INL/RPT-22-69181
Revision 0
Synthesis and
Thermophysical Property
Determination of NaCl-PuCl
3
Salts
September 2022
Toni Karlsson
1
, Manh-Thuong Nguyen
2
, Scott Middlemas
1
, Michael Woods
1
,
Kevin Tolman
1
, Vanda Glezakou
3
, Ryan Johnson
1
, Shawn Reddish
1
, Stephen
Warmann
1
1
Idaho National Laboratory
2
Pacific Northwest National Laboratory
3
Oak Ridge National Laboratory
ii
INL/RPT-22-69181
Revision
0
Synthesis and Thermophysical Property
Determination of NaCl-PuCl
3
Salts
Toni Karlsson, Manh-Thuong Nguyen, Scott Middlemas, Michael Woods, Kevin
Tolman, Vanda Glezakou, Ryan Johnson, Shawn Reddish, Stephen Warmann
September 2022
Idaho National Laboratory
Advanced Reactor Technologies
Pyrochemistry & Molten Salts Systems Group, C420
Idaho Falls, Idaho 83415
htt // t il
Prepared for the
U.S. Department of Energy
Under DOE Idaho Operations Office
Contract DE-AC07-05ID14517
iii
Page intentionally left blank
iv
INL ART Program
Synthesis and Thermophysical Property
Determination of NaCl-PuCl
3
Salts
INL/RPT-22-69181
Revision 0
September 2022
Technical Reviewer: (Confirmation of mathematical accuracy, and correctness of data and
appropriateness of assumptions.)
Guy Fredrickson
Date
Approved by:
Michael E. Davenport
ART Project Manager
Date
Travis R. Mitchell
ART Program Manager
Date
Michelle T. Sharp
INL Quality Assurance
Date
9/13/2022
9/13/2022
9/13/2022
M. Davenport
9/13/2022
v
ABSTRACT
Currently, a knowledge gap exists in the available data and understanding of
thermophysical properties relating to fresh fuel salts, especially those containing
plutonium. This data is necessary for the design and construction of test reactors,
as well as the licensing of future commercial molten-salt reactors. Thermophysical
properties such as melting temperature, salt stability, density, and heat capacity
were determined using synthesized eutectic NaCl-PuCl
3
(36 mol% PuCl
3
) and a
more sodium rich composition containing 25 mol% PuCl
3
. These measurements
document the baseline properties of the salt as a function of temperature for future
experiments on irradiated fuel salt which will provide a holistic perspective on the
change of thermophysical properties during reactor operations. It was determined
that the NaCl-PuCl
3
ingot synthesized for this study contained 63.4 mol% NaCl,
36.3 mol% PuCl
3
and was 99.7% pure. Upon heating the NaCl-PuCl
3
eutectic was
stable at temperatures up to 800°C. The onset of melting occurred at 451°C, and
the enthalpy of fusion was determined to be 140.7 ± 8.4 J/g. Specific heat capacity
measurements showed a slightly decreasing trend with respect to temperature in
the liquid phase ranging from 0.67 to 0.57 J/g·K, with an average value of 0.637 ±
0.03 J/g·K (104 ± 5 J/mol·K) between 500 to 720°C. Three independent trials of
the molten NaCl-PuCl
3
eutectic salt found the density to be ρ(T) = 3.8589 –
9.5342·10
-4
T(°C), validated between 500 to 800°C. In addition to salt synthesis
and experimentally determining thermodynamic properties, ab initio molecular
dynamic (AIMD) simulations were used to calculate density and heat capacity
values.
vi
ACKNOWLEDGEMENTS
The team would like to acknowledge the facilities at the Idaho National Laboratory (INL) for
supporting this research. This work was supported by multiple projects. The salt synthesis work
was supported under CRADA 21CRA25 “Flowing Molten Salt Microloop Testing Using
Actinide Bearing Salts” in collaboration with TerraPower, LLC. Elemental and isotopic analysis
and the majority of the thermophysical property research was supported by the Molten Salt
Reactor Campaign, work package number AT-22IN070502 “Thermochemical and
Thermophysical Property Database Development – INL.” Finally, density measurements were,
in part, supported through the INL Laboratory Directed Research and Development Program
under Department of Energy’s Idaho Operations Office Contract DE-AC07-05ID14517.
vii
Page intentionally left blank
viii
CONTENTS
ABSTRACT .................................................................................................................................................. v
ACKNOWLEDGEMENTS ......................................................................................................................... vi
ACRONYMS ............................................................................................................................................... xi
1. INTRODUCTION............................................................................................................................... 1
2. METHODS ......................................................................................................................................... 2
2.1 Salt Synthesis ........................................................................................................................... 2
2.2 Elemental and Isotopic Analysis .............................................................................................. 4
2.3 Melting Temperature, Salt Stability, and Enthalpy .................................................................. 5
2.4 Heat Capacity ........................................................................................................................... 6
2.5 Density ..................................................................................................................................... 7
2.6 Computation ............................................................................................................................. 8
3. RESULTS ........................................................................................................................................... 9
3.1 Elemental and Isotopic Analysis .............................................................................................. 9
3.2 Salt Stability, Melting Temperature, and Enthalpy ................................................................ 12
3.3 Heat Capacity ......................................................................................................................... 14
3.4 Density ................................................................................................................................... 16
4. CONCLUSION ................................................................................................................................. 17
5. REFERENCES .................................................................................................................................. 19
ix
FIGURES
Figure 1. Plutonium metal. A. metal rods, B. intermediate particles, C. less than 50 mesh powder. ........... 3
Figure 2. Homogenization of chemicals for synthesis of eutectic NaCl-PuCl
3
salt. A. Unmixed
NaCl, NH
4
Cl, and Pu-metal powders; B. Mixed chemicals in glass vial; C. Mixed
material in glassy carbon crucible, reaction vessel. ...................................................................... 4
Figure 3. Experimental density setup showing bottom-loading balance on the stand above a
furnace with quartz lid and thermocouple inserted inside the AFCI glovebox. ........................... 7
Figure 4. NaCl-PuCl
3
eutectic material; A. solid ingot form, B. powder form. ............................................ 9
Figure 5. XRD pattern for eutectic NaCl-PuCl
3
. ......................................................................................... 12
Figure 6. Mass changed curves as a function of time and temperature for the NaCl-UCl
3
ingot: A.
20°C/min, B. 10°C/min, and C. 2°C/min. .................................................................................. 13
Figure 7. Thermograms for three sample each analyzed at a different heating rate. A. heating
curves; B. cooling curves. ........................................................................................................... 14
Figure 8. Summary of heat capacity. A. DSC curves for sapphire standard comparing
experimentally determined and NIST reported sapphire heat capacity. B. Experimental
heat capacity values for three NaCl-PuCl
3
eutectic samples. C. Averaged heat capacity
valued with error bars of one standard deviation derived from experimental Cp
measurements along with IMD values calculated for two compositions. .................................. 15
Figure 9. Experimental and calculated density values for 36 mol% and 25 mol% PuCl
3
in NaCl
carrier salt. .................................................................................................................................. 17
TABLES
Table 1. Properties of NaCl and PuCl
3
in their crystalline form. .................................................................. 2
Table 2. Experimental STA results of zinc run using the temperature and heat flow calibration
files at multiple heating rates. ....................................................................................................... 6
Table 3. Definition of terms used to calculate the error associated with liquid density
measurements. .............................................................................................................................. 8
Table 4. Compositions and temperatures used in simulations. ..................................................................... 9
Table 5. Elemental and isotopic composition of the Pu-metal starting and the synthesized NaCl-
PuCl
3
ingot. ................................................................................................................................. 10
Table 6. Composition of the NaCl-PuCl
3
ingot by conversion to chlorides assumed NaCl and
formation of Pu, U, Np, and Am to their respective trichloride. ................................................ 11
x
Page intentionally left blank
xi
ACRONYMS
AFCI Advanced Fuel Cycle Initiative
AIMD Ab Initio Molecular Dynamics
ASTM American Standard Test Method
CMD Classical force field molecular dynamics
DSC Differential Scanning Calorimeter
ICDD International Centre for Diffraction Database
ICP OES Inductively Coupled Plasma Optical Emission Spectroscopy
INL Idaho National Laboratory
MSR Molten-Salt Reactors
NIST National Institute of Standards and Technology
NPT number of particles, pressure, and temperature are conserved
NVT number of particles, volume, and temperature are conserved
Q-ICP-MS Quadrupole - Inductively Coupled Plasma - Mass Spectrometry
STA Simultaneous Thermal Analyzer
TGA Thermogravimetric Analyzer
UTEVA Uranium and Tetra Valent Actinides
XRD X-ray Diffraction
xii
Page intentionally left blank
1
1. INTRODUCTION
The use of sodium chloride (NaCl) - plutonium trichloride (PuCl
3
) as a fuel in molten-salt reactors
(MSR) has been discussed for decades [1-3]. Today, society’s rapid shift toward renewable energies and
push for carbon neutral energy sources to mitigate climate change have renewed interest in using nuclear
energy as a low-carbon footprint energy source [4]. Advanced MSRs are one potential type of nuclear
reactor design being considered today by several countries including the United States, Canada, China,
South Korea, Norway, and Denmark. Fuel compositions of actinide fluoride and chlorides are typically
considered; however, a recent study on heavier halides indicates that bromide and iodine fuels may also
be possible [5]. A limited amount of experimental thermophysical property data is available on the NaCl-
PuCl
3
fuel salt system; furthermore, most information available relates to the unary actinide salts or
mixtures which have been theoretically calculated and remain unverified experimentally.
