Reaction sequence and kinetics of uranium nitride decomposition

Article abstract of DOI:10.1021/ic901165j

The reaction mechanism and kinetics of the thermal decomposition of uranium dinitride/uranium sesquinitride to uranium mononitride under inert atmosphere at elevated temperature were studied. An increase in the lattice parameter of the UN₂/α-U<

Full text of DOI:10.1021/ic901165j

Inorg. Chem. 2009, 48, 10635–10642 10635
DOI: 10.1021/ic901165j
Reaction Sequence and Kinetics of Uranium Nitride Decomposition
G. W. Chinthaka Silva,† Charles B. Yeamans,‡ Alfred P. Sattelberger,†,§ Thomas Hartmann,† Gary S. Cerefice,†
and Kenneth R. Czerwinski*,†
†Harry Reid Center for Environmental Studies, University of Nevada, Las Vegas, Box 454009, 4505 Maryland
Parkway, Las Vegas, Nevada 89154, ‡Department of Nuclear Engineering, University of California, Berkeley,
4155 Etcheverry Hall, M.C. 1730, Berkeley, California 94720-1730, and §Energy Sciences and Engineering
Directorate, Argonne National Laboratory, 9700 Cass Avenue, Building 208, Argonne, Illinois 60439
Received June 17, 2009
The reaction mechanism and kinetics of the thermal decomposition of uranium dinitride/uranium sesquinitride to
uranium mononitride under inert atmosphere at elevated temperature were studied. An increase in the lattice
parameter of the UN2/R-U2N3 phase was observed as the reaction temperature increased, corresponding to a
continuous removal of nitrogen. Electron density calculations for these two compounds using XRD powder patterns of
the samples utilizing charge-flipping technique were performed for the first time to visualize the decrease in nitrogen
level as a function of temperature. Complete decomposition of UN2 into R-U2N3 at 675 °C and the UN formation after a
partial decomposition of R-U2N3 at 975 °C were also identified in this study. The activation energy for the
decomposition of the UN2/R-U2N3 phase into UN, 423.8 ( 0.3 kJ/mol (101.3 kcal/mol), was determined under an
inert argon atmosphere and is reported here experimentally for the first time.
1. Introduction
results in the formation of a U2N3þx layer and UO2 impurity
phase.9 Low-temperature methods include the reaction of
uranium carbide (UC)10,11 and uranium fluoride (UF4) with
ammonia.12 In a previous study, formation of U2N3 was
detected when UC was heated in a thermogravimetric bal-
ance under N2.13
UN is the lowest-stoichiometry nitride of the U-N solid
system, commonly considered to have three distinct com-
pounds: uranium dinitride (UN2), uranium sesquinitride
(U2N3), and UN.14 Higher uranium nitrides decompose to
UN by the following sequence of reactions:
Uranium mononitride (UN) has been considered as a
potential fuel for Generation IV nuclear reactor systems
due to its favorable physical and chemical properties.1 In
particular, fast breeder systems benefit from the higher fissile-
atom density in UN as compared to UO2, and its superior
mechanical, thermal, and radiation stability as compared to
metallic fuel.2 Additionally, accelerator-driven subcritical
systems for minor-actinide transmutation can also utilize a
nitride-based inert matrix fuel.3 Of the techniques used to
synthesize UN from UO2 for nuclear fuel, carbothermic
reduction4-6 is the most common in industrial applications.
Sol-gel methods3,7 are also proposed for the synthesis of
actinide nitrides. In some studies, both sol-gel methods and
carbothermic reduction have been used,8 as has arc-melting
of pure uranium metal in a nitrogen atmosphere. Arc-melting
1
2UN2 f R-U2N3 þ N2
ð1Þ
2
1
R-U2N3 f 2UN þ N2
ð2Þ
2
*To whom correspondence should be addressed. E-mail: czerwin2@
unlv.nevada.edu. Tel: (702) 895 0501. Fax: (702) 895 3094.
(1) Arai, Y.; Minato, K. J. Nucl. Mater. 2005, 344, 180.
(2) Mizutani, A.; Sekimoto, H. Ann. Nucl. Energy 1998, 25, 1011.
(3) Streit, M.; Ingold, F. J. Eur. Ceram. Soc. 2005, 25, 2687.
(4) Arai, Y.; Fukushima, S.; Shiozawa, K.; Handa, M. J. Nucl. Mater.
