Nitric/Nitrous Acid Equilibria in Supercritical Water

Article abstract of DOI:10.1021/jp983745o

UV - vis spectroscopy was utilized to measure the decomposition of aqueous HNO₃ solutions above 300 °C, in some cases with added NaOH, H₂O₂, and/or NaNO₂, to form NO₂, HNO₂, NO, N₂

Full text of DOI:10.1021/jp983745o

1678
J. Phys. Chem. A 1999, 103, 1678-1688
Nitric/Nitrous Acid Equilibria in Supercritical Water
†
†
‡
,†
Jerzy Chlistunoff, Kirk J. Ziegler, Leon Lasdon, and Keith P. Johnston*
Department of Chemical Engineering and Department of Management Science and Information Systems,
The UniVersity of Texas at Austin, Austin, Texas 78712
ReceiVed: September 16, 1998; In Final Form: January 13, 1999
UV-vis spectroscopy was utilized to measure the decomposition of aqueous HNO
in some cases with added NaOH, H , and/or NaNO , to form NO , HNO , NO, N O, and oxygen. Individual
bands corresponding to HNO and NO were deconvoluted from the spectra up to 400 °C. With constrained
nonlinear optimization, areas of these bands were used to determine extinction coefficients of NO and HNO
+ (1/2)O
3
solutions above 300 °C,
O
2 2
2
2
2
2
2
2
2
2
+
-
and equilibrium constants for the reactions: HNO
3
h H + NO
3
; 2HNO
3
h H
2
O + 2NO
; (Na )(NO
2
2
;
+
-
2
NO + H O h HNO + HNO ; 2HNO h H O + 2NO + (1/2)O
2
2
3
2
2
2
2
; 2NO h N
2
O + (1/2)O
2
3
)
+
-
3
h Na + NO .
Introduction
cadmium, lithium, and sodium nitrates up to 450 °C by Raman
spectroscopy, which is very sensitive to contact interactions.
The dissociation of nitric acid under hydrothermal conditions
Reactions involving NO, NO2, H2O, HNO3, HNO2, and O2
in the gas phase play a crucial role in nitric acid production,
23
up to 250 °C was studied by Raman spectroscopy and up to
1
combustion, and waste incineration. These reactions have been
3
19 °C from the heat of dilution of concentrated HNO3
studied extensively in both the gas phase and for a gas phase
in contact with an aqueous phase, typically at temperatures
24
solutions. Marshall and Slusher obtained HNO3 dissociation
constants indirectly from solubilities of magnesium sulfate (up
to 370 °C) and calcium sulfate (up to 350 °C). However,
Marshall and co-workers
25
2
-5
below 100 °C. The amount of NOx in exhaust gases may be
lowered by catalytic reduction with NH3 to form N2, or
conversely, NOx may be used to oxidize ammonia wastes to
26
26-28
found that nitric acid decomposes
reversibly to nitrogen oxides and water at elevated temperatures
6
-8
N2.
23,25,26
even below the critical point of water.
More recently, NOx chemistry has been studied in hydro-
Chemical reaction equilibria may be measured quantitatively
thermal oxidation (HO) (also called supercritical water oxidation
in hydrothermal solution by UV-vis spectroscopy. As demon-
,10
(
SCWO)).9 In HO, an aqueous waste containing organics is
2
9-31
strated by Seward and co-workers,
complex equilibria in
oxidized by oxygen in supercritical water (SCW), i.e., at
temperatures higher than 374 °C (typically around 500 °C and
above 220 bar). Thermodynamics and kinetics favor significantly
less NOx at the relatively low temperatures in HO compared
with incineration where temperatures are typically 1200 °C or
hydrothermal solutions can be elucidated by spectral deconvo-
2
9
29,31
lution techniques and nonlinear regression procedures.
-
Chemical equilibria involving H2CrO4 and HCrO4 were
measured quantitatively up to 420 °C. A series of relatively
3
2
stable organic pH indicators has been developed to measure
1
1
more. Kinetic studies of oxidation reactions of organic
compounds by nitrates suggested that the reactive species in
pH and monitor acid-base equilibria up to 420 °C.3
3-37
These
studies demonstrate that UV-vis spectroscopy is well suited
for obtaining quantitative equilibrium constants in SCW.
12
SCW solutions of nitrates can be NO2. Recently, it was
suggested that nitrates present in certain high level and mixed
nuclear wastes could be used as an oxidizing agent instead of
The objective of this work was to characterize and quantify
chemical reaction equilibria involving nitric acid and its
decomposition products with UV-vis spectroscopy in SCW.
Because of the possibility for severe corrosion, a new apparatus
was developed in which the solutions contacted only titanium,
gold, and sapphire and could be flushed rapidly. The Results
and Discussion begins with a description of changes in spectra
resulting from adding NaOH, H2O2, NaNO2, or mixtures of these
compounds to HNO3 solutions. Here, the goal is to determine
qualitatively which decomposition products are present based
on these changes in spectra. Next, spectral deconvolution and
peak assignment are discussed. From the areas of the decon-
voluted spectra, a large-scale generalized reduced gradient
oxygen in the HO pretreatment of these wastes.1
3-15
It has been
proposed that ammonium nitrate recovered from demilitarized
rocket motors may be utilized as an oxidizing agent in HO.
NOx chemistry is also important in rapid hydrolysis of metal
1
6
1
7,18
nitrates to form submicrometer metal oxide crystals.
Fundamental studies of nitrogen chemistry in SCW are
relatively rare due to corrosion and the challenges from high
pressures and temperatures. Brill and co-workers studied kinetics
and mechanisms for decomposition reactions of a variety of
nitrogen-containing compounds in SCW including thermal
19
decomposition of hydroxylammonium nitrate, urea and guani-
dinium nitrate,20 and ethylenediammonium dinitrate21 by using
3
8
22
(LSGRG2) optimization model was utilized to determine
optimal values of the extinction coefficients and equilibrium
constants in conjunction with the concentrations of the decom-
position products. The density effect on equilibrium constants
is discussed as a function of changes in polarity upon reaction.
The equilibrium constants are also utilized to calculate speciation
FTIR and Raman spectroscopy. Spohn and Brill studied ion
pair formation in concentrated aqueous solutions of zinc,
*
To whom correspondence should be addressed.
Department of Chemical Engineering.
Department of Management Science and Information Systems.
†
‡
1
0.1021/jp983745o CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/02/1999

Nitric/Nitrous Acid Equilibria in Water
J. Phys. Chem. A, Vol. 103, No. 11, 1999 1679
Figure 1. Stopped flow apparatus for UV-vis spectroscopy of
corrosive solutions in SCW.
diagrams. Quantitative knowledge of chemical reaction equi-
libria of nitrogen chemistry is highly relevant to understanding
hydrolysis and oxidation reactions, corrosion, materials science,
and separation processes in hydrothermal solution.
Experimental Section
The experimental apparatus was similar to one applied
Figure 2. Spectra of 0.0928 mol kg- HNO
1
(filled symbols) and
previously.3
2,39
3
Spectra were obtained in a titanium optical cell
3
NaNO (open symbols), pressure 34.5 MPa.
equipped with two sapphire windows with an aperture of 5 mm
and a path length of 1 cm. A modification was made to avoid
pumping highly corrosive HNO3 solutions through the HPLC
pump (Figure 1). The solutions were added to the top of a 11/
(
NaNO2). Hydrogen peroxide (30%, EM Science) concentration
was determined before each experiment by titration with a
4
1
standard KMnO4 solution.
