D. G. Karraker, T. S. Rudisill, and M. C. Thompson
Westinghouse Savannah River Company
Aiken, SC 29808

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

This report has been reproduced directly from the best available copy.

Available for sale to the public, in paper, from:  U.S. Department of Commerce, National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161,  phone: (800) 553-6847,  fax: (703) 605-6900,  email:  orders@ntis.fedworld.gov   online ordering:  http://www.ntis.gov/support/index.html

Available electronically at  http://www.osti.gov/bridge/

Available for a processing fee to U.S. Department of Energy and its contractors, in paper, from: U.S. Department of Energy, Office of Scientific and Technical Information, P.O. Box 62, Oak Ridge, TN 37831-0062,  phone: (865 ) 576-8401,  fax: (865) 576-5728,  email:  reports@adonis.osti.gov

Key Words: Plutonium, Neptunium, Acetohydroxamic Acid, Solvent Extraction

The distribution coefficients for Pu(IV) and Np(IV) have been measured for the systems HNO₃/acetohydroxamic acid (AHA) and HNO₃/NaNO₃/AHA (aqueous) versus 30 vol% TBP (organic) that apply to the flowsheet for the Uranium Extraction (UREX) solvent extraction process. In brief, AHA was found to be an adequate complexing agent for reducing the extraction of both Pu(IV) and Np(IV) at low HNO₃ concentration. However, the hydrolysis of AHA in acidic nitrate solutions indicates that its employment in a process will require special handling and further testing for stability in a radiation field is necessary.

Accelerator Transmutation of Waste (ATW) under the Advanced Accelerator Applications (AAA) Program is being developed to improve the performance of long term geologic disposition of spent commercial nuclear reactor fuel. The ATW flowsheet chops and dissolves commercial fuel then separates it into a pure uranium stream for disposal, separate ¹²⁹I and ⁹⁹Tc streams for fabrication into targets for transmutation to short-lived radionuclides, and the transuranium actinide stream, which is processed further, converted to ATW fuel, and transmuted by fissioning to produce electrical power.¹

A solvent extraction process based on tributyl phosphate (TBP) called UREX is being developed to separate uranium from the fission products and other actinides. In the UREX process, AHA is used as a reductant/complexant to reject Pu and Np along with Am, Cm, and fission products. Technetium is extracted and stripped in a separate stream. Neptunium and Pu are prevented from extracting by AHA which complexes Np(IV) and Pu(IV) and reduces Np(VI) to inextractable Np(V).² The AHA is readily decomposed to gaseous products during waste evaporation.

Studies of the reaction of AHA and formohydroxamic acid (FHA) are reported by Taylor, May and co-workers.^3,4 Both AHA and FHA act similarly to rapidly reduce Np(VI) to Np(V), but only slowly reduce Np(V) to Np(IV).² Both AHA and FHA strongly complex Np(IV) and Pu(IV) reducing the distribution coefficients of Np(IV) and Pu(IV) into 30 vol% TBP.^2,4 May et al. showed the effect of FHA on Np(IV) distribution as a function of HNO₃ concentration up to about 4M.⁴ However, further work was needed to determine stability constants for the complexation.

A number of scoping tests were performed with AHA and Pu to determine the chemistry in the system. A small volume of 1M AHA was added to a 2 mL aliquot of 16 g/L Pu(IV) solution in a glass centrifuge tube resulting in formation of an intense red color indicative of the Pu complex with AHA.⁵ The color was stable for several days. Additional 1M AHA solution was added to give a total solution volume of 4 mL with no immediate change observed in the color of the solution. However, several days later the color of the solution had changed to blue Pu(III). An identical test with hydroxylamine nitrate (HAN) instead of AHA immediately gave blue Pu(III). Addition of AHA to the Pu(III) solution resulted in immediate formation of intense red color characteristic of the Pu(IV)-AHA complex. Further addition of HAN to the solution showed no visible reaction. These tests demonstrate that the kinetics of reduction of Pu(IV) to Pu(III) with AHA are slow. Apparently, AHA must slowly hydrolyze producing hydroxylamine, which then reduces the Pu(IV).^3,5 The complex formed between Pu(IV) and AHA is very highly colored and apparently strong enough that addition of AHA to Pu(III) solutions results in immediate oxidation of Pu(III) to Pu(IV) and formation of the highly colored Pu (IV) complex. The Pu(IV)-AHA complex is not reduced by addition of hydroxylamine nitrate (HAN) to the solution.