1989, 168, 280.
(5) Ogawa, T.; Shirasu, Y.; Minato, K.; Serizawa, H. J. Nucl. Mater.
1997, 247, 151.
(6) Minato, K.; Akabori, M.; Takano, M.; Arai, Y.; Nakajima, K.; Itoh,
A.; Ogawa, T. J. Nucl. Mater. 2003, 320, 18.
The UN2 compound is a face-centered cubic CaF2-type
structure with a space group of Fm3hm compared to the
(10) Katsura, M.; Hirota, M.; Miyake, M. J. Alloys Compd. 1994, 213/
214, 440.
(11) Katsura, M. Solid State Ionics 1991, 49, 225.
(12) Silva, G. W. C.; Yeamans, C. B.; Ma, L.; Cerefice, G. S.; Czerwinski,
K. R.; Sattelberger, A. P. Chem. Mater. 2008, 20, 3076.
(13) Hanson, L. A. J. Nucl. Mater. 1966, 19, 15.
(14) Morss L. R., Edelstein N. M., Fuger J., Katz J. J., Eds. The Chemistry
of the Actinides and Transactinide Elements, 3rd ed.; Chapman & Hall: New
York, 2006; Vol. 1.
(7) Ganguly, C.; Hegde, P. V. J. Sol-Gel Sci. Technol. 1997, 9, 285.
(8) Mukerjee, S. K.; Dehadraya, J. V.; Vaida, V. N.; Sood, D. D. J. Nucl.
Mater. 1991, 185, 39.
(9) Sole, M. J.; Van Der Walt, C. M. Acta. Metall. 1968, 16, 501.
r
2009 American Chemical Society
Published on Web 10/21/2009
pubs.acs.org/IC

10636 Inorganic Chemistry, Vol. 48, No. 22, 2009
Silva et al.
Figure 1. Equilibrium phase diagram of the pure UNx system.23
body-centered cubic Mn2O3-type structure of R-U2N3 with a
Ia3 space group.15 The fcc UN2 structure shows a range of
compositions described as UNx where 2.0 g x g 1.75.16 The
UN2/R-U2N3 system forms stable solid solutions at all inter-
mediate compositions above UN1.5417 while reports indicate up
to 1.45 N/U molar ratio for the R-U2N3 composition. In these
solid solutions, a continuous change of the uranium and
nitrogen atomic positions within the unit cell with composition
is also reported.16,17 As a result, the names “UN2” and “U2N3”
are often applied interchangeably to the UN2/R-U2N3 solid
solution chemical phase. Since the apparent crystallographic
change from CaF2-type face-centered cubic UN2 to Mn2O3-
type body-centered cubic R-U2N3 occurs at a composition of
reported.12 The current work also observed the formation
of a UO2 secondary phase up to 10 wt % in UN, whereas
only minor levels of UO2 (up to 1 wt %) were identified
in UN2 and U2N3 samples. Pure UN is always slightly
substoichiometric, and has a maximum nitrogen-rich com-
position of UN0.995 at 1100 °C and below.21 Small oxygen
impurities (<1 wt %) stabilize R-U2N3 at higher tempera-
tures,22 so the apparent equilibrium at intermediate com-
positions in the U-N system is between UN1.45 and
UN0.995, having the common designations R-U2N3 and
UN, respectively, as long as some oxygen impurities are
present. A phase diagram for the UNx system is shown in
Figure 1 to provide detail.
Even though a number of reports exist on the presence of
these different uranium nitrides under various experimen-
tal conditions and physical properties, there is little infor-
mation available on the reaction kinetics associated with
the formation of UN from UN2 or R-U2N3. The current
study focuses on confirming that the aforementioned
decomposition reaction sequence (eqs 1 and 2) is correct,
and determining the relevant kinetic parameters and acti-
vation energies. Formation temperatures of R-U2N3 and
UN were identified by systematically decomposing UN2 at
various temperatures under an inert argon atmosphere.
The kinetics of these reactions was analyzed using a
pseudo-first-order model because it shows relatively better
match to the experimental results compared to a zero-order
analysis. Arrhenius plots were generated to determine the
activation energies.
UN1.75
,
18 it is common to refer to solutions with stoichiometry
above UN1.75 as UN2 and below it as R-U2N3. That common
convention will be used in this report for the sake of consistency
with previous studies.