1
6′′ i.d. × 1′′ o.d. 6 foot long stainless steel tube equipped with
With a few exceptions, the feed solutions were thoroughly
deoxygenated by purging with nitrogen. The solutions containing
H2O2 as well as mixtures of HNO3 and NaNO2 solutions were
not deoxygenated, because of H2O2 decomposition upon shaking
and evolution of NO, respectively. In these cases, however, the
solutions were prepared immediately before each experiment
in freshly deoxygenated water. After brief but thorough mixing,
the solution was placed in the stainless steel tube and im-
mediately isolated and pressurized. The estimated concentration
a stainless steel piston and polypropylene o-ring seals. Pure
deionized water was pumped with an HPLC pump into the tube
below the piston to displace the solution through a titanium
preheater (0.03′′ i.d., 2 foot length) into the titanium optical
cell. A bypass line was used to flush the cell with pure water.
The typical time required to exchange the solution in the cell
was around 3 min. After the cell was flushed with water or a
fresh solution, the spectra were recorded. The spectrum for pure
water (baseline) was measured immediately after the temperature
attained its steady preset value. The spectra for the solutions
were measured after additional flushing of the cell with 1 mL
of the solution and reequilibrating the temperature. Typically
four spectra were recorded at different pressures without flushing
the cell between experiments. The solution residence time, which
did not exceed 6 min, was usually sufficiently short to produce
only minor corrosion as well as negligible changes in the spectra.
Carbon dioxide free, approximately 0.5 M stock NaOH
-
5
of oxygen in these feed solutions was around 4 × 10 mol
-
1
kg based on the partial pressure of oxygen in air and the
solubility of O2 in water. Other details of the experimental
3
2
procedures can be found elsewhere.
Results and Discussion
Concentrated NaNO3 and HNO3 Solutions from 20 to 400
°C. The spectrum of nitrate ion in water is characterized by
two principal bands, as shown in Figure 2. The short-wavelength
band (λ ) 200.3 nm at ambient conditions) is very strong (ꢀ )
9600). While the high ꢀ of this band is beneficial, its λ is not,
since a variety of other species also absorb strongly here,
(
analytical grade, EM Science) solution was prepared and
40
standardized using generally accepted procedures. A stock
solution (approximately 0.5 M) of HNO3 (70%, AR, Mallinck-
rodt) was standardized by titration with the above NaOH
standard solution. Approximately 0.25 M NaNO2 solution was
obtained by dissolving a sample of analytical grade NaNO2 (EM
Science) in a known volume of deionized water. Its concentra-
tion was determined by a classical Lunge method,4 i.e., by
titration with a known volume of a standard KMnO4 solution.
This method is accurate to within 0.5-1%. The NaOH and
NaNO2 stock solutions were prepared frequently, to reduce
errors associated with absorption of CO2 or decomposition
-
including other NOx species, OH via charge transfer to solvent,
and the sapphire windows. The long-wavelength band, centered
at around 300 nm, is very weak with an extinction coefficient
on the order of 7, which is too small to be of use.
1
-
1
The effect of temperature on the spectra of 0.0928 mol kg
-1
HNO3 and 0.0928 mol kg NaNO3 solutions is shown in Figure
2. Within limits of experimental error, the spectra of these
solutions at room temperature are identical (Figure 2a), in

1680 J. Phys. Chem. A, Vol. 103, No. 11, 1999
Chlistunoff et al.
TABLE 1: Compositions of the Classes of Solutions Studied.
HNO
mmol kg
3
NaOH
mmol kg
H
mol kg
2
O
2
-1
NaNO
2
mmol kg
-
1
-1
-1
class
I
II
III
IV
V
7.736-38.68 0.0
0.0
0.0
0.0
19.34
1.237-12.37 0.0
23.20
0.0-16.28
15.40
0.0
0.0
7.667
4.617
0.182-0.931 0.0
a
0.0
0.0
0.0
4.52-20.8
5.42
7.03
1
3.73
a
Total nitrogen concentration was 0.0208 mol kg-1
.
accordance with the fact that nitric acid is completely dissoci-
ated. However, an increase in temperature at a constant pressure
of 34.5 MPa leads to changes in HNO3 (open symbols) and
NaNO3 (filled symbols). While changes in NaNO3 spectra up
to 400 °C can be regarded as minor (Figure 2b), temperature
exerts a significant effect on the HNO3 spectra. A temperature
increase from 20 to 200 °C leads to increased overlap of the
3
00 and 200 nm bands. Since nitric acid becomes a weaker
Figure 3. Spectra of 3.868 × 10 mol kg-1 HNO
-2
solutions at two
3
acid at elevated temperatures, these changes may reflect
formation of molecular HNO3.
temperatures and four pressures. Pressure (MPa): 27.6 (1); 31.0 (2);
4.5 (3); 41.3 (4). The spectra are corrected for density of water. Note
3
At temperatures equal to or higher than 300 °C, an additional
broad band centered around 370 nm, appears. This band grows
significantly with temperature. While part of this increase is
reversible and the nitrate band at 300 nm can be restored after
cooling the cell, the measured absorbance does not drop back
to zero at wavelengths longer than those corresponding to the
nitrate band. The reversible changes at temperatures greater than
or equal to 300 °C represent reversible decomposition of nitric
acid, as previously suggested by Marshall and co-workers, based
upon visual observation of HNO3 samples in silica capillary
tubes.27 The residual absorption is still observed even after the
cell is flushed with pure deionized water, due to the presence
of a uniform brown deposit on the windows. The amount of
the deposit increases with the nitric acid residence time in the
cell. The brown deposit is believed to be titanium dioxide, which
is formed when the hot solution containing Ti corrosion products
contacts the cooler windows. Indeed, extensive use of the cell
with these concentrated HNO3 solutions resulted in formation
of pits on the flat Ti surfaces remaining in contact with the
gold seals. The corrosion led to serious leaks and irreproducible
results due to inaccurate correction of the spectra for the
baseline.
similarity of spectrum 1 at 380 °C and spectrum 4 at 400 °C, which
-
3
were obtained at similar densities, i.e., 0.506 and 0.533 g cm
respectively.
,
the oxidation states, to be able to obtain equilibrium constants
from the spectra. The first class consisted of pure nitric acid
solutions (class I). In class II, these solutions were partially
-
neutralized solutions with NaOH to form NO3 , which was
shown above to be much more stable than HNO3. Class III
included H2O2 as an oxidant. Mixtures of HNO3 + NaNO2 were
used in class IV to shift the oxidation states, while NaOH was
added to this mixture in class V.
Typical spectra for a class I experiment are shown in Figure
3. In this class, spectra were time independent for times not
exceeding 20 min. The spectra were reversible with changes in
temperature and pressure, suggesting that decomposition prod-
ucts revert back to the starting material. The large asymmetric
band at 385 nm decreases with an increase in density, especially
at low densities. The band position, shape, and fine structure
are all very similar to that observed for NO2 spectra in the gas
4
2-45
phase.