Although the Pu(IV)-AHA complex is strong, addition of fluoride ion or oxalic acid to the solution resulted in precipitation of all the Pu from solution. In the case of fluoride, the blue PuF₃ was obtained probably because HAN was added to the solution. The Pu valence state in the oxalate was not determined.

Acetohydroxamic acid was titrated with Ce(IV) ion to determine the number of electrons transferred during reductions by AHA. Each mole of AHA required 2 moles of Ce(IV) demonstrating that AHA is a two-electron reducing agent (equation (1)), which is in agreement with the equation shown by Taylor and May.⁵

R-CONHOH + 2H₂O ® R-COOH + HN(OH)₂ + 2H⁺ + 2 e^- (1)

A test was carried out to determine if the Pu-AHA complex was soluble under process conditions. A 2 mL aliquot of 16 g/L Pu(IV) in 0.5M HNO₃ was mixed with 0.5 mL of 1M AHA in a centrifuge cone, the solution centrifuged for 1.25 hours, and samples from top and bottom of the solution analyzed for Pu by alpha counting. Analyses showed no significant difference in the Pu concentration of the two samples proving that no solids were formed in the experiment. Analyses were necessary because the color of the solution was so intense that solids could not be seen if present.

Hydroxamic acids are derivatives of hydroxylamine, which have been known to complex metal ions for many years. Hydroxamic acids are made by reaction of a carbonyl chloride with hydroxylamine and can also be hydrolyzed to form the corresponding carboxylic acid and hydroxylamine (equation (2)).^2,5

R-CONHOH + H₂O + H⁺ ® R-COOH + NH₃OH⁺ (2)

They tend to give very highly colored complexes and absorbance of its complexes with metal ions can be used to analyze for either metal ions or the hydroxamic acid.⁶ Attempts to use the absorbance of the Fe(III) complex at 515 nm to analyze the concentration of AHA in solutions showed that Fe(III) catalyzes the decomposition of AHA. The reaction is second order as shown in equation 3.⁵

-d[AHA]/dt = 0.00205 l/mol-min [AHA][H⁺] (3)

Nunez and Vandegrift show the half-life of AHA destruction decreases as the hydrogen ion activity increases.² This raises questions about the process stability of AHA.

Measurements of the acid hydrolysis of AHA were made at several different acid concentrations and with sodium and uranyl nitrate added. Table 1 gives the half-times for the decomposition in the various solutions. Table 1 shows the concentration of HNO₃, not activity as given in reference 2. Figure 1 shows the half-time for reaction as a function of ionic strength of the solution which decreases as expected since decomposition is a function of the activity of the hydrogen ion. Note that both U and sodium nitrates change the hydrogen activity and the resulting half-life of AHA. The data show that half the AHA will be degraded after about 4 hours in a typical feed solution containing 1.2M U. The hydroxyl amine that results would then slowly reduce Pu(IV) to Pu(III) while the remaining AHA would complex Pu(IV). These data are in direct contrast to the statement by Taylor et al. who found no effect of U(VI) or nitrate ions on the reaction rate with FHA.³ No data were presented by Taylor et al. to demonstrate that conclusion. Taylor and May do note in another report that the presence of U(VI) enhances the reduction of Pu(IV) to Pu(III) and that the hydrolysis of FHA to hydroxylamine affects in reduction of Pu(IV).⁵ Thus, the enhancement of Pu reduction is probably due to more rapid hydrolysis of AHA with U(VI) present in the solution. These results also raise questions about addition of AHA to the UREX feed solution.