We find that the higher-stoichiometry uranium nitrides
are more stable than UN toward adventitious amounts of
oxygen present in our experimental setups.12,19 The higher
nitrides are actually synthesized at lower temperatures
than UN,5,20 which contribute to their greater apparent
oxygen-stability according to the observations previously
(15) Rundle, R. E.; Baenziger, N. C.; Wilson, A. S.; McDonald, R. A.
J. Am. Chem. Soc. 1948, 70, 99.
(16) Tagawa, H. J. Nucl. Mater. 1974, 51, 78.
(17) Serizawa, H.; Fukuda, K.; Ishii, Y.; Morii, Y.; Katsura, M. J. Nucl.
Mater. 1994, 208, 128.
(18) Tagawa, H.; Masaki, N. J. Inorg. Nucl. Chem. 1974, 36, 1099.
(19) Yeamans, C. B.; Silva, G. W. C.; Cerefice, G. S.; Czerwinski, K. R.;
Hartmann, T.; Burrell, A. K.; Sattelberger, A. P. J. Nucl. Mater. 2008, 374
(1-2), 75.
(21) Hoenig, C. L. J. Am. Chem. Soc. 1971, 54, 391.
(22) Benz, R.; Balog, G.; Baca, B. H. High Temp. Sci. 1970, 2, 221.
(20) Mallett, M. W.; Gerds, A. F. J. Electrochem. Soc. 1955, 102(6), 292.

Article
Inorganic Chemistry, Vol. 48, No. 22, 2009 10637
2. Experimental Details
Table 1. Products Observed after Heating UN2 at Different Temperatures and
Time Intervals under Argon
2.1. Uranium Nitride Synthesis. Ammonolysis of UF4 was
used to synthesize UN2. A 1015.76 (5) mg sample of UF4
(International Bio-Analytical Industries, Inc.) was loaded in
a quartz-glass boat wrapped with platinum foil and placed
inside a 25.4 mm diameter quartz-glass tube, capped on either
end with 25 mm quartz-glass Solv-Seal fittings (Andrews Glass
Co., Inc.). Pyrex Solv-Seal caps fitted with 15 mm high vacuum
Teflon stopcocks sealed the tube and allowed a controlled
atmosphere to blanket the sample. The sample was held at
800 °C for 1 h under ammonia gas (research grade, Praxair)
after which 858.53 (5) mg of UN2 was obtained. The mass loss
for the transformation of UF4 into UN2 was 157.23(5) mg. This
indicates an extra 2 mg mass loss compared to the expected value
of 155.2 mg. This extra additional mass loss is attributed to
errors associated with the experimental measurements. A
218.74 (5) mg sample of R-U2N3 was synthesized by decompos-
ing 225.35 (5) mg of the synthesized UN2 under an inert atmo-
sphere (ultrahigh purity argon, 99.9999%, Praxair) at 700 °C for
1 h. The mass loss was 6.6 mg which is relatively close to the
expected value of 5.9 mg. Batches of different UN2 and R-U2N3
masses were used to synthesize UN to determine the reaction
kinetics and temperature effects. A detailed description of these
nitrides syntheses can be found in previous publications.12,19
2.2. Characterization Methods. X-ray powder diffraction
(XRD) patterns were obtained using a Philips PANalytical
X’Pert Pro instrument with a Ni-filtered Cu KR radiation.
The patterns were collected using 40 mA current and 40 kV
tension at room temperature. Chemical phases in the samples
were initially determined matching the XRD powder patterns
with the International Center for Diffraction Data (ICDD)
database patterns. Following are the reference patterns used
for separate compounds: UN2 (01-073-1713), U2N3 (01-073-
1712), UN (00-032-1397), and UO2 (00-041-1422). An internal
LaB6 standard from NIST (SRM 660a) was admixed with the
uranium nitride samples to allow for precise lattice parameter
refinement and to optimize the quality of Rietveld analysis as
performed.
temperature (°C)
time of heating (min.)
products observed
500
600
30
30
240
30
60
30
30
30
30
30
240
30
60
30
30
30
30
UN2
UN2
UN2
UN2
UN2
U2N3
U2N3
U2N3
U2N3
U2N3
U2N3
U2N3
U2N3
U2N3, UN
U2N3, UN
U2N3, UN
UN
650
675
700
750
800
900
950
975
1000
1050
1100
three UN2 samples of approximately 50 mg each at 500,
700, and 1100 °C under flowing high-purity argon
(99.9999%) for 30 min. Heating the first sample at
500 °C resulted in no measurable decomposition of the
UN2. At 700 °C, UN2 was completely converted to
R-U2N3. After heating at 1100 °C for 30 min, the only
uranium nitride phase identified in the samples was UN.