However, these spectra do not correspond to NO2
only. A distinct set of small peaks, whose relative contribution
to the measured absorbance changes with temperature and
pressure, can be seen on top of the large NO2 band. The relative
contribution from the set of small peaks decreases with
increasing concentration (absorbance) of NO2 resulting from
decreasing solution density. A similar but somewhat weaker
effect was also seen at a constant temperature and pressure when
the NO2 concentration increased due to an increase in the HNO3
feed concentration. The fact that the relative contribution of the
small peaks increases when the NO2 concentration decreases
suggests that these small peaks are not due to the presence of
N2O4, i.e., NO2 dimer. The dimerization reaction occurs to a
The corrosion of the cell and deposition of TiO2 on the
windows was curtailed by lowering the acid concentration to a
-2
-1
level of 2-3 × 10 mol kg , which is also more suitable for
equilibrium constant determinations. However, the changes
occurring to the weak 300 nm band could no longer be
accurately measured in our 1 cm cell. Fortunately, the light
absorption by the HNO3 decomposition products in the dilute
HNO3 solutions was sufficiently high to be measured quanti-
tatively at temperatures exceeding 360 °C. Preliminary experi-
ments revealed, however, that the spectra measured at 360 °C
were not reproducible, because of either a relatively slow
approach to equilibria or TiO2 deposition. At 420 °C, on the
other hand, leaks constituted a serious problem. At temperatures
of 380 and 400 °C, equilibrium was achieved rapidly and leaks
were minor, such that quantitative experiments could be
performed.
3,42,46,47
limited extent at room temperature
in the gaseous phase,
and the degree of dimerization decreases strongly with increas-
4
6,47
ing temperature.
Thus, dimerization is extremely unlikely
at the high temperatures employed in this study. In addition,
N2O4 absorbs light at shorter wavelengths, and the absorption
4
2
Spectra of Nitric Acid and Its Mixtures with Sodium
Hydroxide, Hydrogen Peroxide, and Sodium Nitrite at 380
and 400 °C. The experiments were grouped in five distinct
classes depending on the feed solution composition (Table 1).
The various classes were designed to manipulate the relative
equilibrium concentrations of the nitrogen species, especially
band does not exhibit any fine structure.
The general behavior observed for partially neutralized HNO3
solutions (class II, Table 1) was very similar to that exhibited
by pure nitric acid solutions. Solutions with equal feed
+
concentrations of H at ambient temperature, i.e., pure nitric
o
acid solutions of concentration m and partially neutralized

Nitric/Nitrous Acid Equilibria in Water
J. Phys. Chem. A, Vol. 103, No. 11, 1999 1681
Figure 4. The NO
2
band area integrated over 450-600 nm (bottom
Figure 6. Room-temperature spectra of nitrates, nitrites, and the
-
2
line) and 330-450 nm (top line) plotted against the concentration of
nitric acid in pure nitric acid solutions (filled symbols) or excess of
nitric acid in nitric acid/NaOH mixtures (open symbols): temperature
corresponding acids and assignment of the bands: (1) 2.26 × 10
-3 -2 -3
2 3
mol dm NaNO
; (2) 9.280 × 10 mol dm NaNO
; (3) 2.26 ×
; (4) 2.26 ×
10 mol dm-3 NaNO
-
2
+ 9.280 × 10 mol dm HNO
-2
-3
2
3
-
2
-3
-3
3
80 °C, pressure 27.6 MPa.
10 mol dm NaNO
2
3
+ 0.4176 mol dm HNO .
nm as shown in curves 3 and 4 of Figure 5 (class III, Table 1).
The large NO2 band is not affected significantly by H2O2. At
3
80 °C and all the pressures studied, the NO2 band initially
decreases and then increases slightly as the oxygen content
increases. At 400 °C, the NO2 band decreases monotonically
as the feed H2O2 concentration increases from 0.18 to 0.93 mol
-
1
kg .
The stability of NO2 versus the instability of the species
represented by the set of small peaks in the presence of oxygen
curve 1 and 2 in Figure 5) suggests that the yet unidentified
(
species contains nitrogen in an oxidation state lower than +4.
-
Among known HyNOx (x, y g 0) species, only NO2 and
unstable HNO2 (disproportionates to HNO3 and NO) absorb light
5
0-54
in this range.
Their spectra were measured at ambient
50-54
conditions (Figure 6) and agreed well with literature data.
The positions, intensities, and shapes of the spectra shown in
Figure 6 strongly suggest that the set of small bands on top of
the NO2 band in hydrothermal HNO3 solutions corresponds to
-
the HNO2/NO2 couple.
Figure 5. Spectra of 2.320 × 10 mol kg-1 HNO
-2
with addition of
3
The differences in the influence of oxygen on the HNO2/
hydrogen peroxide: temperature 380 °C, pressure 41.3 MPa. Hydrogen
peroxide concentration (mol kg ): 0.0 (1); 0.262 (2); 0.546 (3); 0.906
4).
-
NO2 and NO2 bands can be explained qualitatively in a simple
-1
manner in terms of the following equilibrium:
(
HNO3 solutions with an excess acid concentration equal to mo,
had almost identical spectra in supercritical water. This result
is shown in Figure 4, where the integrated band areas in two
2NO + H O h 2HNO + (1/2)O
2
(1)
2
2
2
Addition of oxygen shifts this equilibrium to the left and
therefore explains the relative decrease of the HNO2/NO2
+
-
wavelength ranges are plotted versus feed excess H concentra-
tion. The band area, integrated over the wavelength range 450-
content versus that of NO2 in Figure 5. One should note,
however, that this explanation does not include HNO3 as well
as any nitrogen species with an oxidation state lower than +3,
such as NO, which, as suggested by literature data,55 can also
be present in our case. A more quantitative explanation, which
includes these species, is given in the speciation diagrams below.
Figure 7 shows the effect of nitrite on the spectra of HNO3
6
00 nm, is believed to correspond to a single species (NO2),
because the ratios of absorbances measured at a given temper-
ature and pressure for any two solutions were essentially
independent of the wavelength in this range. In contrast, the
area integrated over the wavelength range 330-450 nm includes
the contribution from the substance(s) giving rise to the set of
small peaks. As seen in Figure 4 and the associated spectra,
this contribution, even though relatively small, is similar for
pure nitric acid solutions and partially neutralized HNO3
-1
for a fixed total nitrogen concentration of 0.0208 mol kg (class
IV). An increase in the NaNO2 concentration leads to a
significant decrease in the NO2 band. At the same time, the set
of small bands at 370 nm does not seem to be affected strongly
for a molar ratio of NaNO2 to HNO3 up to 1:1. At higher NaNO2
+
solutions with equal excess H feed concentration.
Addition of an excess oxidant such as H2O2, which decom-
poses to oxygen and water,4
8,49
to a 0.0232 mol kg HNO3
solution removes the set of small peaks centered around 370
-1
to HNO3 ratios, the set of small bands (HNO2/NO2 ) is gradually
-
-
replaced by a single band (NO2 ) at a wavelength similar to

1682 J. Phys. Chem. A, Vol. 103, No. 11, 1999
Chlistunoff et al.
Deconvolution of the Spectra. As shown in the preceding
section, the only distinct bands in the measured spectra are those
-
corresponding to NO2 and HNO2/NO2 . Among other species
which can absorb light in the visible and near-UV ranges are
-
HNO3, NO3 , N2O4, and N2O3. The steep increases in absor-
bance at wavelengths shorter than 280 nm cannot be used for
analytical purposes, since so many nitrogen species absorb in
this range. Above, we have already eliminated the possibility
of N2O4.
The HNO3 and NO3- low-energy bands are very weak
57
-
1
(
Figures 2 and 6). The absorbance of a 0.02 mol kg HNO3
solution at 300 nm resulting from the presence of HNO3 or
-
NO3 would not exceed 0.14 if there was no decomposition
and assuming temperature-independent extinction coefficients.
Because a significant amount of HNO3 decomposes, the
remaining absorbance is too low to be determined accurately,
especially after including effects due to the overlap with higher
energy bands (Figure 2), absorbance from decomposition
products, and light scattering by the TiO2 deposit on the
-
windows. Consequently, we decided to exclude NO3 and HNO3
bands from spectral deconvolution.