Table 1: Decomposition of AHA

HNO3 (M)	NaNO3 (M)	UO2(NO3)3 (M)	Ionic Strength (M)	AHA t1/2 (min)
0.92	0	0	0.92	1100
1.98	0	0	1.98	600
3.31	0	0	3.31	205
0.96	0.59	0	1.55	600
1	0	0.92	2.84	258

Reagent grade TBP and n-dodecane were vacuum distilled prior to use. The 30 vol% TBP in n-dodecane was prepared with glass pipettes just prior to use in order to avoid effects of degradation products such as dibutylphosphoric acid.

Plutonium solution for use in these studies was purified by anion exchange yielding a solution that was 16 g/L Pu in 0.5M HNO₃. Solution used for tracer studies of Pu distribution was prepared by diluting the 16 g/L Pu standard solution 100-fold with 1M HNO₃.

Distribution coefficient data were obtained by placing HNO₃ into a centrifuge tube and adding enough 1M AHA so that the desired HNO₃ and AHA concentrations were obtained in a total volume of 5 mL of aqueous solution. A 5 mL aliquot of 30 vol% TBP was then added to the tube, the phases were mixed to equilibrate, and after settling, 0.100 mL of each phase was removed and analyzed for acid concentration by titration with 0.1M standard NaOH to a phenolphthalein end point. The remaining sample was spiked with 0.100 mL of Pu(IV) solution, re-equilibrated, and both phases sampled for Pu analyses.

Determination of acid concentration by titration to the phenolphthalein end point (pH 8.2-10) can lead to titration of some of the AHA, which has a pK_a of 7.397.⁷ A correction for the AHA was determined by titrating separate aliquots to phenolphthalein and methyl orange end points (pH 3.2-4.4) and taking the difference as the amount of AHA titrated. Using this method, a solution of 0.5M HNO₃ with 0.3M AHA gave a correction factor of 0.19M.

Initial tracer distribution studies for Pu(IV) in nitric acid were made at varying AHA concentrations to determine the concentration needed to ensure rejection of Pu. Table 2 shows the experimental data at three different AHA concentrations. The data with no AHA were taken from the literature.⁸ The data are plotted in Figure 2. It is clear that AHA concentrations of about 0.3M and acid concentrations of <2M HNO₃ are needed for Pu to be successfully removed in the UREX process.

Further tracer distribution tests were made at 1.7M HNO₃ with 0.3M AHA with the data given in Table 3 and plotted on Figure 3. The 1.7M HNO₃ was chosen based on estimates of the expected maximum concentration in the extraction section if the scrub acidity is 0.5M in the UREX process. The results show that a concentration of greater than 0.1M AHA will be needed at acid concentrations in this range. A proof of concept test at Argonne National Laboratory (ANL) last year showed that the scrub acidity needed to be lowered from 3M to less than 1M to ensure extraction of Tc along with U.⁹ Thus, a change of the scrub acid concentration is needed to recover Tc as well as ensure that Pu is rejected to the aqueous raffinate.

Table 2: Pu Distribution in 30 Vol% TBP from AHA and HNO₃ Solutions

HNO3(aq) (M)	HNO3(or) (M)	Pu Do/a	AHA (M)
0.51	NM	0.837	0
1	NM	2.7	0
2	NM	8.059	0
3	NM	16	0
4	NM	25	0
6.48	NM	43.4	0
3.099	0.599	4.42	0.02
0.34	0.083	0.035	0.1
0.816	0.248	0.288	0.1
2.79	0.568	2.15	0.1
5.85	0.847	4.42	0.1
0.517	0.155	0.045	0.3
0.888	0.248	0.13	0.3
1.71	0.37	0.167	0.3
2.19	0.444	0.41	0.3
5.527	0.878	1.44	0.3