The latter sample also contained a 5.3 (1) wt % secondary
UO2 phase, most likely due to trace oxygen contamina-
tion in the experimental system. Subsequent experiments
were conducted in two different temperature ranges of
500-700 °C and 700-1100 °C (Table 1) to refine the
temperatures in which the decomposition reactions in eqs
1 and 2 occur. At each of these temperatures, the reactants
were heated for 30 min initially to determine where the
decomposition reactions occur. These experiments
showed that the UN2 decomposition to R-U2N3 starts
at about 675 °C, and the second decomposition to UN
begins near 975 °C. Therefore, the decompositions at 650
and 950 °C were studied further with up to 60 min of
heating. These two experiments showed neither forma-
tion of R-U2N3 at 650 °C nor formation of UN at 950 °C.
Heating of UN2 at 600 and 900 °C was also conducted for
up to 240 min to determine how duration would affect
UN2 decomposition. At 600 °C (a = 0.53050(6) nm), no
R-U2N3 formation occurred. No UN formation was
perceived at 900 °C as well (a = 1.068890(7) nm). Even
though no phase transformations were observed, increase
in the lattice parameters of UN2 and R-U2N3 was detected
as described in Figure 2. Also, over the temperature range
from 975 to 1100 °C, both R-U2N3 and UN were observed
in the product.
Electron density maps of uranium nitrides were calculated
using the XRD powder patterns of the samples. Le Bail decom-
position24 was used to extract the individual observed structure
factor amplitudes (Fobs) of the XRD powder pattern using
Jana2000.25 These observed structure factors were then used
to calculate the electron density maps of the compounds by the
charge-flipping algorithm through the Superflip program.26 The
calculated electron density maps were visualized using UCSF
Chimera.27 A Tecnai-G2-F30 supertwin transmission electron
microscope system with a 300 keV Schottky field emission gun
was used in TEM imaging. Bright field (BF) and the high
resolution (HRTEM) modes of TEM were utilized in sample
characterization. All TEM images were recorded using a slow
scan CCD camera attached to a Gatan GIF 2000 energy filter.
3. Results
3.1. UN2 Decomposition. The thermal behavior of
UN2 in an inert atmosphere was determined by heating
The refined lattice parameters of UN2 and U2N3 were
shown to vary with respect to the temperature (Figure 2).
Lattice parameters of both these nitrides change linearly
as a function of temperature. A 0.0004 nm change was
determined in the lattice parameter of UN2 over a 150 °C
temperature range. The lattice parameter change deter-
mined for the R-U2N3 samples over a 275 °C temperature
range was about 0.004 nm, a 6-fold increase compared to
the change of the UN2 lattice parameter relative to the
final lattice parameter of each compound. Possible rea-
sons for this observation are discussed in the following
paragraph using electron microscopic characterizations.
(23) Yeamans, C. B. Synthesis of Uranium Fluorides from Uranium
Dioxide with Ammonium Bifluoride and Ammonolysis of Uranium Fluor-
ides to Uranium Nitrides. Ph.D. dissertation. University of California,
Berkeley, Berkeley, California, 2008,; pp 12-18.
(24) Bail, A. Le.; Duroy, H.; Fourquet, J. L. Mater. Res. Bull. 1998, 23,
447-452.
::
::
::
^
~
(25) DusIek, M.; PetrIoAcIek, V.; Wunschel, M.; Dinnebier, R. E.; van
Smaalen, S. J. Appl. Crystallogr. 2001, 34, 398-404.
(26) Palatinus, L.; Chapuis, G. J. Appl. Crystallogr. 2007, 40, 786-790.
(27) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.;
Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. J. Comput. Chem. 2004, 25 (13),
1605-1612.

10638 Inorganic Chemistry, Vol. 48, No. 22, 2009
Silva et al.
Figure 5. HRTEM images of UN2 (a) and U2N3 (b). Insets are the
corresponding FFT micrographs of each HRTEM image.
Figure 2. Lattice parameters of UN2 and R-U2N3 as a function of the
temperature used for decomposition.