Some of the challenges that arise when studying systems containing PuCl
3
are the availability of a
pure PuCl
3
feedstock and facilities with characterization equipment that allows the use of transuranic
materials. Synthesis of high-purity PuCl
3
has been demonstrated, at the gram scale, by aqueous, gaseous,
and electrochemical methods [6-8]. There are several documented chemical reactions for producing
PuCl
3
, each using a different precursor. PuCl
3
can be synthesized from the reaction of PuO
2
or plutonium
(III) oxalate with PCl
5
, SCl
2
, HCl-H
2
, COCl
2
, and CCl
4
. However, these are either not scalable reactions
or do not yield a PuCl
3
product >98% pure [6] except with COCl
2
[9]. Aqueous preparation of PuCl
3
can
be complicated by difficulties associated with obtaining a pure solution of Pu(III) and removing all water
of crystallization; however, 98% pure PuCl
3
has been achieved [10]. Therefore, synthesis and purification
of PuCl
3
are typically achieved through high-temperature pyrochemical processing where the PuO
2
feedstock is converted to Pu metal by direct reduction in CaCl
2
through the multicycle direct oxide
reduction process [11]. However, if complete chlorination of Pu is not ensured, it can have unintended
effects on thermophysical properties, especially melting temperature [12].
A lesser utilized synthesis route is proposed in this research, where NH
4
Cl is used as the chlorinating
agent at elevated temperatures and has been shown to yield a high-purity UCl
3
product [13]. The overall
reaction is represented in Reaction 1. NH
4
Cl sublimates at 338°C [14] represented by Reaction 2, and its
decomposition is therefore the source of HCl gas. Reaction of Pu metal with HCl gas is highly
exothermic, Reaction 3, and because this route of chlorination is a solid-gas reaction, the Pu metal must
be pulverized to increase the surface area and ensure complete chlorination. Extreme care should be taken
when handling Pu-metal powders which are pyrophoric and must be handled in an inert atmosphere.
Because the Pu-Cl reaction is highly exothermic, an inert carrier salt, such as NaCl, can be added to the
reaction vessel to absorb some of the energy produced during the chlorination. The complete synthesis
reaction follows Reaction 4, where the intended product was the eutectic NaCl-PuCl
3
binary containing
36 mol% PuCl
3
reported by Bjorklund et al. [15], and NaCl is present as both a product and reactant.
Gibbs free energy calculations were performed by HSC Software [16].
+ 3
+ 3
+ 1.5
,
= 505
Reaction 1
()
() + ()
,
= 505 Reaction 2
() + 3
(
)
() + 1.5
()
,
= 527 Reaction 3
+
+
+ +
+
Reaction 4
2
One issue that currently inhibits development and deploying advanced MSRs is a lack of
understanding of molten-salt behavior and molten-salt thermophysical property changes due to impurities,
such as those arising from salt fabrication, handling (i.e., oxygen and moisture), and corrosion. Therefore,
it is necessary to understand the fundamental behaviors such as melting and polymorphic temperatures,
heat capacity, and density of actinide-bearing salts proposed for use as fuels in MSRs. This type of data
along with other information such as corrosion kinetics will aid in modeling and simulation.
Thermophysical properties of NaCl are well known. However, limited data is present in literature on
the unary PuCl
3
salt and even less information is available for mixtures of NaCl and PuCl
3
.
Thermophysical properties for NaCl and PuCl
3
are provided in Table 1. In this work, eutectic NaCl-PuCl
3
(64 mol% NaCl – 36 mol% PuCl3) was synthesized along with 75 mol% NaCl - 25 mol% PuCl
3
salt.
Both were characterized to provide a better understanding of NaCl-PuCl
3
mixtures melting temperature,
salt stability, density, and heat capacity in the temperature range or an operating MSR. In addition,
experimental data for density and heat capacity are compared to computational values determined by ab
initio molecular dynamic (AIMD) simulations to provide further insights on the similarities and contrasts
existing in both approaches.
Table 1. Properties of NaCl and PuCl
3
in their crystalline form.
Property
NaCl (cr)
Ref.
PuCl
3
(cr)
Ref.
Appearance Clear to
white
[17] Green-blue [18]
Melting Temperature, °C
801 [17] 736 ± 10
760
767 ± 2
760 to 765
768
[19]
[10]
[15]
[20]
[21]
Boiling Temperature, °C
1413 [17] 1679
1730
[19]
[5]
Density (solid),
g/cm
3
2.165 [17] 5.708 [22]
Specific Heat (J/mol·K)
50.33
[23]
101.2
[5, 22]
Heat of Fusion, kJ/mol 517 [17] 55.0 ± 5.0
63.6
[24]
[25]
2. METHODS
All material handling and experimentation were performed in an inert atmosphere glovebox with
oxygen and moisture levels maintained below 5 and 0.1 ppm, respectively. Physical measurements of all
masses were measured using a calibrated analytical balance. Balances are calibrated annually by the
standards and calibration laboratory (S&CL) at the Idaho National Laboratory (INL), and they are
assigned an uncertainty for their measurement range. Daily checks are performed on these balances using
National Institute of Standards and Technology (NIST) traceable external calibration weights, also
calibrated by the S&CL, to confirm the balance functionality and accuracy before use.
2.1 Salt Synthesis
Prior to use, sodium chloride (NaCl, 99.99% metals basis) was heated under a 17.5 kPa absolute
vacuum to 350°C for 3.5 hours in an argon glovebox to ensure it was fully dry before being transferred
and weighed in the Advanced Fuel Cycle Initiative (AFCI) Glovebox at INL. Ammonium chloride
3
(NH
4
Cl, 99.99%) was heated to 100°C and held for 8 hours. The particle size of NaCl and NH
4
Cl was
reduced using an agate mortar and pestle to less than 50 mesh. Plutonium metal, in rodlet form, was
selected as the feedstock material because of its purity and ease of size reduction. Visually, the Pu-metal
rod appeared to be free from surface oxidation (Figure 1A). All labware in contact with materials during
synthesis and experimentation was cleaned with soapy water, rinsed in deionized water, dried with
ethanol, rinsed with deionized water, dried in an oven, and then heated to temperature prior to use in a
furnace.
Prior to hydriding the Pu-metal rods (Figure 1A), further size reduction was performed using sheers
to cut segments into small pellets approximately 3–5 mm in diameter as seen in Figure 1B. To increase
the surface available for reaction, the Pu-metal pellets were exposed to hydrogen gas at elevated
temperatures. Exposure of Pu metal to hydrogen gas leads to the formation of PuH
x
where the
stoichiometry is dependent on temperature and pressure [26]. The reaction for hydriding follows Reaction
5; this reaction will proceed at a rapid rate, independent of temperature but dependent on the hydrogen
partial pressure and active surface area [27].
+
Reaction 5
The furnace hydriding furnace was prepared by placing the Pu-metal sample in a tapered quartz
crucible (4.5 cm inner diameter at base), evacuating the furnace chamber to approximately 3.07 kPa, and
backfilling with a flowing hydrogen purge gas (H
2
, 99.99%) at a rate of 20 standard cubic centimeter per
minute. The furnace was then heated to 250°C and held for 60 minutes with the H
2
purge. After the 60
minutes, the furnace was cooled to room temperature, evacuated, and reheated to 250°C with a H
2
purge.
