CONP-891013—8
DE90 013887
TRUEX PROCESS SOLVENT CLEANUP WITH SOLID SORBENTS
Pui-Kwan Tse, Lucinda Reichley-Yinger, and George F. Vandegrift
Argonne National Laboratory
9700 South Cass Avenue
Argonne, IL 60439
ABSTRACT
Solid sorbents, alumina, silica gel, and Amberlyst A-26
have been tested for the cleanup of degraded TRUEX-NPH
solvent.
A sodium carbonate scrub alone does not
completely remove acidic degradation products from highly-
degraded solvent and cannot restore the stripping
performance of the solvent. By following the carbonate
scrub with either neutral alumina or Amberlyst A-26 anion
exchange resin, the performance of the TRUEX-NPH is
substantially restored. The degraded TRUEX-NFH was
characterized before and after treatment by supercritical
fluid chromatography. Its performance was evaluated by
americium distribution ratios, phase-separation times,
and lauric acid distribution coefficients.
INTRODUCTION
The TRUEX process has been developed over the last decade (1-7)
as a treatment for PUREX process raffinates and other acidic nitrate
waste streams contaminated by transuranics
(TRU).
The key
ingredients in the TRUEX solvent extraction process are
octyl(phenyl)N,N-diisobutylcarbamoylmethylphosphine oxide
(CMPO),
a
bifunctional extractant, and tri-n-butyl phosphate
(TBP),
which acts
as a phase modifier and in some cases as a co-extractant. Both of
these extractants are subject to degradation under processing
conditions for nuclear waste, and their degradation products affect
both the extraction and stripping effectiveness of the TRUEX
process.
A normal paraffinic hydrocarbon (NPH) is used as diluent.
The concentration of CMPO in the TRUEX-NPH solvent affects
metal distribution ratios and, therefore, the success of the TRUEX
process.
Four studies of TRUEX solvent degradation have been
MASTER
DJSTRIBUTION OF THIS DOCUMENT IS UNulMlTED
TSE,
REICHLEY-YINGER, VANDEGRIFT
reported
(7-10).
In three studies
(7-9),
the amount of CMPO in the
degraded TROEX solvent was determined from americium distribution
ratios between the degraded TRUEX solvent and 2M
HN03,
assuming a
third-order extractant dependency for americium. In the other study
(10),
the amount of CMPO present in the degraded solvent was
determined by gas chromatography (GC). Gas chromatography is a very
powerful tool for qualitative and quantitative measurements of
volatile organic compounds, but it is less suitable for analyzing
thermally unstable compounds such as CMPO. CMPO decomposes at
~180°C,
a temperature much lower than normal gas chromatographic
conditions.
Supercritical fluid chromatography has been found to be
a useful tool for analyzing thermally unstable and non-volatile
compounds such as CMPO (11).
The presence of acidic impurities and degradation products has
a drastic effect on the stripping properties of the TRUEX solvent;
their formation, therefore, is usually of greater concern than the
loss of CMPO to successful processing using the TRUEX process.
Measurements of the distribution ratios of americium between
degraded TRUEX solvents and 0.01M and
0.05M
HNO3 solutions were the
basis for measuring the effects of the formation of acidic
degradation products in three studies
(7-9)-
Measurements of
concentrations of acidic degradation products by methylation and GC
analysis were the basis of the fourth study (10).
Solvent cleanup methods used and proposed for use in the PUREX
process include 1) washing the solvent with sodium carbonate
solution; 2) washing the solvent with hydrazine carbonate (12); and
3) passing the solvent through a column of solid sorbents such as
macroreticular resins (13), titanium dioxide (12), sodium silica gel
(14),
and alumina (15). The first two methods are subject to
operational difficulties because of slow phase disengagement
separation, gassing, and interfacial "crud" formation. Organic
resins used as solid sorbents are susceptible to chemical and
radiation damage and work well only with acid-free solvents. The
titanium dioxide method has not been demonstrated on a large scale.
The silica gel and alumina methods have shown promise for
eliminating many of the problems encountered with organic resins.
The purpose of the present work is to develop a procedure for
cleaning up degraded TRUEX-NPH process solvent, which initially
contains 0.2M CMPO and 1.4M TBP in an NPH diluent. Development of
the procedure has been based on 1) the effectiveness of a carbonate
wash followed by treatment with a solid sorbent compared to a
carbonate wash alone and 2) the relative effectiveness of the
alumina, silica gel, and Amberlyst A-26 sorbents. Phase-separation
times,
distribution coefficients for lauric acid, and americium
distribution ratios have been measured to determine the
effectiveness of the solvent treatments. The solvent has also been
characterized by supercritical fluid chromatography (SFC) before and
after degradation and after each treatment.
TRUEX PROCESS SOLVENT CLEANUP WITH SOLID SORBENTS
EXPERIMENTAL
ar——
Materials
Extraction-grade CMPO with a purity of approximately 98% was
purchased from M&T Chemical Company, Rahway, NJ. The material used
for the experiments was further purified by a modified MIX procedure
(16).