Figure 3. HRTEM images of UN2 particles. Insets are the BF images of
the particles used in the HRTEM imaging. Squares indicate amorphous
areas while circled areas demonstrate disruptions with some amorphous
characteristics especially in the lower end circle.
Figure 6. Change in the lattice parameter of the UN2/R-U2N3 system
with respect to the N/U molar ratio.
areas of the imaged particle. Therefore, the disorders
identified in these samples at nanoscale are due to a second
phase other than UO2. Because UNx represents a range of
“x” values, another UNx with different chemical composi-
tion compared to the main chemical phase in the sample is
the probable reason for such nanostructural disruptions.
Presence of another minor UNx phase within the main
UN2 is therefore the reason for the solid solution char-
acteristics in the sample. These separate phases could not
be differentiating from the XRD powder patterns but
rather from the nanostructural changes or with lattice
parameter changes in the compound. Therefore, these
nanostructural changes in UN2 are due to a UNx phase
inferring in solid solution characteristics. Change in the
UN2 lattice parameter as a function of temperature imply-
ing the range of compositions of UNx present (Figure 2)
also supports the observation of solid solution behavior.
The U2N3 sample also showed similar solid solution
behavior from its nanostructure characteristics. Observa-
tion of nanosized crystalline domains (Figure 4a) and
nanostructural disruptions (Figure 4b) due to dislocations
of atomic layers and incomplete crystallization of grains at
this scale in some of the U2N3 particle areas verify this
presumption. Presence of nanosized crystalline domains
Figure 4. HRTEM images of U2N3 particles shown in the inset (the BF
image). Circle shows a disrupted nanoparticle area.
HRTEM images of both UN2 (Figure 3) and U2N3
(Figure 4) micro particle areas presented here contained
some deformations in their nanostructures. In the case of
UN2, bulk particle areas included single crystal charac-
teristics. However, some of these grains had amorphous
areas as shown in Figure 3a (with an arrow and the
rectangle box). As a specific example, a single crystal
particle (inset of Figure 3b) exhibited some disruptions at
the nanoscale. These abnormal observations are high-
lighted in Figure 3b. These nanostructural disorders
occur either from the specimen preparation techniques
or the formation of a secondary phase. If the first reason is
dominant, these disruptions should extend up to bulk
particle areas and such large disruptions could not be
observed (inset of Figure 3b). The lattice fringes do not
represent a secondary UO2 phase in the nano particle

Article
Inorganic Chemistry, Vol. 48, No. 22, 2009 10639
smaller than that of the UN2 also denotes a lower particle
crystallization of U2N3 than that of UN2. Low point-
to-point resolution of the nanostructure of U2N3 depicted
in Figure 5b compared to that of UN2 in Figure 5a
supports this observation. Low crystallization of the sam-
ple and the presence of small crystalline domains in U2N3
suggest a faster change in structure with respect to the
temperature than that of UN2 explaining the larger change
in the lattice parameter of U2N3 as a function of tempera-
ture compared to UN2 as mentioned above. This behavior
should also promote formation of UN from the U2N3
phase at elevated temperatures such as 1000 and 1100 °C
because UN forms more crystalline12 chemical phase with
a low degree of solid solution behavior with the U2N3
precursor.
The N/U molar ratios of these UNx samples were
estimated by correlation plots for UN2 and R-U2N3
(Figure 6).15-18,28-33 A value of 0.521 nm was taken as
the lattice parameter of the stoichiometric UN2 by assum-
ing behavior analogous to AmO2 as previously employed
by Tagawa et al.18 The lattice parameter values of the
UN2 samples synthesized and annealed up to 650 °C
correlated with those found for samples with approxi-
mate N/U molar ratios of 1.75-1.8. However, a lack of
data for the UNx system in the 1.8 e x e 2.0 region may
imply a higher error in the N/U molar ratio estimations
compared to the 1.45 e x e 1.75 region. Lattice para-
meters of R-U2N3 samples synthesized in this study show
variation comparable to that seen by other authors.
Approximate N/U molar ratios determined using the
correlation plots in Figure 6 for UN2 and R-U2N3 are
summarized in Table 2.