This heating, cooling, and evacuation cycle was repeated five times in an attempt to dislodge the hydride
material from the Pu-metal pellets, exposing more metal surface for reaction with H
2
gas. The material
was removed from the furnace and sieved. Material <50 mesh was held to the side while material >50
mesh went through the hydride process until all material was below 50 mesh (Figure 1C). To de-hydride,
a final step was performed where the material was heated to 325°C under vacuum and held for several
hours. No hydrogen analysis was performed on the material to confirm completeness as the presence of
some PuH
2
is not detrimental to the chlorination step due to the high favorable thermodynamic
chlorination reaction.
Figure 1. Plutonium metal. A. metal rods, B. intermediate particles, C. less than 50 mesh powder.
The Pu-powder was chlorinated using NH
4
Cl which is an exothermic reaction; to absorb some of the
energy, sodium chloride was added to the mixture as a “heat sink.” In addition, the eutectic NaCl-PuCl
3
was the desired product so adding NaCl to the reaction container eliminated a further blending step.
Two batches of Pu metal were size reduced using the hydride, de-hydride process, approximately
150.0g each. Two batches of NaCl-PuCl
3
eutectic salt were synthesized, approximately 250g in mass
each. First, the NaCl and NH
4
Cl were mixed, with NH
4
Cl added in 2% excess. Excess chlorine gas
causing the formation of PuCl
4
was not of concern as PuCl
4
has been shown, by calculations, to not exist
4
in the solid state but to exist in the gaseous state in an atmosphere of chlorine over PuCl
3
[28]. Though
PuCl
4
has not been isolated in the solid or gaseous state, its presence has been inferred due to increased
volatility of PuCl
3
in a stream of chlorine at 400°C relative to H-HCl stream. However, it is thought that
gaseous PuCl
4
decomposes back to PuCl
3
and chlorine gas when condensed. This is in contrast to UCl
3
which has been shown to form volatile UCl
4
when in the presence of excess chlorine [29].
Pu-metal powder was added to the NaCl-NH
4
Cl mixture in a glass jar (Figure 2A) and then lightly
mixed into a homogenous blend as seen in Figure 2B. Finally, the homogenous mixture was poured into a
glassy carbon crucible (Sigradur, GAT 32, 320 ml) shown in Figure 2C and placed into a modified
benchtop furnace (Kerr, Auto Electro-Melt Furnace, Maxi 3 kg). A glassy carbon cover (Sidradur GAD 3)
was placed on the crucible. Steel wool was placed on top of the crucible lid to react with any chlorine gas
should it escape the reaction vessel. This was all contained within the furnace which included an insulated
lid. The synthesis reaction vessel and furnace were located within the argon atmosphere AFCI glovebox.
During synthesis, ammonia and HCl gas detectors (Dräger diffusion tubes) were placed on top of the
furnace and at various location in the glovebox.
Figure 2. Homogenization of chemicals for synthesis of eutectic NaCl-PuCl
3
salt. A. Unmixed NaCl,
NH
4
Cl, and Pu-metal powders; B. Mixed chemicals in glass vial; C. Mixed material in glassy carbon
crucible, reaction vessel.
Because the Pu-chlorination reaction is exothermic and has the potential to instantaneously release a
large amount of energy, the synthesis reaction was performed over approximately a 60-hour period. Time
zero began at room temperature, and the furnace was heated to 450°C with a heating rate of 10°C/hour.
The critical temperature for the reaction was thought to be 338°C, the sublimation temperature of NH
4
Cl.
However, NH
3
gas was detected when the furnace reached 175°C, suggesting the chlorination reaction
initiated at lower temperatures than expected. Once the furnace reached 450°C, the material was heated at
5°C/min to 800°C and then cooled to room temperature. Once cool, the NaCl-PuCl
3
ingot was easily
removed from the glassy carbon crucible. No discoloration was seen on the steel wool, the crucible lid, or
the walls of the crucible that would indicate volatility of the NaCl-PuCl
3
product.
2.2 Elemental and Isotopic Analysis
Both the Pu-metal powder and the resulting NaCl-PuCl
3
salt samples were dissolved in a solution
containing several strong inorganic acids to ensure the sample and any metal impurities would be
dissolved. Before trace analysis of impurities could be carried out by inductively coupled optical
emissions spectrometry (ICP OES), a portion of the parent solution was chemically separated to remove
the plutonium from the matrix. Plutonium has a rich emission spectrum which can interfere with
5
determining other analytes of interest, particularly if they are present in only trace quantities. The
separations were carried out using auto-gas-pressurized extraction chromatography (auto-GPEC). Once
the sample is loaded into the auto-GPEC, a valve switches/rotates, opening to a pressurized-gas line, and
the sample is then pushed through the chromatographic column, and the eluent is collected on the other
side. To remove the plutonium, a uranium and tetra valent actinides (UTEVA, Eichrom) 100–150 µm
resin was employed as the stationary phase. The removal of the plutonium is required before the ICP-OES
analysis because emission from plutonium can interfere with determining other elements of interest,
specifically iron. Isotopic analysis was performed on the quadrupole – inductively coupled plasma - mass
spectrometer (Q-ICP-MS) which does not require separation prior to analysis, and therefore, a portion of
the parent solution was simply diluted to prepare it for analysis.
Crystallographic X-ray diffraction (XRD) data were acquired on a Panalytical Empyrean X-ray
diffractometer (Malvern-Panalytical Ltd United Kingdom) operating with a Bragg-Brentano geometry
monochromatic Cu-K
α
beam radiation (λ=1.540593 Å). The salt sample(s) were analyzed by adjusting the
voltage and current values to 40-kV and 45-mA, respectively, and rotating the sample(s) on the
diffractometer reflection sample holder spinner at a rotational speed of 15 rpm. Scanning was conducted
from 10° to 70° 2θ with a 0.026-degree step size interval. A NIST LaB6 standard (SRM 660c) was used
to determine instrument parameters and verify instrument alignment. The background was fitted with a
Sonneveld & Visser polynomial [30] with a zero bending factor and eight granularity to fit the
background. Rietveld fits to the data were conducted using a computer software program (HighScore
Plus; Malvern-Panalytical) and using International Centre for Diffraction Database (ICDD) PDF4+.
Refinements for plutonium trichloride were conducted in hexagonal space group P63/m (No.176) using
ICDD reference 00-006-0222 and for sodium chloride in cubic space group Fm-3m (No.225) using ICDD
reference 00-002-818.
2.3 Melting Temperature, Salt Stability, and Enthalpy
Transition temperatures and enthalpies of transition were determined employing a simultaneous
thermal analyzer (STA), Netzsch 449 F1, with differential scanning calorimeter (DSC)/thermogravimetric
analyzer (TGA) type-S sample carrier, located inside an inert argon glovebox having oxygen and
moisture levels below 5 ppm and 0.1 ppm, respectively. Inert, glassy carbon crucibles (diameter 6 mm, 50
μl) and lids, rated to 2400°C, were used to contain the salt samples and calibration standards. Crucible
lids contained a small hole in the center to prevent a buildup of pressure in the crucible and allow the
escape of off gasses (if they were to occur).
Calibration of the STA for temperature and heat flow was performed using five high-purity standards
for each of the three heating rates: 20, 10, 2°C/min. The transition temperatures and enthalpy of the
standards were calculated by averaging the onset and peak area for three heating cycles [31]. Enthalpies
of transition were determined by integrating the full area under each transition peak from the transition
start to the transition finish. The temperature and sensitivity calibrations were accurate in the range of 164
to 962°C by verification using a zinc standard (Table 2) where the experimental melting temperature was
within ±3°C and < ±1°C with a heating rate of 20 and 2°C/min, respectively. The enthalpy of zinc at
different heating rates was accurate to ±3%.
6
Table 2. Experimental STA results of zinc run using the temperature and heat flow calibration files at
multiple heating rates.
Heating
Rate
Cycle 1 Cycle 2 Cycle 3 Average Deviation Error
°C/min
°C
°C
°C
°C
°C
%
20
417.0
416.9
416.9
416.9
2.6
-0.61
10
418.4
418.2
418.2
418.3
1.2
-0.29
2
419.1
419.0
419.0
419.0
0.5
-0.11
°C/min
J/g
J/g
J/g
J/g
J/g
%
20
105.5
105.4
104.9
105.3
2.2
-2.08
10
108.3
108.0
107.8
108.0
-0.5
0.50
2
104.8
104.4
103.9
104.4
3.1
-2.91
The reported eutectic and melting point temperatures were determined using the average of the onset
temperature or ending peak temperature for melting, derived from the heating segment of three separate
heating and cooling cycles. The reported onset and melting temperatures were determined using linear
regression to remove the effects of thermal lag from the samples. All temperatures are reported with an
accuracy of < ± 5°C [32].
