of 1 ꢂ 107
reaction could be interpreted as a substitution of the (SCN)2
radical for NO in ONSCN, releasing NO and forming (SCN)3
,
M
ꢀ1sꢀ1 (i.e., k23), which coincides reasonably well with the value
ꢀ
which persists in equilibrium with (SCN)2 and SCNꢀ.34 It is of
interest to compare the magnitude of k15 suggested in this study
to the rate constants of nitrosation reactions involving ONSCN
to determine if the required values of k15 are plausible. In the
case of S-nitrosation, the high reactivity of many substrates makes
it impossible to determine the rate constant, as is the case for the
reaction between ONSCN and a range of heterocyclic thiones,12
while the reaction of ONSCN with mercapto-carboxylic acids
occurs with a rate constant of ∼1 ꢂ 104 Mꢀ1sꢀ1. In the case of
N-nitrosation, the rate constant for the reaction between
ONSCN and aniline and several primary amino acids43 is
∼1 ꢂ 108 Mꢀ1sꢀ1, while a range of aniline derivatives react with
rate constants between 1 ꢂ 105 and 5 ꢂ 108 Mꢀ1sꢀ1. Thus, the
rate constants required for reaction 15 in the proposed mechan-
ism lie within the range of values typically encountered for
nitrosation reactions effected by ONSCN. In the case of
of 1.7 ꢂ 107 Mꢀ1sꢀ1 determined in the present study. Experi-
mentation with different values of k23 in the simulations showed
that the previously reported value could be accommodated by
making appropriate changes to K21 and k22 with no reduction in
the goodness of fit. Numerous intermediates contribute to the
hydrolysis of (SCN)2 and it is plausible that they too participate
in reactions with nitrous acid or nitrosyl thiocyanate. However,
owing to the large rate constants for their consumption,33 the
concentration of such intermediates (e.g., HO2SCN, HO3SCN)
is predicted to be low, resulting in minimal contributions from
these reactions.
Formation of N2O. Small amounts of N2O arise in the product
gases at levels of less than 3%, as confirmed both by FTIR and
MIMS. Because of its low abundance and subsequently insignif-
icant impact on the overall decomposition kinetics, we did not
attempt to model the formation of this species. A potential
pathway for the formation of N2O involves the decomposition of
nitrosyl cyanide, ONCN, which reportedly hydrolyzes to pro-
duce N2O and CO2.48 Cyanide ions, formed via the hydrolysis of
(SCN)2 (generated from the decomposition of ONSCN) could
react with nitrous acid to produce ONCN, in the same way as
thiocyanate ions react with nitrous acid to form nitrosyl thiocya-
nate. An alternative pathway may consist of ONSCN hydrolyzing
ꢀ
the substitution of the (SCN)2 radical for NO in ONSCN,
the analogous reactions between thiyl radicals and thionitrites
are reportedly barrierless,44 suggesting a reaction rate close to
the diffusion controlled limit. Reactions 18 and 27, which offer
alternate pathways for the consumption of (SCN)2ꢀ, do not have
a significant impact on the kinetics. Although the rate constant
for reaction 27 is large, it is second order in (SCN)2ꢀ, the
7450
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
to form ONH, leading to formation of N2O (via combination
of ONH49).
(3) da Silva, G.; Dlugogorski, B. Z.; Kennedy, E. M. AIChE J. 2006,
52, 1558–1565.
(4) Nguyen, D. A.; Iwaniw, M. A.; Fogler, H. S. Chem. Eng. Sci. 2003,
58, 4351–4362.
(5) Stamler, J. S.; Singel, D. J.; Loscalzo, J. Science 1992, 258,
1898–1902.
(6) Williams, D. L. H. Acc. Chem. Res. 1999, 32, 869–876.
(7) Choi, J.; Valentine, R. L. Environ. Sci. Technol. 2003, 37, 4871–4876.
(8) Walse, S. S.; Mitch, W. A. Environ. Sci. Technol. 2008, 42,
1032–1037.