58
N2O3 has a weak band (ꢀ ≈ 20) centered around 620 nm
and absorbs strongly in the UV region. The estimated extinction
5
9
coefficient of N2O3 at 280 nm is 290. Since the measured
-
2
Figure 7. Spectra of HNO
3
/NaNO
content of 0.0208 mol kg : temperature 380 °C, pressure 31.0 MPa.
NaNO mole fraction: 0.218 (1); 0.260 (2); 0.385 (3); 0.442 (4); 0.549
5); 0.635 (6); 0.778 (7); 1.0 (8).
2
mixtures with a total nitrogen
absorbance at 280 nm did not exceed 0.4 in a 2 × 10 mol
-
1
-1
kg HNO3 solution, the maximum possible N2O3 concentration
2
-3
-1
could not exceed 1.4 × 10 mol kg resulting in a normalized
absorbance of 0.03 at 620 nm. Similar calculations for HNO3/
NaNO2 mixtures reveal that the absorbance due to N2O3 at 620
nm is also insignificant. In addition, gas-phase data suggest that
N2O3 should dissociate almost completely to NO and NO2 at
(
that measured in pure NaNO2 solution (curve 8). As could be
expected for a salt of a relatively weak acid, sodium nitrite
solution was found to be a rather strong base in supercritical
water, as manifested by significant corrosion of the sapphire
6
,47,60,61
our temperatures.
As was the case for nitrate, we think
3
2
that the N2O3 band at 620 nm, if present, is too small to be
accurately measured and deconvoluted from the spectra.
windows.
The observed changes in the spectra resulting from addition
of NaNO2 to nitric acid (Figure 7) further support our assignment
of the bands. Because the total nitrogen content is fixed, the
amount of HNO3 in the feed mixture decreases. Consequently
products from the decomposition of HNO3, i.e., NO2 and HNO2/
Another species that was considered was the solvent separated
+
-
+
+
ion pair (H )(NO2 ) or, in solutions containing Na , (Na )-
-
(NO2 ). These species will now be shown to be much less stable
than HNO2. The dissociation constant of nitrous acid under our
experimental conditions can be roughly estimated to be on the
-
NO2 decrease. However, the total nitrite concentration (small
-
-8
bands) does not decrease since the loss in HNO2/NO2 is
order of 10 or less at 380 °C and solution densities in the
-
3
62-64
compensated for by the greater amount of nitrite introduced in
the feed. Alternatively, the same effects may be explained in
terms of the net oxidation state of nitrogen in the mixture. The
decrease in the net oxidation state of nitrogen in the feed mixture
leads to the relative increase of the lower oxidation state (HNO2/
range 0.5-0.6 g cm based on low-temperature data
and
comparison with other acids of comparable strength in super-
Dissociation constants for 1:1 alkali metal
salts at densities from 0.5 to 0.6 g cm are typically between
10 and 10 at 380 °C.
constants, comparable concentrations of HNO2 and (Na )(NO2 )
would exist for H to Na equilibrium concentration ratios lower
6
5,66
critical water.
-
3
-
1
-2
66,67
Based on these equilibrium
-
+
-
NO2 ) versus that of the higher oxidation state (NO2).
+
+
For mixtures with a high NaNO2 to HNO3 ratio (spectra 6
and 7 in Figure 7), the concentration of NO2 is negligible. The
-6
+
-8
-1
than 10 , i.e., H concentration on the order of 10 mol kg ,
which is unlikely in acidic media. For similar reasons, the
-
composite HNO2/NO2 band is also much smaller than expected
-
+
-
for pure NO2 or HNO2 solutions with identical total nitrogen
concentration of (H )(NO2 ) may be expected to be insignifi-
+
-
+
content. While part of the nitrogen in these solutions can be in
cant. Therefore, we chose to exclude (H )(NO2 ) and (Na )-
(NO2 ) bands from the spectral deconvolution.
-
-
the form of NO3 , which has a weak absorption band around
3
00 nm (see Figure 6 and Figure 7 curve 6), other nitrogen
Our final procedure was to deconvolute the spectra with only
species may also be present such as NO or N2O which absorb
five Gaussian peaks, four of them representing HNO and one
2
below 280 nm.56
representing NO . Fixed widths and positions were used for the
2
Experiments for mixtures of nitric acid, sodium hydroxide,
and sodium nitrite (class V) were performed to further increase
the ratio of HNO2/NO2- to NO2 over the values in class IV. As
shown by the reaction
HNO bands as determined from the spectra where HNO bands
2
2
were the strongest (see above). A wavenumber range of 3 ×
4
4
-1
10 to 5 × 10 cm (333-500 nm) was used to achieve an
accurate fit for the asymmetric NO peak. The position and
2
width of this peak was allowed to vary. Extensive calculations
revealed that the NO2 peak position peak width varied randomly
+
-
2
H + 2NO3 h 2NO + H O + (1/2)O
(2)
2
2
2
(Table 2), most likely because of the large peak width and the
the addition of a base (NaOH) may be expected to lower the
NO2, as was observed (not shown).
imperfect baseline correction. Spectra obtained from class IV
experiments with a NaNO2 to HNO3 ratio higher than ap-

Nitric/Nitrous Acid Equilibria in Water
J. Phys. Chem. A, Vol. 103, No. 11, 1999 1683
TABLE 2: Deconvolution of the Spectra with Five Gaussian Peaks at 380-400 °C and 27.6-41.3 MPaa
peak area ratiob
peak
width cm
1
-1
class IIc
class IIId
band
species
NO
λ nm
1/λ cm-
class I
class IV
1
2
3
4
5
2
385 ( 7
386.1
371.7
358.4
348.4
1134 ( 48
HNO
HNO
HNO
HNO
2
2
2
2
25900
26900
27900
28700
50
90
1
1
1
1
0.33 ( 0.04
0.64 ( 0.10
0.58 ( 0.20
0.32 ( 0.07
0.65 ( 0.16
0.69 ( 0.23
0.25 ( 0.09
0.99 ( 1.00
2.31 ( 7.21
0.34 ( 0.03
0.70 ( 0.11
0.69 ( 0.14
70
130
a
The errors listed are mean standard deviations. b Peak area relative to band 2. c After removing four data points which deviated strongly from
d
the mean. Significant scatter in these data result from the extremely small nitrous acid bands in the presence of oxygen.
however, could not be determined directly, since their extinction
coefficients are unknown under hydrothermal conditions. The
NO2 extinction coefficients in the gas phase have been
4
3-45,68-71
determined accurately
temperature.
and are strongly dependent upon
4
5,70
The NO2 band fine structure is significantly
reduced in SCW, especially at higher densities (see above). Data
are unavailable for nitrous acid extinction coefficients versus
temperature, most likely due to the low stability of the acid at
low temperatures. However, since measured absorbances are
unique functions of extinction coefficients and all the equilibria
in the system, both the extinction coefficients and the equilib-
rium constants may be optimized numerically to the experi-
2
9,30
mental data.
While details of the numerical method are
7
2
described elsewhere, the basic method and assumptions are
presented here.
The equilibrium between most probable species may be
represented by the following basic set of reactions:
Figure 8. Peak fitting results for a 0.03480 mol kg-1 HNO
3
solution
at 27.6 MPa and 380 °C. The individual bands obtained from the fitting
are drawn with dotted lines. Thick and thin full lines in the upper portion
of the graph represent the measured and fitted spectra, respectively.