The stability of the salts was studied by monitoring the mass change as a function of temperature. A
correction file was generated, prior to sample analysis, using two empty crucibles (i.e., sample and
reference) with identical heating conditions under which the sample was run to account for artificial
changes in mass due to buoyancy effects. Once the correction was generated, the sample was loaded into
the same sample crucible used in the correction file.
2.4 Heat Capacity
Specific heat capacity measurements were conducted using a Netzsch 404 F1 DSC with a type-S
sample carrier and rhodium furnace. Prior to heat capacity measurements, the DSC was calibrated for
temperature and heat flow as described previously for the STA. A heating rate of 5°C/min was used for
all calibration standards and during specific heat capacity measurements. The accuracy of the temperature
and sensitivity calibration was verified using one or more standards and performed according to ASTM
E967-18 [31]. Verification of the calibration curves showed that the DSC was calibrated to have deviation
in experimental melting temperature from theoretical values of < ±1°C, and the heat flow was calibrated
to be < 2% for aluminum.
Heat capacity was determined using the C
p
ratio method where the measurement consisted of several
experimental runs including baseline (empty crucible, no sample), standard + baseline (sapphire with same
crucibles as baseline), and sample + baseline (same crucible as baseline and standard). When determining
heat capacity, a new sensitivity calibration file may be generated from the sapphire standard. However, the
sensitivity calibration generated using the high-purity standards may also be used depending on which one
provides the most accurate heat capacity for sapphire within the temperature range of interest. Specific heat
as a function of temperature at constant pressure was calculated by Equation 1 and requires three
measurements be taken under the same experimental conditions.
=
Equation 1
7
Measurement of heat capacity involved nine separate measurements using the same sample and
reference crucible. First, three baseline correction measurements were performed. The first was discarded,
and the second and third baseline correction were used to verify reproducibility. Next, three
measurements were run using a NIST sapphire standard with the selected baseline. The first sapphire run
was used to generate a sapphire heat flow calibration, replacing the heat flow calibration generated using
the standards. The second and third sapphire runs were used to verify that the measured specific heat
capacity of sapphire matched the values for theoretical sapphire heat capacity. Finally, the salt sample
was loaded into the crucible and run three times using the same baseline as the sapphire sample. The
average heat capacity of the three sample measurements is reported as the heat capacity for that sample.
Specific heat capacity determination was performed in accordance with American Standard Test Method
(ASTM) standard procedure [33].
2.5 Density
For the NaCl-PuCl
3
eutectic density determination, 35.250g of salt was added to a glassy carbon
crucible previously cleaned with deionized water and isopropyl alcohol and baked in a furnace at 800°C
for 2 hours. The glassy carbon crucible containing the salt sample was then placed in a benchtop furnace
(Kerr, Auto Electro-Melt Furnace, Maxi 3 kg) modified for density experiments. Figure 3 shows the
Archimedes densitometer setup within the glovebox including the furnace, stand, and bottom hanging
balance and is described in greater detail by Duemmler et al. [34]. A Ni bobber was tied onto the wire and
its volume calculated in 10 mL of both deionized water and ethanol, which have well defined densities
near room temperature. The average of these calculations gave a volume of 1.0408 ± 0.0006 cm
3
. The
mass of the bobber and wire was then measured in argon using the hang down balance (Mettler Toledo
WXSS204, tolerance 0.8 mg). The crucible with salts, quartz lid, and bobber with wire were added to the
setup and the salt melted. Once eutectic NaCl-PuCl
3
measurements were completed, an addition of 5.60g
of NaCl was added to eutectic salt in the crucible to achieve a 25mol% PuCl
3
in NaCl composition and
the density of this mixture was determined.
Figure 3. Experimental density setup showing bottom-loading balance on the stand above a furnace with
quartz lid and thermocouple inserted inside the AFCI glovebox.
8
Mass readings were performed after 5 minutes of thermal stability at each temperature as measured
by an Inconel-sheathed K-type thermocouple (OMEGA) inserted directly into the salts. This equilibration
was on average 60 minutes after each temperature change of approximately 50°C. An internal adjustment
of the balance was performed before each set of mass measurements at a unique temperature. The balance
was tared before each measurement. Density measurements were first taken at increasing temperature
intervals between 525 and 800°C and then with decreasing temperatures between 800 to 525°C and
finally a randomized temperature schedule.
The density calculations were performed using the direct Archimedean method based on
measurement of buoyancy force exerted on a bobber submerged in molten salts. The density of the salt at
an experimental temperature is calculated by Equation 2 where
is the measured mass of the
bobber and wire suspended in argon,
is the measured weight of the bobber and wire suspended in
the salts, g is the acceleration due to the gravity of Earth, α is the linear thermal coefficient of expansion
of nickel, T is the salt temperature, and
is the reference temperature for
, the reference volume of the
nickel bobber. The linear thermal coefficient of expansion of nickel was calculated based on a polynomial
fit of reference data [35].
=
(
(
)
)
Equation 2
The experimental uncertainty of the density was calculated by the propagation of individual
uncertainties. Equation 3 shows the fundamental form of the uncertainty propagation of Equation 2.
Equation 4 shows the simplified experimental uncertainty function, and Table 3 lists the individual
uncertainties and their explanations.
=
+
+
+
+
Equation 3
=
+
+
(
)
(
)
+
(
)
Equation 4
Table 3. Definition of terms used to calculate the error associated with liquid density measurements.
Symbol
Explanation
Standard deviation of the five weight
measurements in argon
Standard deviation of the five weight
measurements in salt
Standard deviation of the two calculated
volumes from water and ethanol benchtop
trial
Assigned as 1% of value of
OMEGA assigned 0.05% * T + 0.3°C
2.6 Computation
Since research on transuranic salts is sparce in literature, timely, expensive, and hazardous, AIMD
simulations were used to investigate two NaCl-PuCl
3
salt compositions at different temperatures; a
9
summary of compositions is provided in Table 4. Comparing experimental and calculated, density and
heat capacity results, will aid in refining the modeled data and yield more accurate models.
All simulations were conducted using the CP2K package [36, 37]. Classical force field molecular
dynamics (CMD) were used initially to relax the system and provide starting configurations for the AIMD
simulations. For CMD, the polarized ionic model (PIM) [38] was used to represent interatomic
interactions. The density of the system was optimized within the “number of particles, pressure, and
temperature are conserved” (NPT) ensemble, where the pressure and temperature were maintained with
the Parrinello-Raman barostats [39] and Nose-Hoover thermostat [40], respectively.
The AIMD simulations were performed with the Perdew-Burke-Ernzerhof density functional [41]
including Grimme’s D3 van der Waals corrections [42] with the Goedecker-Teter-Hutter pseudopotentials
[43, 44]. The double zeta valence polarized [44-46] Gaussian basis functions were used for the valance
states of Pu(6s
2
6p
6
5f
6
7s
2
), Na(2s
2
2p
6
3s
1
), and Cl(3s
2
3p
5
) with a plane wave cutoff of 900 Ry in
conjunction with the Gaussian plane wave hybrid basis set framework [47]. Molecular dynamics
simulations were conducted within both the NPT and “number of particles, volume, and temperature are
conserved” (NVT) ensembles. In NPT simulations, the pressure and temperature were maintained with
the Parrinello-Raman barostats [39] and Nose-Hoover thermostat [40], respectively.
Table 4. Compositions and temperatures used in simulations.
16 mol% PuCl
3
25 mol% PuCl
3
36 mol% PuCl
3
Number of Cl atoms
66
78
86
Number of Pu atoms
8
13
18
Number of Na atoms
42
39
32
Temperatures (K)
1000, 1257
898, 973, 1073
730, 900, 1100
3. RESULTS
Two separate salt syntheses were performed. Both yielding an approximate 250g NaCl-PuCl
3
ingot
close to the eutectic composition of 64 mol% NaCl – 36 mol% PuCl
3
. Each ingot was easily released
from the glassy carbon crucible and appeared to be black with a teal hue in color (Figure 4A). After size
reduction, the powdered material was blue green in color (Figure 4B)—closely matching the color
reported for PuCl
3
[18].