ONCN þ 2H2O h ONH þ CO2 þ NH3
ONSCN þ H2O h ONH þ HOSCN
2ONH f N2O þ H2O
ð32Þ
ð33Þ
ð34Þ
(9) Wu, H.; Loeppky, R. N.; Glaser, R. J. Org. Chem. 2005, 70,
6790–6801.
’ CONCLUSIONS
(10) Ridd, J. H. Q. Rev., Chem. Soc. 1961, 15, 418–441.
(11) Meyer, T. A.; Williams, D. L. H. J. Chem. Soc., Perkin Trans. 2
1988, 517–521.
(12) Amado, S.; Blakelock, L.; Holmes, A. J.; Williams, D. L. H.
J. Chem. Soc., Perkin Trans. 2 2001, 441–447.
(13) Hughes, E. D.; Ridd, J. H. J. Chem. Soc. 1958, 82–88.
(14) Francisco, V.; Garcia-Rio, L.; Antonio Moreira, J.; Stedman, G.
New J. Chem. 2008, 32, 2292–2298.
(15) Garley, M. S.; Stedman, G. J. Chem. Res. 1996, 9.
(16) Williams, D. L. H. Nitrosation Reactions and the Chemistry of
Nitric Oxide; Elsevier B. V.: Amsterdam, 2004.
(17) Stedman, G.; Whincup, P. A. E. J. Chem. Soc. 1963, 5796–5799.
(18) Bourne, N. K.; Field, J. E. Proc. R. Soc. London, Ser. A 1991, 435,
423–435.
(19) Lecher, H.; Graf, F. Ber. Dtsch. Chem. Ges. B 1926, 59B, 2601–2.
(20) Doherty, A. M.; Garley, M. S.; Haine, N.; Stedman, G. J. Chem.
Soc., Dalton Trans. 1997, 2163–2166.
The mechanism of the decomposition of nitrosyl thiocyanate
involves a complex system of reactions, with the rate suppressed
by the accumulation of the product, nitric oxide, and catalyzed by
thiocyanate ions. We examined potential reaction mechanisms
by fitting the kinetic parameters to the experimental measure-
ments, and subjected these kinetic parameters to thermodynamic
and kinetic scrutiny, to identify three parallel pathways for the
decomposition of ONSCN. The first two pathways, which
describe the kinetics at high thiocyanate ion concentrations
comprise a reaction that is second order in ONSCN to generate
NO and (SCN)2 and a reaction between ONSCN and SCNꢀ to
form NO and (SCN)2ꢀ, with the latter pathway being reversible.
The rate limiting step in the mechanism corresponds to the
ꢀ
consumption of (SCN)2 radicals by ONSCN, which could
occur via either radical substitution or S-nitrosation. The third
reaction pathway, which becomes significant at low thiocyanate
concentrations, involves the formation of a previously unre-
ported species, ONOSCN, via reaction between ONSCN and
HOSCN, the latter being an intermediate in the hydrolysis of
(SCN)2. ONOSCN contributes to NO formation via homolysis
of the OꢀNO bond and subsequent dimerization and hydrolysis
of OSCN. The proposed kinetic mechanism provides an
excellent fit to the experimental measurement, allowing the rate
of NO formation from the decomposition of ONSCN to be
modeled accurately.
(21) Jones, E.; Munkley, C. G.; Phillips, E. D.; Stedman, G. J. Chem.
Soc., Dalton Trans. 1996, 1915–1920.
(22) Bazsa, G.; Epstein, I. R. Int. J. Chem. Kinet. 1985, 17, 601–612.
(23) Nagy, P.; Lemma, K.; Ashby, M. T. Inorg. Chem. 2007, 46,
285–292.
(24) Bunton, C. A.; Dahn, H.; Loewe, L. Nature 1959, 183, 163–165.
(25) Armor, J. N. J. Chem. Eng. Data 1974, 19, 82–84.
(26) Tu, C.; Swenson, E. R.; Silverman, D. N. Free Radical Biol. Med.
2007, 43, 1453–1457.
(27) Kuzmic, P. Anal. Biochem. 1996, 237, 260–273.
(28) da Silva, G. R.; Dlugogorski, B. Z.; Kennedy, E. M. Int. J. Chem.