Table 3: Pu Distribution in 30 Vol% TBP from 1.7M HNO₃ with AHA

The data collected for acid solutions were inadequate for modeling work at ANL. As a result, tracer distribution coefficients for Pu(IV) with NaNO₃ salting at 0.5M HNO₃ and 0.3M AHA were measured with the results shown in Table 4 and on Figure 4. Table 5 shows the data for NaNO₃ salting without AHA present. Note that the aqueous acid concentrations given in Table 4 have been corrected for AHA. Table 4 has the only data for which an AHA correction was made. The results of NaNO₃ salting with AHA are not what would be expected based on the two points without AHA. The distribution of Pu would be expected to increase with nitrate salting at constant acid concentration with increasing nitrate salting similar to the behavior in acid solutions shown in Table 2. However, the reverse is true with distribution decreasing as in the case of increasing AHA concentration at constant acid (see Table 3 and Figure 3). It appears that the complexation by AHA is enhanced with added NaNO₃, which might be associated with changes in the activity of AHA. It is also possible that the increase in nitrate concentration increases the rate of destruction sufficiently that some Pu(IV) is reduced to Pu(III) with the remaining Pu(IV) complexed by AHA. The one point at 0.02M HNO₃ and 5.9M NaNO₃ indicates the effect of AHA without Pu reduction. The observed value of 5 is lower than the value at 2.36M NaNO₃ and 0.5M HNO₃. The modeling performed at ANL may help explain the changes occurring in solution.

Table 4: Pu Distribution from NaNO₃ Solutions with 0.3M AHA

(1) – Corrected for AHA
NM – Not Measured

 

Table 5: Pu Distribution from NaNO₃ Solutions with No AHA

Reagent grade TBP and n-dodecane were vacuum distilled prior to use. The 30 vol% TBP in n-dodecane was prepared with glass pipettes just prior to use in order to avoid effects of degradation products such as dibutylphosphoric acid.

Neptunium solution for use in these studies was prepared by spiking Np into a high acid solution, adding ferrous sulfamate (FeSA), extracting into 30 vol% TBP, and back extracting into 1M HNO₃. The solution contained 4.23 x 10⁵ d/m/mL with 80 % Np alpha, 16 % ²³⁹Pu alpha and 4 % ²³⁸Pu alpha.

Acid distribution data measurements involved mixing 30 vol% TBP with an equal volume of the desired HNO₃, centrifuging, and titrating 0.100 mL of each phase to a phenolphthalein end point. The mixture was then spiked with 0.100 mL of Np tracer, 0.025 mL of 0.5M FeSA, and the chosen volume of 3M AHA; the mixture was re-equilibrated, centrifuged and sampled. Organic solutions were analyzed by liquid scintillation counting; aqueous samples were plated for alpha PHA. It was assumed that the organic count was all Np; the counts from the aqueous plates were corrected using the PHA results.

In the presence of solid salt (NaNO₃), liquid scintillation counting of the aqueous phase was erratic, since only a 0.010 mL sample could be tolerated because of nitrate quenching. The small sample size led to large errors, so the distribution coefficients (D_o/a’s) were calculated from the addition of a constant amount of Np tracer to the equilibrations. Assuming only Np(IV) would be extracted, D_o/a can be calculated from equation (4),

D_o/a = Np_(o)/(Np_Total-Np_(o)) (4)

where Np_Total and Np_(o) are the total and organic phase Np activities, respectively. Total Np was measured separately from counting plates.