Figure 4. NaCl-PuCl
3
eutectic material; A. solid ingot form, B. powder form.
3.1 Elemental and Isotopic Analysis
Elemental and isotopic results from Q-ICP-MS, ICP-OES, and gamma spectroscopy for the Pu-metal
and synthesized NaCl-PuCl
3
ingot are provided in Table 5. Elements which were below the detection limit
10
of each instrument are not listed in Table 5. It can be concluded from the elemental and isotopic analysis
of the salt that there are no major impurities in the Pu-metal with the exception of a small amount of iron,
as well as uranium, and americium present as decay products.
Table 5. Elemental and isotopic composition of the Pu-metal starting and the synthesized NaCl-PuCl
3
ingot.
Q-ICP-MS
Pu-metal
NaCl-PuCl
3
salt
Analyte
ug/g
% Error
ug/g
% Error
U-234
118
±10%
60.4
±10%
U-235
792
±5%
406
±5%
U-236
556
±5%
283
±5%
m/z -238
430
±5%
246
±5%
Np-237
390
±5%
212
±5%
Pu-239
807000
±5%
469000
±5%
Pu-240
157000
±5%
91600
±5%
m/z-241
10100
±5%
6040
±5%
m/z-242
3350
±5%
1960
±5%
Sr-88
<40
N/A
8.06
±15%
Ag-109
10.7
±20%
11.7
±20%
Ba-138
<30
N/A
11.7
±15%
ICP-OES
Analyte
ug/g
% Error
ug/g
% Error
Fe
91.0
± 25 %
72.7
± 30 %
Na
<320
N/A
94900
±5%
Pr
<30
N/A
83.8
± 20 %
Gamma Spectroscopy
Species
uCi/g
% Error
uCi/g
% Error
Am-241*
25100
±3%
14100
±3%
* The specific activity of Am-241 is 3.428 Ci/g [48].
Analysis results of the ICP-OES, Q-ICP-MS and gamma spectroscopy data are provided in Table 6.
Mass balance suggests there was a 106% recovery of the sample but within the expected error. The
quantities of Am-241 and Pu-241 were determined by subtracting the Am-241 concentration (gamma
spectroscopy) from the total m/x-241 value (Q-ICP-MS). NaCl and PuCl
3
values indicate that the fabricated
NaCl-PuCl
3
ingot contains 63.4 mol% NaCl and 36.3 mol% PuCl
3
. The experimentally reported eutectic
composition contains 36 mol% PuCl
3
[15]; therefore, the salt fabricated for this study is very nearly the
reported eutectic composition. Other studies calculate the eutectic composition of PuCl
3
to be 38.3 mol%
[49] or 37.4 mol% [50]. Impurities and compounds other than NaCl and PuCl
3
account for < 0.32 mol% of
the ingot. The concentrations of Sr-88, Ag-109, Ba-137, Fe, and Pr are not accounted for since they are <
0.03% of the total sample. It was therefore determined by elemental and isotopic evaluation that the NaCl-
PuCl
3
ingot material synthesized had a purity of 99.7%.
11
Table 6. Composition of the NaCl-PuCl
3
ingot by conversion to chlorides assumed NaCl and formation of
Pu, U, Np, and Am to their respective trichloride.
Element Metal Metal
Chloride
wt. % mol %
g/g
g/g
Na
0.0949
0.2412
22.66
63.41
Total U
0.0007
0.0011
0.10
0.05
Total Pu
0.5647
0.8159
76.65
36.27
Np
0.0002
0.0003
0.03
0.01
Am
0.0041
0.0059
0.56
0.26
Total
0.6647
1.0644
100.00
100.00
A XRD pattern of 84 mg NaCl-PuCl
3
powder was analyzed for phase purity. This sample was
hygroscopic and needed an inert atmosphere during XRD. The sample was loaded into a gas-tight-dome
holder to keep argon gas atmosphere around this sample during the data collection inside the
diffractometer. The black tick marks in Figure 5 represents the combined reported lines used to fit this
pattern for PuCl
3
hexagonal space group P63/m (No.176) using ICDD reference 00-006-0222 and NaCl
cubic space group Fm-3m (No.225) using ICDD reference 00-002-818. The blue curve is the collected
diffraction data for this sample with identified peaks for PuCl
3
and NaCl. These peaks are identified with
their respective phases and indicated by green and red dots above the identified peaks. The XRD holder is
equipped with a polymer dome on this gas-tight sample holder and resulted in amorphous x-ray scatter
and can be seen in the XRD data as the hump between 10 to 25° 2-theta. There were some additional
peaks that could be seen in the data and are likely from a monoclinic phase, which could indicate the
presence of an oxy-chloride phase. Several XRD patterns were collected over a 2-day period, as time
passed peaks 15 and 16°, 2-theta became more visible suggesting that air was contaminating the sample
over time. The peak shown in Figure 5 was collected less than 1 hour after removing the sample from the
glovebox.
The lattice dimensions, for PuCl3 determined by x-ray diffraction are a
1
= 7.380 ± 0.001 Å and a
3
=
4.238 ± 0.001 Å [51]. The refined structure for eutectic NaCl-PuCl
3
showed lattice dimensions of a
1
=
7.506Å and a
3
= 4.237Å. The lattice constants obtained in this work show that the eutectic NaCl-PuCl
3
composition has an increase lattice constant for a
1
of 0.126Å, while the a
3
remained unchanged from
PuCl
3
reported values in literature. Similar behavior was exhibited in the eutectic NaCl-UCl
3
system when
compared to UCl
3
[13, 52].
12
10 20 30 40 50 60 70 80 90
0.0
0.2
0.4
0.6
0.8
1.0
Height (A.U.)
Position (2θ)
NaCl-PuCl
3
XRD pattern
NaCl-PuCl
3
(fit)
PuCl
3
NaCl
Figure 5. XRD pattern for eutectic NaCl-PuCl
3
.
3.2 Salt Stability, Melting Temperature, and Enthalpy
Prior to melting temperature and other property measurements the stability of the salt at elevated
temperatures was investigated on the STA by monitoring the mass change as a function of temperature.
Each heating rate used a new sample. Results for mass change as a function of heating rate and time are
provided in Figure 6. The sample with the fastest heating rate was exposed to the highest temperature. No
clear trend with heating and cooling was observed for the sample as is evident by the lack of “stairstep” in
the mass change curves. The overall sample mass change was observed to be < ± 0.25% for all samples.
13
0 50 100 150 200 250 300 350
99.0
99.5
100.0
100.5
101.0
Mass
Temperature
Time (min)
Mass (%)
A.
0
100
200
300
400
500
600
700
800
Temperature (°C)
0 100 200 300 400
99.0
99.5
100.0
100.5
101.0
Mass Temperature
Time (min)
Mass (%)
B.
300
400
500
600
700
Temperature (°C)
0 100 200 300 400 500 600 700
99.0
99.5
100.0
100.5
101.0
Mass Temperature
Time (min)
Mass (%)
C.
350
400
450
500
550
Temperature (°C)
Figure 6. Mass changed curves as a function of time and temperature for the NaCl-PuCl
3
ingot: A.
20°C/min, B. 10°C/min, and C. 2°C/min.
The melting point was investigated using three separate samples, each experiencing a different heating
rate: 20, 10, and 2°C/min. Each sample was subjected to four heating and cooling cycles, the first was
discarded while inflections from the other three were averaged and used to report temperature dependent
transitions. The second heating thermogram for each heating rate is shown in Figure 7A during heating and
Figure 7B for cooling. Upon heating, only one peak was observed, attributed to the solid to liquid transition
at the eutectic temperature. The eutectic onset temperature is 451°C, while the peak and end temperature
are 455 and 456°C, determined using linear regression from the average values determined at each heating
rate. The eutectic temperature, 451°C, determined in this work closely matches other reported eutectic
temperatures of 453°C [15] and 450°C [50].