Table 2. Approximated N/U Molar Ratios Determined for UNx Samples from
Correlation Plots in Figure 6
estimated N/U
molar ratio
temperature (°C)
lattice parameter: a (nm)
500
600
650
700
750
800
900
950
975
0.53027 ( 0.00003
0.53050 ( 0.00006
0.53060 ( 0.00006
1.06540 ( 0.00014
1.06570 ( 0.00008
1.06690 ( 0.00011
1.0688 ( 0.0006
1.765
1.757
1.753
1.678
1.664
1.607
1.517
1.493
1.479
1.06930 ( 0.00011
1.0696 ( 0.0002
Figure 7. Average electron density maps of UN2 calculated from XRD
powder patterns. Same density level was used in all the images for
comparison. A model of the UN2 unit cell in [001] direction is also shown
at the beginning of the series.
Figure 7 shows a model of UN2 unit cell together with
the average electron density maps in the [001] direction.
The electron density at nitrogen positions decreases as the
temperature is increased. The decrease in N/U molar ratio
with respect to the sample temperature is thus supported
by these electron density maps calculated from XRD
powder patterns of the samples.
Electron density maps of U2N3 samples were also
calculated (Figure 8). Due to the complex nature of the
U2N3 unit cell as indicated in the model, observation of
the decrease in average electron density at nitrogen posi-
tions could not be easily observed as in UN2. A close-
packed space filling of U2N3 unit cell was identified in the
electron density maps of first two samples synthesized at
675 and 700 °C. Samples synthesized at 750-950 °C
showed more loose packing characteristics of the unit cell
Figure 8. Average electron density maps of U2N3 unit cell in the [001]
direction.
Figure 9. XRD powder refinement of the UN (97 wt %) sample synthesized by heating a UN2 sample at 1100 °C for 23 min under an inert atmosphere.
Highlighted is the calculated pattern of UN.

10640 Inorganic Chemistry, Vol. 48, No. 22, 2009
Silva et al.
indicating low density. At 750 and 800 °C temperatures,
the samples contain comparatively high electron den-
sity at nitrogen atom locations. The electron density
decreased when the temperature was raised to 900 or
950 °C. Thus, a reduction of U-N bonding characteri-
stics due to a decrease in the N/U molar ratio can be seen
when the temperature is increased in the sample synthesis.
These density maps calculated correlate with the observed
reaction speciation.
A number of UN samples were synthesized by decom-
posing UNx at different temperatures. Secondary chemi-
cal phases, including a UO2 impurity phase (<10 wt. %)
and the incompletely decomposed R-U2N3 phase were
observed in these samples. The UN samples prepared by
decomposition of the UN2 at 1050 and 1100 °C showed
greater product phase content than samples prepared at
975 and 1000 °C due to the presence of larger amounts of
incompletely decomposed R-U2N3 phase at the lower
temperatures. The UO2 phase impurity was typically less
than 5 wt. % at 1100 °C (Figure 9). A plot of the lattice
parameters of these UN samples as a function of tem-
perature is shown in Figure 10. The overall change in
lattice parameter of UN synthesized at the reported four
different temperatures in Figure 10 is about 0.0001 nm. At
975 °C, in which the lowest UN wt. % was observed, the
lattice parameter of UN was the highest. At other three
temperatures, UN shows a range of lattice parameters
even with high phase wt. % at 1100 °C. These observa-
tions demonstrate a distribution of the UN lattice para-
meter over a considerably wide range even at one specific
temperature, as was reported elsewhere confirming the
formation of UN and its behavior with respect to tem-
perature.33 Furthermore, no intermediate chemical
phases such as β-U2N3 were identified between UN and
R-U2N3 explaining such a variation of lattice parameter
of UN phase. Absence of such an intermediate phase was
confirmed by heating UN2 at temperatures near 1000 °C
(Table 3) for different time periods. In these experiments
no additional phases other than UO2 were observed.
Another experiment illustrated that heating UN2 under
a N2-H2 mixture (8% H2) for 30 min at 1100 °C, for an
effective nitrogen pressure of about 92 kPa, produced
R-U2N3 with no trace of UN. This is consistent with the
species predicted by the phase diagram in Figure 1 near
the K = 5 (N2 pressure = 105 Pa) isobar at 1100 °C.
3.2. Kinetics of UN2 Decomposition. UN2 decomposi-
tion studies revealed that R-U2N3 formed at temperatures
between 675 and 975 °C. Due to the solid solution
Figure 10. Lattice parameter of UN as a function of the temperature
used for decomposition.