Kinet. 2007, 39, 645–656.
(29) Barnett, J. J.; McKee, M. L.; Stanbury, D. M. Inorg. Chem. 2004,
43, 5021–5033.
’ ASSOCIATED CONTENT
(30) Schwartz, S. E.; White, W. H. Adv. Environ. Sci. Technol. 1983,
12, 1–116.
(31) Stedman, G.; Whincup, P. A. E. J. Chem. Soc. 1969, 1945–1948.
(32) Figlar, J. N.; Stanbury, D. M. Inorg. Chem. 2000, 39, 5089–5094.
(33) Figlar, J. N.; Stanbury, D. M. J. Phys. Chem. A 1999, 103,
5732–5741.
S
Supporting Information. Kinetic data for ONSCN de-
b
composition, typical UVꢀvis and FTIR spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
(34) Barnett, J. J.; Stanbury, D. M. Inorg. Chem. 2002, 41, 164–166.
(35) Collings, P.; Garley, M. S.; Stedman, G. J. Chem. Soc., Dalton
Trans. 1981, 331–335.
Corresponding Author
*Phone: +61 2 4985-4433. Fax: +61 2 4921-6893. E-mail:
Bogdan.Dlugogorski@newcastle.edu.au.
(36) Garley, M. S.; Stedman, G.; Miller, H. J. Chem. Soc., Dalton
Trans. 1984, 1959–1963.
(37) Czapski, G.; Holcman, J.; Bielski, B. H. J. J. Am. Chem. Soc. 1994,
116, 11465–11469.
’ ACKNOWLEDGMENT
(38) Milosavljevic, B. H.; LaVerne, J. A. J. Phys. Chem. A 2005, 109,
165–168.
We acknowledge the Australian Research Council and Dyno
Nobel Asia Pacific Pty Ltd. for funding the research and Ms.
Wendy Venpin for assistance with membrane inlet mass spectro-
metry experiments. M.S.R. thanks The University of Newcastle
for a Research Higher Degree Scholarship.
(39) Neta, P.; Huie, R. E.; Ross, A. B. J. Phys. Chem. Ref. Data 1988,
17, 1027–1284.
(40) Park, J. Y.; Lee, Y. N. J. Phys. Chem. 1988, 92, 6294–302.
(41) Huie, R. E.; Neta, P. J. Phys. Chem. 1986, 90, 1193–1198.
(42) Dꢀozsa, L.; Szilassy, I.; Beck, M. T. Inorg. Chim. Acta 1976,
17, 147–153.
’ REFERENCES
(1) Hunger, K.; Mischke, P.; Rieper, W.; Raue, R.; Kunde, K.; Engel,
A. Azo Dyes; Wiley-VCH Verlag GmbH & Co. KGaA, 2000.
(2) Rollefson, G. K.; Oldershaw, C. F. J. Am. Chem. Soc. 1932, 54,
977–979.
(43) da Silva, G.; Kennedy, E. M.; Dlugogorski, B. Z. J. Am. Chem.
Soc. 2005, 127, 3664–3665.
(44) Stamler, J. S.; Toone, E. J. Curr. Opin. Chem. Biol. 2002, 6,
779–785.
7451
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452

Inorganic Chemistry
ARTICLE
(45) Ferranti, F.; Indelli, A. Gazz. Chim. Ital. 1980, 110, 273–277.
(46) Goldstein, S.; Lind, J.; Merenyi, G. J. Phys. Chem. A 2004,
108, 1719–1725.
(47) Madej, E.; Folkes, L. K.; Wardman, P.; Czapski, G.; Goldstein,
S. Free Radical Biol. Med. 2008, 44, 2013–2018.
(48) Shirota, F. N.; Goon, D. J. W.; DeMaster, E. G.; Nagasawa,
H. T. Biochem. Pharmacol. 1996, 52, 141–147.
(49) Bazylinski, D. A.; Hollocher, T. C. Inorg. Chem. 1985, 24, 4285–
4288.
7452
dx.doi.org/10.1021/ic102445d |Inorg. Chem. 2011, 50, 7440–7452