As labeled in Figure 7B, there are two distinct peaks upon cooling where peak 1 occurs at various
temperatures and is attributed to recrystallization of the eutectic NaCl-PuCl
3
mixture. Peak 2 occurs at
approximately 455°C, and while the cause of this peak is unclear, it could be due to cooling kinetics within
the sample crucible. Referring to the elemental and isotopic analysis of the ingot material, no metal
impurities were detected, and the concentration of PuCl
3
was determined to be 36.6 mol%. Analysis of the
2°C/min heating peak shows the eutectic transition peak has a width of 10°C, occurring between 450–
460°C. Although there is only one peak shown experimentally to occur in this temperature range, if the salt
differs very slightly from the eutectic composition, then a small peak may be hidden within the larger
eutectic transition peak. Referencing the NaCl-PuCl
3
phase diagram developed by Bjorklund et al. [15]
14
indicates that a 10°C increase in temperature at most could be the result of a salt having a PuCl
3
composition
between 35.47–37.52 mol%. Although it is not recommended to use temperatures from cooling curves
when analyzing molten salts, due to supercooling effects, a solidus temperature at 455°C would indicate a
PuCl
3
composition of 36.8 mol%, which is 0.5 mol% higher than analytical results indicate. It can therefore
be concluded from analytical results as well as analysis of the solidus and liquidus temperature that the
NaCl-PuCl
3
used in this study has a PuCl
3
concentration between 36.3 to 36.8 mol%.
100 200 300 400 500 600 700 800
0.0
0.5
1.0
1.5
2.0
2.5
DSC (mW/mg)
Temperature (°C)
20°C/min
10°C/min
2°C/min
Exo.
A.
100 200 300 400 500 600 700 800 900
-3.5
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
DSC (mW/mg)
Temperature (°C)
20°C/min
10°C/min
2°C/min
Exo.
Peak 1
Peak 2
B.
Figure 7. Thermograms for three sample each analyzed at a different heating rate. A. heating curves; B.
cooling curves.
T
he enthalpy of fusion,
, for the NaCl-PuCl
3
ingot was determined by integrating the area under
the eutectic peak transition, where
ranged from 134.8 to 148.4 J/g with an average for all three heating
rates of 140.7 ± 8.4 J/g. The enthalpy of crystallization,
, was determined to be -130.9 ± 11.4 J/g.
Th
e eutectic and liquidus temperature of 25 mol% PuCl
3
in NaCl was determined to be 450 and 599°C,
respectively. These values were determined following the experimental approach described for the
eutectic composition, where three heating rates were used to determine the onset eutectic temperature and
liquidus temperature. The enthalpy of fusion for the 25 mol% PuCl
3
in NaCl was determined to be 171.4
± 11.4 J/g.
3.3 Heat Capacity
Three individual liquid state heat capacity measurements were performed using a new sample for each
group of measurements, and in total, three samples were analyzed. Before heat capacity measurements, heat
flow calibrations were verified using the sapphire calibration standard on each sample. If the experimentally
determined sapphire heat capacity deviated from standard values [53] by > 5%, the sapphire heat capacity
was deemed inadequate, and it was performed again. Heat capacity was determined following standard test
methods [33]. An example of sapphire heat capacity measurement is shown in Figure 8A.
For NaCl-PuCl
3
eutectic samples, the experimental heat capacities for all samples are shown in Figure
8B where the heat capacity was done in triplicate for each sample. The reported heat capacity was
determined by taking the average values for all heat capacity as shown in Figure 8C with error bars
representing one standard deviation. Experimentally determined heat capacity was shown to vary slightly
with temperature between 0.67 to 0.57 J/g·K, with an average value of 0.637 ± 0.03 J/g·K (104 ± 5 J/mol·K)
within the range of temperatures from 500 to 720°C, as shown in Figure 8C.
15
450 500 550 600 650 700 750
0
1
2
3
4
5
Sapphire 3
Sapphire 4
Cp, Sapphire (exp.)
Cp, Sapphire (theo.)
Temperature (°C)
DSC (mW)
A.
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
Cp (J/(g⋅K))
2.0% deviation
at 550°C
3.5% deviation
at 700°C
500 550 600 650 700 750
0.0
0.2
0.4
0.6
0.8
1.0
Cp, Sample 1
Cp, Sample 2
Cp, Sample 3
Cp (J/(g⋅K))
Temperature (°C)
B.
450 500 550 600 650 700 750 800 850
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Cp, Avg. Experimental
AIMD, 36 mol% PuCl
3
AIMD, 25 mol% PuCl
3
Cp (J/(g⋅K))
Temperature (°C)
Linear fit
Cp = 0.832 - 3.78E
-4
⋅T(°C)
R
2
= 0.785
300 400 500 600 700 800 900 1000 1100
Temperature (K)
C.
Figure 8. Summary of heat capacity. A. DSC curves for sapphire standard comparing experimentally
determined and NIST reported sapphire heat capacity. B. Experimental heat capacity values for three NaCl-
PuCl
3
eutectic samples. C. Averaged heat capacity valued with error bars of one standard deviation derived
from experimental Cp measurements along with AIMD values calculated for two compositions.
The specific heat was calculated from AIMD simulations using Equation 5, where H is the total
energy, V is the volume, and P is the pressure of the during NPT simulations,
is a temperature change,
and the notation < > indicates the ensemble average values. Ideally,
should be small enough to reduce
the numerical error. However, we found that <H + VP>
NPT
is linear with temperature as seen in Figure
8C; therefore, we can use the slope of the line over the range to simulate temperatures to estimate heat
capacity.
=
Equation 5
16
Experimentally determined heat capacity values were higher than the respective computational values
for the eutectic mixture. For pure PuCl
3
, the calculated specific heat capacity has been reported to be
0.293 J/g·K (101.2 J/mol·K) [54] while the specific heat capacity for NaCl is known to be 50.3 J/mol·K
(0.856 J/g·K). An estimated heat capacity, based on the molar ratio and assuming the molecular weight
for Pu-239 trichloride, is calculated to be 68.4 J/mol·K (0.419 J/g·K). The estimated heat capacity value is
slightly lower than the AIMD-derived value of 0.45 J/mol·K as well as the experimentally determined
value.
The Pu-metal starting material is not isotopically pure Pu-239 as shown in Table 5 which may alter
the experimental heat capacity slightly. Studies to determine the energy produced by alpha disintegrations
in a pure Pu-239 source found that the total energy produced per alpha-disintegration in Pu-239 is 5.23
MeV (8.38·10
-6
erg/atom) [55]. In addition, a study to determine the heat capacity of PuC synthesized
from a 99.93% Pu-metal found that the sample exhibited “self-heating” due to alpha disintegrations and
enthalpy values had to be corrected to account for heat generated within the sample [56]. The
experimental heat capacity data in Figure 8C are uncorrected for internal heating sources, and this may
lead to the differing values between the experiment and calculations.
This work highlights an important point, in that, for salts containing radioactive substances, such as
fuel salts where the sample can exhibit self-heating, heat capacity values may vary from theoretical
calculations and will vary experimentally depending on the isotopic composition. Deviation between
AIMD and experimentally determined values may also arise if the enthalpy of mixing predicted by
simulations does not accurately predict the modeled system; variation in modeled enthalpy is evident in
literature [49, 50, 57].
3.4 Density
The density of the molten NaCl-PuCl
3
mixtures was studied by measuring the buoyancy of a nickel
bobber (99.999%) in the liquid salts at temperatures between 500–800°C. The experimentally determined
values of density for each temperature profile and composition are shown in Figure 9; it is apparent that
the effects of the heating/cooling schedule are non-existent. Figure 9 shows the combined results and
linear equation of fit, representing the equation for density as a function of temperature. Independent
experiment trials determined the 36 mol% and 25 mol% PuCl
3
in NaCl salt to have liquid state densities
as outlined in Equations 6 and 7, respectively.
%
= 3.8589 9.5324 · 10
(, )
= 0.9795
Equation 6
%
= 3.3945 8.4546 · 10
(, )
= 0.9912
Equation 7
17
450 500 550 600 650 700 750 800 850 900 950 1000
2.5
2.6
2.7
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
36 mol% PuCl
3
Heating
Cooling
Random
Linear Fit (exp.)
AIMD
PIM
ρ salt g/cm
3
Temperature (°C)
R
2
= 0.9795
R
2
= 0.9912
25 mol% PuCl
3
Heating
Random
Linear Fit (Exp.)
AIMD
800 850 900 950 1000 1050 1100
Temperature (K)
Figure 9. Experimental and calculated density values for 36 mol% and 25 mol% PuCl
3
in NaCl carrier salt.
Figure 9 shows the density determined by using CMD and AIMD approaches with experimental data.