+
-
HNO h H + NO
(3)
(4)
(5)
(6)
(7)
3
3
2
HNO h H O + 2NO + (1/2)O
3 2 2 2
proximately 1:1 (curves 6-8 in Figure 7) were excluded, since
these solutions obviously contained nitrites whose bands
interfered with these of HNO2. The parameters for the HNO2
bands were averages of the values obtained from deconvoluting
individual spectra; typical results of deconvolution are quite
accurate as shown in Figure 8. As shown in Table 2, the ratio
of each band to band 2 was relatively constant, indicating only
one species produced the fine structure and that the fits were
quite accurate. An exception is H2O2 class III experiments where
nitrous acid peaks were extremely small (Figure 5) and the errors
were correspondingly large (Table 2).
The rest of this section describes other deconvolution
techniques that were unsuccessful and not adopted. Because the
NO2 band was extremely broad, it was attempted to fit the band
with two Gaussian bands. However, the changes in the positions
and widths of these bands were unacceptably large. As indicated
2NO + H O h HNO + HNO
2 2 3 2
2HNO h H O + 2NO + (1/2)O
2
2
2
2NO ) N O + (1/2)O
2
2
-
+
H O h OH + H
(8)
2
+
-
NaOH h Na + OH
(9)
+
-
3
+
-
-
(
Na )(NO ) h Na + NO
(10)
3
Less important reactions are
+
-
2
+
(Na )(NO ) h Na + NO
(11)
(12)
2
by the spectra measured for NaNO2/HNO3 mixtures (see Figure
2NO h N O
4
2
2
-
7
), the content of NO2 in some of these solutions could
+
-
2
potentially be significant. Therefore, it was also attempted to
fit the spectra with six Gaussian peaks representing a single
HNO h H + NO
(13)
(14)
(15)
2
-
NO2 band, the four strongest HNO2 bands, and a single NO2
2
HNO h N O + H O
2
2
3
2
band (see Figure 6). However, consistent peak positions and
widths were obtained only for the four peaks representing HNO2
in spectra where the HNO2 bands were the strongest. Because
N O h NO + NO
2
3
2
-
the single NO2 band was not different enough from the NO2
For reasons specified immediately below, these equilibria (eqs
11-15) were not considered in our model. As explained above,
reactions 11 and 12 do not occur to a significant extent under
our experimental conditions. As a result of the low dissociation
constant of nitrous acid, reaction 13 is shifted far to the left in
all the acidic solutions. At ambient conditions, reaction 14 occurs
to a significant extent in strongly acidic dehydrating media5
but it is not likely to generate significant quantities of N2O3
band, it was not possible to deconvolute reliable values of NO2-
over a wide pH range. Therefore, spectra were not deconvoluted
in this study in solutions where the NaNO2 to HNO3 ratio in
the feed was greater than 1.22:1.
Determination of the Equilibrium Constants. Among
possible decomposition products of nitric acid, only NO2 and
HNO2 could be observed in our spectra. Their concentrations,
2,53,73

1684 J. Phys. Chem. A, Vol. 103, No. 11, 1999
Chlistunoff et al.
under our experimental conditions because of reaction 15 as
discussed above. The following set of equations defined the
equilibrium constants in our model:
GRG2) optimization routine.38 To properly scale the equations
to aid optimization, a logarithmic transformation was made for
7
2
each equation as described elsewhere. The optimization was
carried out on eight data sets corresponding to the different
temperatures and pressures. Initially, all concentrations, extinc-
tion coefficients, and equilibrium constants were not bounded.
These values were further refined by reducing the bounds of
the variables to achieve more linear behavior in the log of the
equilibrium constants versus density. Typically, the final values
for the equilibrium constants were at least half an order of
magnitude away from a bound.
2
K m
- mH
+
mNO3
-
γ( ) 0
(16)
(17)
(18)
(19)
(20)
a
HNO3
2
2
1/2
K m
- mNO2 mO2 ) 0
4
HNO3
2
K m
- mHNO3
m
) 0
5
NO2
HNO2
2
2
1/2
K m
- mNO mO2 ) 0
Integrated areas of NO2 and HNO2 bands from deconvolution
of the spectra were used in the determination of the equilibrium
constants. The optimization was performed by using eqs 16-
29 as constraints and the equilibrium constants, species con-
centrations, ionic strength, and activity coefficient as variables.
Each of the eight data sets (T,P) included 35-48 experiments
6
HNO2
2
1/2
K m
- m mO2 ) 0
7
NO
N O
2
2
γ( ) 0
(21)
(22)
(23)
w
K - mH
+
mOH
-
7
2
from classes I-IV (see Ziegler et al. ). This resulted in the
simultaneous solution of 497-679 variables by using 490-672
equations as constraints. The objective function was chosen to
be
2
K m
- mNa
+
mOH
-
γ( ) 0
b
NaOH
2
Kd,ipmNaNO3 - mNa
+
mNO3
-
γ( ) 0
N
where Ka is the dissociation constant of nitric acid, Kw is the
ion product of water,74 Kb is the inverse of the sodium hydroxide
2
minφ ) w1
(ANO2,measured - ꢀNO2lmNO2) +
∑
i)1
7
5
association constant as determined by Ho and Palmer, and
N
+
-
Kd,ip is the (Na )(NO3 ) ion pair dissociation constant. The
subscripts defining the remaining equilibrium constants refer
to the reaction numbers in the equations above. It was assumed
that the individual ionic activity coefficients are not sensitive
to specific properties of ions and can be approximated by the
2
w2 (A
- ꢀHNO2lmHNO2) (30)
HNO ,measured
2
∑
i)1
where ꢀNO , and ꢀHNO are the integrated extinction coefficients.
2
2
The HNO2 residuals were weighted so as to be the same order
of magnitude as the NO2 residuals; w1 and w2 were chosen to
be 10 and 1000, respectively. This approach allowed the small
integrated area of HNO2 to affect the final equilibrium values.
Further improvement to the model was achieved by applying
the “jackknife” statistical procedure as described elsewhere.72
Table 3 lists values of the equilibrium constants and molar
absorptivities of NO2 and HNO2 obtained from the optimization.
For a discussion of local optima and estimates on the error
associated with these values see Ziegler et al.
Figure 9 shows dissociation constants for nitric acid and the
Na )(NO3 ) ion pair plotted against the water density. Similarly
to other strong acids, HNO3 becomes a much weaker acid at
high temperatures in supercritical water. Part of this effect is
due to the exothermic character of HNO3 dissociation at
moderate and high densities. The second part is due to the
isobaric decrease of water density and dielectric constant, which
favor the associated species, i.e., HNO3. The nitric acid
dissociation constants at 380 °C agree well with the data of
76
mean ionic activity coefficient, γ(, given by Pitzer:
x
I
2
1.2
ln γ ) -Aæ
+
ln(1 + 1.2
x
I) (24)
(
[
]
1
+ 1.2 I
x
76
where Aæ is the Debye-Huckel parameter and I is the ionic
strength given by
1
2
72
I )
m z
(25)
∑i i i
2
+
-
(
where mi and zi are the molalities and charges of all the ionic
species in solution.