The modification included reducing the reaction temperature
to room temperature and increasing the reaction time to two hours
for treatment with each ion exchange resin. Finally, CMPO was
recrystallized from heptane. The resulting white crystalline
material was analyzed by supercritical fluid chromatography and
found to be >99% pure. Gold-labeled TBP was obtained from Aldrich
Chemical Company Conoco
C^2~^14
was obtained from Vista Chemical
Company, Westlake LA.
Solid sorbents tested included alumina, sodium silica gel, and
Amberlyst A-26. Activated neutral alumina was obtained from Aldrich
Chemical Co. as the Brockman 1, 150 mesh. Davisil silica gel (grade
634,
100-200 mesh) from Aldrich Chemical Co. was treated with NaOH,
water washed, and air dried before use. Amberlyst A-26 anion
exchange resin (chloride-form) was obtained from American Scientific
and was converted to the hydroxide-form by treating the resin with
1M NaOH. All other chemicals used were reagent grade or better.
Solvent Degradation
One liter of TRUEX-NPH solvent was degraded by stirring and
heating it for four hours under refluxing conditions with an equal
volume of 8M
HNO3.
The phases were then separated, and the organic
phase was washed with water to remove nitric acid.
Supercritical Fluid Chromatography
Analyses of the solvent were performed on a Lee Scientific
Model 622 supercritical fluid chromatograph/gas chromatograph
(SFC/GC) with flame ionization detector
(FID).
The properties of
SFC Have been discussed in detail elsewhere (11). Experimental
conditions used were:
1) Carrier fluid: SFC grade C0£ from Scott Specialty Gas
(linear flow rate was controlled by the length of the frit
restrictor and was usually set at 10 times the minimum
linear
velocity),
2) Injection temperature: room temperature,
3) Density (pressure) program: initially 0.25 g/mL held for
10 min; ramp 0.02 g/mL/min to 0.6 g/mL; held for 2 min,
A) Oven temperature: 100°C,
5) Detector temperature: 325°C, and
6) Column: 50 /Jm i.d. x 10 m SB-Methyl-100 (Lee Scientific)
TSE,
REICHLEY-YINGER, VANDEGRIFT
The samples for SFC analysis were prepared by dissolving 150
flL
(~0.12
g) of the analyte in 10 mL of dichloromethane (B&J GC/MS
grade).
Concentrations of CMPO, TBP, NPH, and their degradation
products were calculated by comparison with an external standard,
the undegraded TRUEX-NPH.
Phase-Separation Times
Phase-separation times were used to estimate the ability of
contacting equipment to mix and separate the two phases in a
process.
The phase-separation times were determined by the Mailen
et al. method (14). Accordingly, 2 mL of the organic solution and
2 mL of
0.25M
Na2C03 were placed in a 13 x 100 mm culture tube and
mixed by slow inversion for 15 sec. The separation time was taken
to be the time for the emulsion to collapse to one layer of drops at
the interface of the two phases.
Distribution Measurements
Distribution measurements for atnericium between TRUEX-NPH and
HNO3 and for lauric acid between TRUEX-NPH and solid sorbents were
performed at 25°C. Americium distribution ratios (D^) were
determined by contacting aliquots of washed and nitric-acid-
preequilibrated TRUEX-NPH with 0.01, 0.05, and 2M HNO3 containing
trace amounts of
Am-241.
Lauric acid distribution coefficients (K,j)
were measured by adding a trace amount of [1-^C] lauric acid to the
washed TRUEX-NPH and contacting it with a fixed amount of solid
sorbent,
which had been pre-equilibrated with heptane and dried
under nitrogen. For D^, two aliquots from each phase were analyzed
by gamma counting (MINAXI, United Technologies,
Parkard).
The
initial and final activities of lauric acid in the organic phase
were counted with a Parkard Tri-Carb 2000 Liquid Scintillation
Analyzer. Distribution measurements were reproducible within +5%.
RESULTS AND DISCUSSION
Characterization of Degraded TRUEX-NPH
The extent of degradation in the hydrolyzed TRUEX-NPH solvent
was characterized by SFC and D^ measurements. Supercritical fluid
chromatograms of undegraded and degraded TRUEX-NPH are shown in
Figs.
1 and 2, respectively. The concentrations of CMPO and TBP in
the degraded solvent were determined by SFC analysis to be
0.055M
and
1.14M,
28% and 82% of their initial values. The former compares
well with a CMPO concentration of
0.057M
calculated by a third-order
dependence on the D^m value for 2M HNO3 (see Table 1). Based on the
decrease in peak areas, the NPH concentration in the degraded
solvent was 80% of that in the undegraded solvent.
Nineteen additional peaks are found in the chromatogram of the
degraded solvent (Fig. 2). These degradation products appear
predominantly at retention times between 15 and 20 min. When a
TRUEX PROCESS SOLVENT CLEANUP WITH SOLID SORBENTS
0 10 20 30
Time,
min
1
CH
2
CI
2
2 C
12
H
26
3 C
13
H
28
4 c
14
H
30
5 TBP
6 CMPO
Fig.
1. Supercritical Fluid Chromatogram of Undegraded TRUEX-NPH
Solvent