Table 3. Phase Distribution of the UN2 Sample after Heating It at 1000 °C for
Different Time Periods
phase distribution (wt%)
Table 4. Rate Constants of the UNx Decomposition Reaction at 1000, 1050, and
1100 °C
time (min.)
UN
U2N3
UO2
temperature (°C)
rate constant: k (s-1 ꢀ 103)
ln (k /10-6 s-1
)
30
4.9
17.3
79.4
75.0
76.3
95.1
82.7
18.4
22.6
20.5
120
240
300
360
2.2
2.4
3.2
1000
1050
1100
0.07 ( 0.01
0.21 ( 0.02
1.3 ( 0.3
4.22 ( 0.16
5.37 ( 0.10
7.14 ( 0.24
Figure 11. Pseudo-first-order kinetics of UNx decomposition at 1050 °C (a) and Arrhenius plot for the reaction (b).

Article
Inorganic Chemistry, Vol. 48, No. 22, 2009 10641
Figure 12. XRD powder refinement of U2N3 sample after heating it at 1100 °C for 30 min under N2- 8% H2. The U2N3 calculated pattern is highlighted.
behavior of UN2 and U2N3 and overlapping peaks in the
XRD, the kinetics of this decomposition reaction could
not be determined. For the conversion of UN2/R-U2N3 to
UN, the reaction kinetics was evaluated at temperatures
of 1000, 1050, and 1100 °C. The kinetic data associated
with this unimolecular reaction at 1050 °C are shown in
Figure 11a, revealing an approximate match between the
growth of UN and the decay of R-U2N3. Pseudo-first-
order was used in determining the kinetics involved in the
UN formation since it correlates well with the exponential
growth of UN. The zero-order is not valid in this case
since at a given time the amount of UN formed was not
always a constant value but it was a function of the
amount of UN2 used at the beginning of each experiment.
The slight deviation observed in Figure 11 is attributed to
the formation of the secondary UO2 chemical phase.
These observations suggest the transition through an
intermediate chemical phase of R-U2N3-x, likely having
the uranium-rich boundary composition of UN1.54 that
reaches a steady-state concentration close to zero. Similar
behavior was found for the samples decomposed at 1000
and 1100 °C. Figure 11b displays the Arrhenius plot for
this decomposition reaction, and activation energy of
423.8 ( 0.3 kJ/mol was determined. Table 4 summarizes
the rate constants determined in the experiment.
UNx were identified. The UN2 phase is the only species present
at temperatures less than 675 °C. In the second temperature
region, 675-975 °C, UN2 completely decomposes to R-U2N3.
Between 975 and 1100 °C, both UN and R-U2N3 were detected
depending on the temperature and time of heating, with UN
being the stable phase in the higher end of the region.
Higher crystallinity of UN2 in its nanostructure was identi-
fied by HRTEM imaging compared to that of U2N3. Solid
solution behavior of the UN2 and U2N3 samples respectively
in the above-mentioned two temperature ranges was also
identified with nanostructural changes of the samples. Lower
crystallinity and the larger disruptions of the nanostructure of
U2N3 compared to UN2 suggested a rapid change in its crystal
structure forming a more and well crystallized chemical phase
UN. This explains why the lattice parameter in the UNx
system changed much more rapidly above 675 °C. The result
supports the conclusion that the reduction of the N/U ratio
through decomposition of the UNx system is kinetically
favorable at elevated temperatures. This effect is large enough
that even though UN is thermodynamically favorable at a
lower temperature, higher-composition UNx phases are kine-
tically stable up to 675 °C. A linear increase of the lattice
parameters as a function of reaction temperature was mea-
sured for both UN2 and R-U2N3 using room temperature
XRD. This indicates a continuous removal of nitrogen from
the UNx system, lowering the N/U molar ratio per the
correlation reported in Figure 6. Average electron densities
calculated on UN2 and U2N3 also showed a decrease
of nitrogen in the samples with respect to the increase in
temperature confirming the above observation. Heating both
UN2 and U2N3 under N2-8% H2 for 30 min, even at 1100 °C,
produces only R-U2N3 (Figure 12). Given that UN2 and
R-U2N3 have solid solution behavior, formation of R-U2N3
under N2 at 1100 °C was expected. If any external N2(g)
pressure was introduced during the decomposition of the
higher uranium nitrides, removal of nitrogen from the UNx
is difficult and may require temperatures greater than 1100 °C
as well as heating times above 30 min.