The model also included, a charge balance,
mNa
+
+ mH
+
- mNO3
-
- mOH
-
) 0
(26)
a sodium mass balance,
0
0
Marshall and Slusher obtained from calcium and magnesium
m
+ m
- mNa
+
- mNaOH - m
+
-
) 0 (27)
NaOH
NaNO2
(Na )(NO
)
25,26
3
sulfate solubilities in nitric acid solutions
to 0.45 g cm (see Figure 9a). However, this new spectroscopic
technique offers a means to measure equilibrium constants at
at densities down
-3
a nitrogen mass balance,
-
3
0
0
very low densities, in this case down to 0.24 g cm .
m
+ m
NO3-
- mHNO3 - m - mHNO2 - mNO2 -
HNO3
NaNO2
+
-
The dissociation constants for the (Na )(NO3 ) ion pair are
close to corresponding data for NaCl, KCl, KOH, NaOH.6
The change in Kd,ip with a decrease in density (Figure 9b) reflects
increased interionic interactions in low dielectric permittivity
media.
7,75,77
mNO - 2m - m
+
-
3
) 0 (28)
)
N O
(Na )(NO
2
and a redox balance,
1
mO2 - (m
2
0
0
1
4
Figures 10-13 show the equilibrium constants for reactions
4-7 plotted against the water density together with the gas-
phase values calculated from existing thermodynamic data.61
Reactions 4-7 are nonionic, and density effects on equilibrium
constants are smaller than these observed for ionic reactions.78
Linear extrapolation of the log K versus density plots to zero
- m
2
+ mHNO2) - (m + 3mNO) -
H O
NaNO2
NO2
2
2m
) 0 (29)
N O
2
The equilibrium constants and extinction coefficients were
fitted through a large-scale generalized reduced gradient (LS-

Nitric/Nitrous Acid Equilibria in Water
J. Phys. Chem. A, Vol. 103, No. 11, 1999 1685
and HNO
TABLE 3: Equilibrium Constants for the Reactions 3-7 and 10 and Integrated Extinction Coefficients for NO
from Optimization of the Measured NO
2
2
2
and HNO
2
Absorbancesa
1
05 ꢀNO2b
105 ꢀHNO2b
2
2
-1
-1
1
04 KHNO3
K
4
10 K
6
K
7
10 Kd,ip
(kg mol
cm ) cm
(kg mol
cm ) cm
-
1
(mol kg-1)0.5
2
(mol kg-1)0.5
(mol kg-1)0.5
-1
-1
-1
-1
-1
T °C
P MPa
mol kg
18.2
5.15
10 K
5
mol kg
-
2
2
2
2
3
80
27.6
1.09
0.231
0.168
0.086
24.4
2.47
1.06
2.98
6.30
5.28
8.67
0.05
0.60
2.17
5.45
3.83
7.23
12.5
6.76
41.6
30.9
22.8
12.2
4.65 × 10
4.65
12.3
5.13
6.24
6.40
6.05
6.04
7.06
6.32
6.68
7.08
1.43
1.77
2.39
2.16
7.46
3.74
2.08
1.82
-
31.0
34.5
41.4
2.36 × 10
-
4.20
5.64
0.006
0.041
0.586
5.65
2.56 × 10
-
0.47 × 10
16.6
4
00
27.6
490.91
12.39
0.53
0.003
0.615
0.582
5.15
3
3
4
1.0
4.5
1.4
-
2
0.333
7.14 × 10
a
Class V experiments were not used in deconvolution. b Integrated extinction coefficient with respect to wavenumber (cm-1).
Figure 10. Equilibrium constants for reaction (4) in supercritical water
at 380 °C (circles) and 400 °C (rectangles) plotted against solution
density. For comparison, gas-phase data61 calculated from thermo-
chemical data are also presented (filled symbols).
3 3
and the (Na )(NO
-) ion
+
Figure 9. Dissociation constants of HNO
pair at 380 °C (circles) and 400 °C (rectangles) plotted against solution
2
5
3
density. (a) Literature data for dissociation of saturated HNO at 325
(
[), 350 (2), 360 (1), and 370 °C (+) are also presented. (b) For
77
comparison, the literature data for the dissociation of NaCl are also
presented at identical temperatures and pressures (filled symbols).
density results in values which are very close to the gas-phase
data, a result which supports our methodology. Relatively linear
plots of log K vs F (in some cases log F) have been observed
74
previously for equilibrium constants in hydrothermal solutions.
The hydration of neutral molecules in SCW is most likely
limited to the first coordination shell and is relatively weak.
Figure 11. Equilibrium constants for reaction (5) in supercritical water
at 380 °C (circles) and 400 °C (rectangles) plotted against solution
density. For comparison, gas-phase data calculated from thermo-
39,79,80
For polar molecules such as HNO3 and NO2, the density effect
is likely to be influenced more by the solvation energies resulting
from dipole-dipole interactions than from partial molar volume
61
chemical data are also presented (filled symbols).
8
1
changes. The dipole moments for polar molecules in eqs 4-7
are 2.17 (HNO3), 1.423 (cis-HNO2), 1.855 (trans-HNO2), 0.313
the equilibrium constants with increasing density (dielectric
constant) for these three reactions is consistent with this decrease
in polarity. Similarly, the observed increase of the equilibrium
constant for the disproportionation reaction 5 is consistent with
the net increase in polarity upon the reaction.
8
2
(
NO2), 0.159 (NO), 0.167 (N2O), and 1.854 (H2O). A net
decrease of dipole moment is observed for reactions 4, 6, and
on the basis of the stoichiometries. The observed decrease of
7

1686 J. Phys. Chem. A, Vol. 103, No. 11, 1999
Chlistunoff et al.
Figure 12. Equilibrium constants for reaction (6) in supercritical water
at 380 °C (circles) and 400 °C (rectangles) plotted against solution
6
1
density. For comparison, gas-phase data calculated from thermo-
chemical data are also presented (filled symbols).
Figure 14. Speciation diagram of the major species in class I
experiments at 380 °C and 31.0 MPa.
Figure 13. Equilibrium constants for reaction (7) in supercritical water
at 380 °C (circles) and 400 °C (rectangles) plotted against solution
6
1
density. For comparison, gas-phase data calculated from thermo-
chemical data are also presented (filled symbols).
Speciation Diagrams Based on the Regressed Equilibrium
Constants. The concentrations of the various species in equi-
librium were calculated from eqs 16-29 on the basis of the
feed concentrations and equilibrium constants regressed from
the model. The results obtained at 380 °C and 31 MPa for three
classes (I, III, and IV) are presented in Figures 14-16,
respectively. For pure nitric acid solutions (class I, Figure 14),
the concentration of each species increases in approximately
the same manner with an increase in the feed HNO3 concentra-
tion. This result is consistent with the relatively constant HNO2
to NO2 molar ratio in Figure 4.
Figure 15. Speciation diagram of the major species in class III
experiments at 380 °C and 31.0 MPa. The initial concentration of HNO
3
-
1
is 0.0232 mol kg
.
observed at 380 °C (Figure 5) but not at 400 °C. This behavior
may result in part from the error introduced in correcting the
measured absorbances (Figure 5) for density when the actual
solution density is approximated by the water density at the
same T and P. Electrolytes increase the density, and the actual
concentration of ions increases upon addition of H2O2 (Figure
15). Therefore, the density corrected absorbances will be
progressively overestimated as the H2O2 feed concentration
increases.