Predicting the density of actinide molten-salt systems, especially with actinide-rich composition, is a
challenging task due to high computing cost associated with electronic structure calculations of the
system. Given how close the CMD and AIMD values are, CMD should provide a good starting
configuration for AIMD. For the eutectic composition, the PIM density appeared to be lower than the
AIMD density at a wide range of temperature; nonetheless, they are very consistent with experimental
data. Both computational and experimental results show a linear dependence of the density on the
temperature: =
× . The temperature-density correlation coefficient from AIMD and
experiment is 9.3×10
-4
and 9.5×10
-4
(g/cm
3
·K), respectively. The same approach was used to determine
the density for 25 mol% PuCl
3
systems.
4. CONCLUSION
A non-traditional, gas:solid, synthesis path using NH
4
Cl for chlorinating Pu-metal to obtain PuCl
3
salts was demonstrated. NaCl was added to the reaction to form the eutectic composition, 64 mol% NaCl
- 36 mol% PuCl
3
, and was determined to have a purity of 99.7% with trace amounts of iron, americium,
and uranium. By adding NaCl to the eutectic salt, 25 mol% PuCl
3
(in NaCl) was made. Both salt
compositions were shown to be stable up to 800°C, and their melting points and enthalpies of fusion were
measured with high repeatability. The melting point of the 36 mol% PuCl
3
was determined to be 451°C
while the melting point of 25 mol% PuCl
3
was determined to be 598°C. Enthalpy of these salt systems
had not been experimentally determined previously; however, this work determined the enthalpy of fusion
for the 36 mol%PuCl
3
and 25 mol% PuCl
3
to be 140.7 ± 8.4 J/g and 171.4 ± 11.4 J/g, respectively.
The heat capacity of the salt was measured and reinforced the idea that for studying thermal
properties of radioactive salts, the self-heating due to alpha disintegrations may need to be considered to
explain discrepancies in measured properties containing different ratios of isotopes and therefore decay
products. The experimentally determined heat capacity was shown to be slightly higher than AIMD
18
simulation values. However, both experimental and calculated values showed little dependance on
temperature within the range measured. For the eutectic salt, heat capacity was determined experimentally
to be 0.637 ± 0.030 J/g·K between 500 to 720°C while AIMD simulations determined a Cp of 0.45 J/g·K
over the same temperature range. These discrepancies are expected given the experimental challenges of
measuring heat capacity. From the theory point of view, constant pressure molecular simulations are also
extremely sensitive and require extensive molecular models and sampling. Overall, this work shows that
modeling captures the essential trends in these systems and can be used to supplement experimental data.
A combination of experimental measurements and molecular dynamics simulations were used to
determine the density of the two salt compositions. Experimental densities, in the liquid state, were shown
to have a linear decrease with temperature. As expected, the density of the 36 mol% PuCl
3
mixture was
higher than that of the 25 mol% PuCl
3
salt. Experimental and AIMD values for density of the eutectic salt
system were in good agreement and of similar slope; while computed density values for the 25 mol%
PuCl
3
salt showed a differing slope, values were also in close agreement with experimental.
The data reported in this work are both experimental and computation values. Some of the data
presented here are the only known data available for these salt systems in literature. Further work is
needed to verify accuracy and reduce errors, but this serves as a key baseline study for the otherwise
unknown NaCl-PuCl
3
binary salt system.
19
5. REFERENCES
1. Taube, M. and J. Ligou. 1974. “Molten plutonium chlorides fast breeder reactor cooled by
molten uranium chloride.” Annals of Nuclear Science and Engineering 1 (4): 277–281.
https://doi.org/10.1016/0302-2927(74)90045-2
.
2. Nelson, P. A. D. K. Butler, M. G. Chasanov, and D. Meneghetti. 1967. “Fuel Properties and
Nuclear Performance of Fast Reactors Fueled with Molten Chlorides.” Nuclear Applications
3 (9): 540–547. https://doi.org/10.13182/NT67-A27935
.
3. Chasanov, M. G. 1965. “Fission-Product Effects in Molten Chloride Fast-Reactor Fuels.”
Nuclear Science and Engineering 23 (2): 189–190. https://doi.org/10.13182/NSE65-A28145
.
4. International Atomic Energy Agency. 2021. “The Potential Role of Nuclear Energy in
National Climate Change Mitigation Strategies.” IAEA-TECDOC-1984, IAEA.
5. Faure, B. and T. Kooyman. 2022. “A comparison of actinide halides for use in molten salt
reactor fuels.” Progress in Nuclear Energy 144: 104082.
https://doi.org/10.1016/j.pnucene.2021.104082
.
6. Snyder, C., M. H. West, M. D. Ferran, and K. W. Fife. 1988. “The chlorination of plutonium
dioxide.” United States, 18.
7. Tolley, W. B. 1953. “Plutonium Trichloride: Preparation By Reaction With Phosgene or
Carbon Tetrachloride, and Bomb Reduction to Metal.” United States.
https://doi.org/10.2172/4198998
.
8. Stroud, M. A., R. R. Salazar, K. D. Abney, E. A. Bluhm, and J. A. Danis. 2006. “Purification
of Plutonium Chloride Solutions via Precipitation and Washing.” Separation Science and
Technology 41 (10): 2031–2046. https://doi.org/10.1080/01496390600742997
.
9. Boreham, D., J. H. Freeman, E. W. Hooper, I. L. Jenkins, and J. L. Woodhead. 1960. “The
preparation of anhydrous plutonium trichloride.” Journal of Inorganic and Nuclear
Chemistry 16 (1): 154–156. https://doi.org/10.1016/0022-1902(60)80102-9
.
10. Garner, C. S. 1944. “The Preparation of Plutonium Trichloride.” Los Alamos National
Laboratory. https://permalink.lanl.gov/object/tr?what=info:lanl-repo/lareport/LA-00112.
11. Paget, T., J. Mcneese, K. Fife, M. Jackson, and R. Watson. “Chapter 6: Molten Salt
Chemistry of Plutonium.” in Plutonium Handbook (La Grange Park: American Nuclear
Society, 2018).
12. Johnson, K. W. R. and J. A. Leary. 1964. “The Pu-PuCl3 system.” Journal of Inorganic and
Nuclear Chemistry 26 (1): 103–105. https://doi.org/10.1016/0022-1902(64)80236-0.
13. Herrmann, S. D., H. Zhao, K. K. Bawane, L. He, K. R. Tolman, and X. Pu. 2022. “Synthesis
and characterization of uranium trichloride in alkali-metal chloride media.” Journal of
Nuclear Materials 565, 153728. https://doi.org/10.1016/j.jnucmat.2022.153728
.
14. Voronenkov, V., et al. 2020. “Hydride Vapor-Phase Epitaxy Reactor for Bulk GaN Growth.”
2019 Compound Semiconductor Week 217 (3): 1900629.
https://doi.org/10.1109/ICIPRM.2019.8819335
.
15. Bjorklund, C. W., J. G. Reavis, J. A. Leary, and K. A. Walsh. 1959. “Phase Equilibria in the
Binary Systems PuCl3–NaCl and PuCl3–LiCl.” The Journal of Physical Chemistry 63 (10):
1774–1777. https://doi.org/10.1021/j150580a049.
16. Outotec. 2019. “HSC Chemistry 9.” Metso Outotec Finland Oy.
17. Feldman, S. R. 2011. “Sodium Chloride.” in Kirk‐Othmer Encyclopedia of Chemical
Technology. 1-27. https://doi.org/10.1002/0471238961.1915040902051820.a01.pub3.
18. Beard, C., et al. 2008. “Actinide Research Quarterly, in Plutonium Processing at Los
Alamos.” ed. M. Coonley. Los Alamos National Laboratory: Stockpile Manufacturing and
Support (SMS) Directorate. p. 20.
19. Phipps, T. E., G. W. Sears, R. L. Seifert, and O. C. Simpson. 1950. “The Vapor Pressure of
Plutonium Halides.” The Journal of Chemical Physics 18 (5): 713–723.
https://doi.org/10.1063/1.1747733
.
20
20. Robinson, H.P. 1949. “Determination of the Melting Points of Plutonium (III) Chloride and
Plutonium (III) Bromide.” The Transuranium Elements. 952–956.
21. Konings, R. J. M., L. R. Morss, and J. Fuger. “The Chemistry of the Actinide and
Transactinide Elements Thermodynamic Properties of Actinides and Actinide Compounds.”
ed. L. R. Morss, N. M. Edelstein, and J. Fuger. Vol. 4. The Netherlands: Springer. 2113–
2224. https://doi.org/10.1007/1-4020-3598-5_19
.