4. Discussion
Decomposition of UN2 under inert atmospheric condi-
tions forms R-U2N3 and UN. Obtained XRD patterns for
UN2 and R-U2N3 showed single-phased characteristics of the
product, with minimal secondary oxide chemical phases. The
UN product, however, showed significant phase impurities,
with secondary phase levels of UO2 approaching 10 wt. %.
Depending on the decomposition time and conditions, R-
U2N3 phases, due to incomplete decomposition, were also
present up to 80 wt %. Given a heating time between 30 and
240 min, three primary temperature stability regions for the
In contrast to the R-U2N3/UN2 samples, UN showed no
distinct pattern in the variation of lattice parameter as a
function of reaction temperature. This is consistent with the
knowledge that UN has a very narrow range of compositions
at temperatures below 1200 °C.21 Also, considerable levels of
uranium oxide impurity (up to about 10 wt. %) had only
a small effect on the lattice parameter of the UN phase.
(28) Berthold, H. J.; Delliehausen, C. Angew. Chem. Int. 1966, 5, 726.
(29) Sasa, Y.; Atoda, T. J. Am. Chem. Soc. 1970, 53, 102.
(30) Counsell, J. F.; Dell, R. M.; Martin, J. F. Trans. Faraday Soc. 1966,
62, 1736.
(31) Anselin, F. J. Nucl. Mater. 1963, 10, 301.
(32) Price, C. E.; Warren, I. H. Inorg. Chem. 1965, 4, 115.
(33) Berthold, H. J.; Hein, H. G. Angew. Chem. Int. Ed. 1969, 8, 891.
(34) Cordfunke, E. H. P. J. Nucl. Mater. 1975, 56, 319.

10642 Inorganic Chemistry, Vol. 48, No. 22, 2009
Silva et al.
Because oxide phases are known to have a very low solubility
in UN22 this consistency supports the conclusion that the
lattice parameter of UN will not vary significantly as a
function of oxide impurity levels across a large composition
range of the UN/UO2 system.
temperature. An intermediate R-U2N3 phase was seen at
temperatures g675 °C under inert atmosphere with a
negligible nitrogen pressure. The UN was formed at
temperatures greater than 975 °C. A continuous removal
of nitrogen was observed in the UN2 decomposition
process to R-U2N3 and to UN with a continuous crystal
structure changes from fcc to bcc and from bcc to fcc.
Electron density maps calculated for the higher uranium
nitrides also supported the change in lattice parameters
and thus the removal of nitrogen from the crystal systems
as a function of temperature. The complete decomposition
of about 50 mg of UNx sample to pure UN could be
completed in less than 30 min at 1100 °C. The lattice
parameter of UN did not vary significantly with changing
reaction temperature or purity of the bulk sample. The
activation energy for the formation of UN via R-U2N3
decomposition under inert atmospheric conditions was
determined to be 423.8 ( 0.3 kJ/mol.
The activation energy for the formation of UN under
argon atmosphere starting from UNx with a composition
of UNx with x ≈ 1.75-1.8 was determined to be 423.8 (
0.3 kJ/mol. A theoretical study reported -504.2kJ/mol as the
amount of stabilizing energy for a nitrogen atom occupying a
lattice site in R-U2N3 with a 1.75 N/U molar ratio.35 This
indicates about 504.2 kJ/mol activation energy requirement
to decompose the material into UN according to the eq 2,
provided the freeing of nitrogen from the lattice and the
creation of a vacancy is the rate-determining step. This result
provides reasonable agreement with the activation energy
calculated for the decomposition of the higher nitrides to UN
(423.8 ( 0.3 kJ/mol) determined in this study as well as the
pseudo-first-order behavior of the reaction kinetics.
Acknowledgment. We thank Dr. Anthony Hechanova
for administrating the UNLV Transmutation Research
Program under the financial support of the U.S. Depart-
ment of Energy (Grant DE-FG07-01AL67358). We are
indebted to Tom O’Dou and Trevor Low for outstanding
laboratory management and radiation safety.
5. Conclusion
The decomposition of UN2 samples progressed slowly
below 675 °C, but showed a rapid rate increase above that
(35) Fujino, T.; Tagawa, H. J. Phys. Chem. Solids 1973, 34, 1611.