Addition of NaNO2 to a HNO3 solution (Figure 16) leads to
a significant increase in the N2O and NO concentrations and to
a substantial decrease in oxygen concentration. These changes
may be understood easily. Addition of an N (+3) species
(NaNO2) to a solution of N (+5) species (HNO3) destabilizes
The addition of H2O2 reduces the decomposition of HNO3
modestly (Figure 15). The relative decrease in the concentrations
-
6
-10
of the products follows the order N2O (from 10 to 10 mol
-
1
kg , not shown) > NO > HNO2 > NO2, a result consistent
with the oxidation states of nitrogen. The weak decrease in NO2
concentration shown in Figure 15 was not observed experimen-
tally over the whole range of H2O2 concentrations. A decrease
followed by a small increase of the NO2 content was actually

Nitric/Nitrous Acid Equilibria in Water
J. Phys. Chem. A, Vol. 103, No. 11, 1999 1687
The equilibrium constants for all the reactions change with
water density in the direction expected, on the basis of the
change in polarity. In most cases, the measured equilibrium
constants are in good agreement with the gas-phase value when
extrapolated to zero density. While the behavior of ionic
reactions 3 and 10 may be quantified in terms of a modified
3
3,36
Born model,
nonionic reactions 4-7 require other types of
7
9
models to predict solvation effects. These models are useful
for extrapolation of the equilibrium constants to even higher
temperatures, which are also of interest in hydrothermal
technology.
Acknowledgment. We gratefully acknowledge support from
the Department of Energy (DE-FG07-96ER14687), from the
U.S. Army for a University Research Initiative Grant (30374-
CH-URI), from the R.A. Welch Foundation, and from the
Separations Research Program at the University of Texas, a
consortium of over 30 companies.
References and Notes
(
1) Miller, J. A.; Bowman, C. T. Prog. Energy Combust. Sci. 1989,
5, 287.
(2) Wendel, M. M.; Pigford, R. L. AIChE J. 1958, 4, 249.
(3) England, C.; Corcoran, W. H. Ind. Eng. Chem. Fundam. 1974, 13,
1
Figure 16. Speciation diagram of the major species in class IV
experiments at 380 °C and 31.0 MPa. Total nitrogen concentration is
-
1
0
.0208 mol kg
.
373.
(
(
(
4) Kaiser, E. W.; Wu, C. H. J. Phys. Chem. 1977, 81, 187.
5) Kaiser, E. W.; Wu, C. H. J. Phys. Chem. 1977, 81, 1701.
6) Odenbrand, C. U. I.; Andersson, L. A. H.; Brandin, J. G. M.;
products with an oxidation state higher than +3 (NO2, sum of
nitrate ion and nitric acid) and stabilizes products with an
oxidation state lower than +3 (N2O, NO). A relatively constant
Lundin, S. T. Appl. Catal. 1986, 27, 363.
7) Kiovsky, J. R.; Koradia, P. B.; Lim, C. T. Ind. Eng. Chem. Prod.
Res. DeV. 1980, 19, 218.
8) Kato, A.; Matsuda, S.; Kamo, T.; Nakajima, F.; Kuroda, H.; Narita,
(
-
concentration is observed for HNO2. The small increase in NO3
(
concentration upon addition of NaNO2 is caused by an increase
T. J. Phys. Chem. 1981, 85, 4099.
-
-10
-8
-1
in OH concentration (from 10 to 10 mol kg , not shown)
resulting from the decrease in the HNO3 feed concentration and
increase in NaNO2, which exhibits basic properties.
(9) Tester, J. W.; Holgate, H. R.; Armellini, F. J.; Webley, P. A.;
Killilea, W. R.; Hong, G. T.; Barner, H. E. In Emerging Technologies in
Hazardous Waste Management; Tedder, D. W., Pohland, F. G., Ed.; ACS
Symposium Series 518; American Chemical Society: Washington, DC,
1993; pp 35-76.
The quantitative predictions of our model (see above, Figures
(
10) Gloyna, E. F.; Li, L. Waste Manage. 1993, 13, 379.
14-16) agree very well with the measured HNO2 and NO2
(11) Killilea, W. R.; Swallow, K. C.; Hong, G. T. In 2nd International
absorbance bands. Examples include the roughly proportional
increase of both HNO2 and NO2 bands generated by an increase
in HNO3 concentration (Figures 4 and 14), the relative effect
of oxygen on the NO2 and HNO2 bands (Figures 5 and 15),
and the weak sensitivity of the HNO2 band in the HNO3/NaNO2
mixtures to the solution composition (Figures 7 and 16).
The predictions of the model for the species whose absor-
bances were not measured directly also appear to be realistic.
Class IV (mixtures of HNO3 and NaNO2) experiments (Figure
Symposium on Supercritical Fluids; Department of Chemical Engineering,
Johns Hopkins University: Boston, MA, 1991; pp 173-176.
(12) Dell’Orco, P. C. Ph.D. Dissertation Thesis, University of Texas at
Austin, 1994.
(13) Dell’Orco, P. C.; Foy, B. F.; Robinson, J. M.; Buelow, S. J. Haz.
Waste Haz. Mater. 1993, 10, 221.
(14) Dell’Orco, P.; Foy, B.; Wilmanns, E.; Le, L.; Ely, J.; Patterson,
K.; Buelow, S. In InnoVations in Supercritical Fluids. Science and
Technology; Hutchenson, K. W., Foster, N. R., Ed.; American Chemical
Society: Washington, D. C., 1995; pp 179-196.
(15) Oldenborg, R.; Robinson, J. M.; Buelow, S. J.; Dyer, R. B.;
Anderson, G.; Dell’Orco, P. C.; Funk, K.; Willmanns, E.; Knutsen, K.
Hydrothermal Processing of Inorganic Components of Hanford Tank
Wastes; Los Alamos National Laboratories report LA-UR-94:3233; Los
Alamos National Laboratories, 1994.
7) suggested that significant amounts of a species with a nitrogen
oxidation state lower than +3 were formed. Indeed, a significant
contribution from NO is predicted by the model (Figure 16).
-
Concentration changes for other species including HNO3, NO3 ,
(16) Proesmans, P. I.; Luan, L.; Buelow, S. J. Ind. Eng. Chem. Res.
1997, 36, 1559.
+
-
N2O, oxygen, and the ion pairs NaOH and (Na )(NO3 ) follow
trends, which may be expected from the feed solution composi-
tion and/or oxidation state of nitrogen.
(
17) Adschiri, T.; Kanazawa, K.; Arai, K. J. Am. Ceram. Soc. 1992,
7
7
5, 2615.
(18) Adschiri, T.; Kanazawa, K.; Arai, K. J. Am. Ceram. Soc. 1992,
5, 1019.
(
(
19) Shoppelrei, J. W.; Brill, T. B. J. Phys. Chem. A 1997, 101, 8593.
20) Shoppelrei, J. W.; Kieke, M. L.; Wang, X.; Klein, M. T.; Brill, T.
Conclusions
B. J. Phys. Chem. 1996, 100, 14343.
Chemical equilibria in nitric acid solutions in SCW may be
measured directly from the absorbance bands of NO2 and HNO2
from 380-400 °C by UV-vis spectroscopy. These absorbance
bands are similar to those at ambient temperature, except the
fine structure of NO2 is far less than that in the gas phase. All
of the spectra appear to be nearly completely reversible upon
cooling. Equilibrium constants for reactions 3-7 and 10 can
be determined by optimization of the measured and calculated
absorbances of NO2 and HNO2. Other important decomposition
products that do not absorb in the wavelengths studied include
O2, NO, and, in mixtures of HNO3 and NaNO2, N2O.
(
(
(
21) Maiella, P. G.; Brill, T. B. Appl. Spectrosc. 1996, 50, 829.
22) Spohn, P. D.; Brill, T. B. J. Phys. Chem. 1989, 93, 6224.
23) Ratcliffe, C. I.; Irish, D. E. Can. J. Chem. 1985, 63, 3521.
(24) Oscarson, J. L.; Gillespie, S. E.; Izatt, R. M.; Chen, X.; Pando, C.