22. Burns, J. H., J. R. Peterson, and J. N. Stevenson. 1975. “Crystallographic studies of some
transuranic trihalides:
239
PuCl
3
,
244
CmBr
3
,
249
BkBr
3
and
249
CfBr
3
.” Journal of Inorganic and
Nuclear Chemistry. 37 (3): 743–749. https://doi.org/10.1016/0022-1902(75)80532-X.
23. Maenpaa, L. and A. Roine. HSC Chemistry 9, in H, S, and Cp Database. 2019, Outotec
Research Center.
24. Lemire, R. J., et al. 2001. “Chemical Thermodynamics of Neptunium and Plutonium.”
Chemical Thermodynamics 4, ed. O.N.E. Agency. North Holland Elsevier Science.
25. Cleveland, J. M. 1980. “Plutonium Handbook: A Guide to the Technology.” Compounds of
Plutonium, ed. O. J. Wick. Vol. 1. La Grange Park, IL: The American Nuclear Society.
26. Richmond, S., et al. 2010. “The solubility of hydrogen and deuterium in alloyed, unalloyed
and impure plutonium metal.” IOP Conference Series: Materials Science and Engineering, 9:
012036.
27. Haschke, J.M., T.H. Allen, L. A. Morales. 2000. “Surface and Corrosion Chemistry of
Plutonium.” Los Alamos Science 26: 252–273.
28. Bradley, D. 1957. “The Preparation and Properties of the Chlorides of Uranium, Plutonium,
Thorium and of the Fission Product Chlorides, A.E.R.” United Kingdom Atomic Energy
Authority Research Group.
29. Meier, R., et al. 2012. “Recycling of Uranium from Uranium-Aluminum alloys by
Chlorination with HCl(g).” Procedia Chemistry 7: 785–790.
https://doi.org/10.1016/j.proche.2012.10.119.
30. Sonneveld, E. J. and J. W. Visser. 1975 “Automatic collection of powder data from
photographs.” Journal of Applied Crystallography 8 (1): 1–7.
31. American Society for Testing and Materials. 2018. “Standard Test Method for Temperature
Calibration of Differential Scanning Calorimeters and Differential Thermal Analyzers.”
ASTM E967-18, ASTM International.
32. American Society for Testing and Materials. 2018. “Standard Test Method for Thermal Lag
of Thermal Analysis Apparatus.” ASTM E3142-18a, ASTM International.
33. American Society for Testing and Materials. 2018. “Standard Test Method for Determining
Specific Heat Capacity by Differential Scanning Calorimetry.” ASTM E1269-11, ASTM
International.
34. Duemmler, K., et al. 2022. “Evaluation of thermophysical properties of the LiCl-KCl system
via ab initio and experimental methods.” Journal of Nuclear Materials 559.
https://doi.org/10.1016/j.jnucmat.2021.153414
.
35. Hidnert, P. 1957. “Thermal Expansion of Some Nickel Alloys.” Journal of Research of the
National Bureau of Standards 58 (2): 89–92.
36. VandeVondele, J., et al. 2005. “QUICKSTEP: Fast and accurate density functional
calculations using a mixed Gaussian and plane waves approach.” Computer Physics
Communications 167 (2): 103–128. https://doi.org/10.1016/j.cpc.2004.12.014
.
37. Kühne, T. D., et al. 2020. “CP2K: An electronic structure and molecular dynamics software
package - Quickstep: Efficient and accurate electronic structure calculations.” The Journal of
Chemical Physics 152 (19): 194103. https://doi.org/10.1063/5.0007045
.
38. Salanne, M. and P. A. Madden. 2011. “Polarization effects in ionic solids and melts.”
Molecular Physics 109 (19): 2299–2315. https://doi.org/10.1080/00268976.2011.617523
.
21
39. Parrinello, M. and A. Rahman. 1981. “Polymorphic transitions in single crystals: A new
molecular dynamics method.” Journal of Applied Physics 52 (12): 7182–7190.
https://doi.org/10.1063/1.328693
.
40. Martyna, G.J., M.L. Klein, and M. Tuckerman. 1992. “Nosé–Hoover chains: The canonical
ensemble via continuous dynamics.” The Journal of Chemical Physics 97 (4): 2635–2643.
https://doi.org/10.1063/1.463940
.
41. Perdew, J. P., K. Burke, and M. Ernzerhof. 1996. “Generalized gradient approximation made
simple.” Physical Review Letters 77 (18): 3865–3868.
https://doi.org/10.1103/PhysRevLett.78.1396
.
42. Grimme, S., et al. 2010. “A consistent and accurate ab initio parametrization of density
functional dispersion correction (DFT-D) for the 94 elements H-Pu.” The Journal of
Chemical Physics 132 (15): 154104. https://doi.org/10.1063/1.3382344
.
43. Goedecker, S., M. Teter, and J. Hutter. 1996. “Separable dual-space Gaussian
pseudopotentials.” Physical Review B 54 (3): 1703–1710.
https://doi.org/10.1103/PhysRevB.54.1703
.
44. Lu, J.-B., et al. 2021. “Norm-Conserving Pseudopotentials and Basis Sets to Explore Actinide
Chemistry in Complex Environments.” Journal of Chemical Theory and Computation 17 (6):
3360–3371. https://doi.org/10.1021/acs.jctc.1c00026
.
45. VandeVondele, J. and J. Hutter. 2007. “Gaussian basis sets for accurate calculations on
molecular systems in gas and condensed phases.” Journal of Chemical Physics 127 (11).
https://doi.org/10.1063/1.2770708
.
46. Lu, J. B., et al. 2019. “Norm-Conserving Pseudopotentials and Basis Sets To Explore
Lanthanide Chemistry in Complex Environments.” Journal of Chemical Theory and
Computation 15 (11): 5987–5997. https://doi.org/10.1021/acs.jctc.9b00553
.
47. Lippert, G., J. Hutter, and M. Parrinello. 1997. “A hybrid Gaussian and plane wave density
functional scheme.” Molecular Physics 92 (3): 477–487.
https://doi.org/10.1080/002689797170220
.
48. Clow, J., et al. 1994. “Specific activities and DOE-STD-1027-92 hazard Category 2
thresholds, LANL fact sheet.” LA-12846-MS, Los Alamos National Laboratory.
https://doi.org/10.2172/10190930
.
49. Beneš, O. and R. J. M. Konings. 2008. “Thermodynamic evaluation of the NaCl–MgCl2–
UCl3–PuCl3 system.” Journal of Nuclear Materials 375 (2): 202–208.
https://doi.org/10.1016/j.jnucmat.2008.01.007
.
50. Yin, H., et al. 2020. “Thermodynamic description of the constitutive binaries of the NaCl-
KCl-UCl3-PuCl3 system.” Calphad 70. https://doi.org/10.1016/j.calphad.2020.101783
.
51. Zachariasen, W. 1948. “Crystal chemical studies of the 5f-series of elements. I. New structure
types.” Acta Crystallographica 1 (5): 265–268.
https://doi.org/10.1107/S0365110X48000703
.
52. Zachariasen, W. H. and F. H. Ellinger. 1959. “Unit cell and thermal expansion of β-
plutonium metal.” Acta Crystallographica 12 (3): 175–176.
https://doi.org/10.1107/S0365110X59000524
.
53. Archer, D. G. 1993. “Thermodynamic Properties of Synthetic Sapphire (α‐Al2O3), Standard
Reference Material 720 and the Effect of Temperature‐Scale Differences on Thermodynamic
Properties.” Journal of Physical and Chemical Reference Data 22 (6): 1441–1453.
https://doi.org/10.1063/1.555931
.
54. Guillaumont, R., et al. 2003. “Update on the Chemical Thermodynamics of Uranium,
Neptunium, Plutonium, Americium and Technetium.” Vol. 5. OECD Nuclear Energy
Agency.
55. Stout, J. W. and W. M. Jones. 1947. “A Calorimetric Determination of the Energy Produced
by Plutonium (239).” Physical Review 71 (9): 582–585.
https://doi.org/10.1103/PhysRev.71.582
.
22
56. Kruger, O. L. and H. Savage. 1964. “Heat Capacity of Plutonium Monocarbide from 400° to
1300°K.” The Journal of Chemical Physics 40 (11): 3324–3328.
https://doi.org/10.1063/1.1725001
.
57. Schorne-Pinto, J., et al. 2022. “Correlational Approach to Predict the Enthalpy of Mixing for
Chloride Melt Systems.” ACS Omega 7 (1): 362–371.
https://doi.org/10.1021/acsomega.1c04755
.