J. Solution Chem. 1992, 21, 789-801.
(
(
25) Marshall, W. L.; Slusher, R. J. Inorg. Nucl. Chem. 1975, 37, 2165.
26) Marshall, W. L.; Slusher, R. J. Inorg. Nucl. Chem. 1975, 37, 1191.
(27) Marshall, W. L.; Jones, E. V. J. Inorg. Nucl. Chem. 1974, 36, 2313.
(28) Barton, C. J.; Hebert, G. M.; Marshall, W. L. J. Inorg. Nucl. Chem.
961, 21, 141.
1
(
29) Seward, T. M. Geochim. Cosmochim. Acta 1984, 48, 121.
(30) Heinrich, C. A.; Seward, T. M. Geochim. Cosmochim. Acta 1990,
54, 2207.

1688 J. Phys. Chem. A, Vol. 103, No. 11, 1999
Chlistunoff et al.
(
31) Gammons, C. H.; Seward, T. M. Geochim. Cosmochim. Acta 1996,
(60) Verhoek, F. H.; Daniels, F. J. Am. Chem. Soc. 1931, 53, 1250.
(61) Chase, M. W., Jr.; Davies, C. A.; Downey, J. R., Jr.; Frurip, D. J.;
McDonald, R. A.; Syverud, A. N. JANAF Thermochemical Tables, 3rd ed.;
American Chemical Society and the American Institute of Physics for the
National Bureau of Standards, 1985; Vol. 14.
(62) Lumme, P.; Lahermo, P.; Tummavuori, J. Acta Chem. Scand. 1965,
19, 2175.
6
0, 4295.
(
993.
(
32) Chlistunoff, J. B.; Johnston, K. P. J. Phys. Chem. B 1998, 102,
3
33) Xiang, T.; Johnston, K. P. J. Phys. Chem. 1994, 98, 7915.
34) Xiang, T.; Johnston, K. P.; Wofford, W. T.; Gloyna, E. F. Ind.
(
Eng. Chem. Res. 1996, 35, 4788.
(
(
35) Xiang, T.; Johnston, K. P. J. Solution Chem. 1997, 26, 13.
36) Ryan, E. T.; Xiang, T.; Johnston, K. P.; Fox, M. A. J. Phys. Chem.
(63) Lumme, P.; Tummavuori, J. Acta Chem. Scand. 1965, 19, 617.
(64) Tummavuori, J.; Lumme, P. Acta Chem. Scand. 1968, 22, 2003.
A 1997, 101, 1827.
(65) Mesmer, R. E.; Marshall, W. L.; Palmer, D. A.; Simonson, J. M.;
(
(
(
(
37) Shin, T. W.; Kim, K.; Lee, I.-J. J. Solution Chem. 1997, 26, 379.
38) Smith, S.; Lasdon, L. J. Comput. 1992, 4, 1.
39) Bennett, G. E.; Johnston, K. P. J. Phys. Chem. 1994, 98, 441.
40) Official Methods of Analysis of AOAC International, 16th ed.;
Holmes, H. F. J. Solution Chem. 1988, 17, 699.
(
66) Chen, X.; Izatt, R. M.; Oscarson, J. L. Chem. ReV. 1994, 94, 467.
(67) Ho, P. C.; Palmer, D. A. Geochim. Cosmochim. Acta 1997, 61,
3027.
AOAC International: Arlington, 1995; Vol. II, pp A.1.12.
41) Kolthoff, I. M.; Sandell, E. B. Textbook of QuantitatiVe Inorganic
Analysis, 3rd ed.; The Macmillan Company: New York, 1952.
(
(
68) Dixon, J. K. J. Chem. Phys. 1940, 8, 157.
69) Nakayama, T.; Kitamura, M. Y.; Watanabe, K. J. Chem. Phys. 1959,
(
30, 1180.
(
42) Hall, T. C.; Blacet, F. E. J. Chem. Phys. 1952, 20, 1745.
(
70) Corcoran, T. C.; Beiting, E. J.; Mitchell, M. O. J. Mol. Spectrosc.
(43) Norris, M. S.; Fleck, S. A.; Lichtenfels, D. H. Anal. Chem. 1955,
1
992, 154, 119.
2
7, 1565.
44) Schneider, W.; Moortgat, G. K.; Tyndall, G. S.; Burrows, J. P. J.
Photochem. Photobiol. 1987, 40, 195.
45) Mihalcea, R. M.; Baer, D. S.; Hanson, R. K. Appl. Optics 1996,
5, 4059.
(
71) Armerding, W.; Walter, J.; Ruger, Chr.; Spiekermann, M.; Comes,
F. J. Ber. Bunsen-Ges. Phys. Chem. 1993, 97, 1440.
72) Ziegler, K. J.; Chlistunoff, J.; Lasdon, L.; Johnston, K. P., 1999,
submitted.
(
(
(
3
1
(73) Singer, K.; Vamplew, P. A. J. Chem. Soc. 1956, 3971.
(46) Chao, J.; Wilhott, R. C.; Zwolinski, B. J. Thermochim. Acta 1974,
(74) Marshall, W. L.; Franck, E. U. J. Phys. Chem. Ref. Data 1981, 10,
0, 359.
1.
(
(
(
(
(
(
(
(
47) Hisatsune, I. C. J. Phys. Chem. 1961, 65, 2249.
48) McLane, C. K. J. Chem. Phys. 1949, 17, 379.
(75) Ho, P. C.; Palmer, D. A. J. Solution Chem. 1996, 25, 711.
(
76) Pitzer, K. S. In ActiVity Coefficients in Electrolyte Solutions, 2nd
49) Takagi, J.; Ishigure, K. Nucl. Sci. Eng. 1985, 89, 177.
50) Strickler, S. J.; Kasha, M. J. Am. Chem. Soc. 1963, 85, 2899.
51) Longstaff, J. V. L.; Singer, K. J. Chem. Soc. 1954, 2604.
52) Bunton, C. A.; Stedman, G. J. Chem. Soc. 1958, 2440.
53) Bayliss, N. S.; Watts, D. W. Aust. J. Chem. 1956, 9, 319.
54) Bennet, M. R.; Brown, G. M.; Maya, L.; Posey, F. A. Inorg. Chem.
ed.; CRC Press: Boca Raton, 1991; pp 75-153.
(77) Ho, P. C.; Palmer, D. A.; Mesmer R. E. J. Solution Chem. 1994,
2
1
3, 997.
(78) Balbuena, P. B.; Johnston, K. P.; Rossky, P. J. J. Phys. Chem. 1996,
00, 2706.
1
982, 21, 2461.
(79) Luo, H.; Tucker, S. C. J. Phys. Chem. 1997, 101, 1063.
(80) Sitkoff, D.; Sharp, K. A.; Honig, B. J. Phys. Chem. 1994, 98, 1978.
(81) Ji, Q. E. M. E.; van Eldik, R.; Johnston, K. P.; Goates, S. R.; Lee,
M. L. J. Phys. Chem. 1995, 99, 13461.
(82) CRC Handbook of Chemistry and Physics, 78th ed.; CRC Press:
Boca Raton, 1997.
(
(
(
(
(
55) Wikstrom, M. M.; Nobe, K. I&EC Process Des. DeV. 1965, 4, 191.
56) Sponer, H.; Bonner, L. G. J. Chem. Phys. 1940, 8, 33.
57) Friend, J. A.; Lyons, L. E. J. Chem. Soc. 1959, 1572.
58) Mason (Banus), J. J. Chem. Soc. 1959, 1288.
59) Beattie, I. R.; Vosper, A. J. J. Chem. Soc. 1961, 2106.