Neptunium distribution between 30 vol% TBP and varying concentrations of HNO₃ was measured with 0.1M AHA and without AHA. The data are shown in Table 6 and plotted on Figure 5. The curves are non-linear fits to the data. Table 7 and Figure 6 show similar data from May et al. with and without FHA (as interpolated from the figures).⁴ Figure 7 combines the data from Tables 6 and 7 to better show the comparison. The present data for Np distribution without complexant matches well except for the last point, which was not "smoothed" using equation (4). Thus, the present method appears to produce data with only slightly more scatter than that shown by May. The lower curve in Figure 7 shows a fit of both the May data for FHA and the present data for AHA. The present data has more scatter, but it would appear that reduction of Np distribution into TBP is about the same for both hydroxamic acids. The reduction in distribution is reduced at higher acid concentrations as expected due to reduced ionization of the hydroxamic acid function. The data confirm that addition of AHA to the UREX process would result in Np rejection to the waste at acid concentrations below about 1.5M.

Table 6: Distribution of Np(IV) in 30 Vol% TBP from AHA/HNO₃ Solutions

Table 7: Effect of Increased HNO₃ on Np(IV) Distribution between
30 Vol% TBP and An Aqueous Phase Containing 0.1M FHA⁴

May et al. also showed a plot of Np distribution as a function of the mole ratio of AHA to Np (0.005M) at 1.2M HNO₃.⁴ The data interpolated from the plot are shown in Table 8 and on Figure 8. Figure 8 also shows a point from the present work interpolated from the data in Table 6. The point is lower than the data from May, which may result from the difference in Np concentrations.

Table 8: Effect of Increased AHA Concentration on
Np(IV) Distribution between 30 Vol% TBP and HNO₃

[Np] = 0.005M
[HNO₃]_(aq) = 1.2M

Table 9 shows Np distribution at constant starting acid concentration with varying NaNO₃ concentrations with and without AHA. Figure 9 shows a plot of the data. Both sets of data are scattered, but clearly show the effect of addition of 0.1M AHA in reducing the distribution of Np in these solutions. This data set needs to be repeated to obtain less scatter in the data.

Table 9: Distribution of Np(IV) in 30 Vol% TBP from AHA/NaNO₃/HNO₃ Solutions

Laboratory experiments showed that AHA is hydrolyzed in acid solutions in times that are short with respect to plant operation. Addition of uranium or sodium nitrate increases the rate of hydrolysis by increasing the activity of hydrogen ion in solution. Distribution of Pu(IV) and Np(IV) from HNO₃ and HNO₃-NaNO₃ solutions into 30 vol% TBP is reduced by complexation with AHA making it feasible to reject both actinides to the waste during solvent extraction processing for recovery of U and Tc. In NaNO₃ solutions with constant HNO₃, Pu distribution decreases with increasing NaNO₃ concentration indicating that Pu(IV) is being reduced as well as being complexed by AHA.

While AHA can complex Pu(IV) and Np(IV) to prevent their extraction with U in the proposed flowsheet, it is necessary also to ensure that Pu(IV) is not re-extracted in the extraction section of the A contactor. Should the Pu(IV) be re-extracted, a reflux situation would be created where Pu would accumulate in the bank, with possibly dangerous consequences. The stability of AHA in the system is a major concern in this respect. The hydrolysis of AHA in acid solution would suggest that the 1AS cold feed should be introduced into the contactor by combining two separate streams, 0.5M HNO₃ and 0.6-1M AHA, at the contactor rather than as a single stream. The make-up of a single stream could have a substantial loss of AHA during the time necessary for the stream to be sampled, analyzed, and the results reported to the operating staff.

The effect of radiolysis on AHA has not been investigated; but until this has been done, the flowsheet should be regarded as tentative.

1. J. J. Laidler, L. Burris, E. D. Collins, J. Duguid, R. N. Henry, J. Hill, E. J. Karell, S. M. McDeavitt, M. C. Thompson, M. A. Williamson, and J. L. Willit, Chemical Partitioning Technologies for an ATW System, Progress in Nuclear Energy, 38, 65-79 (2001).
2. L. Nunez and G. F. Vandegrift, Evaluation of Hydroxamic Acid in Uranium Extraction Process: Literature Review, ANL-00/35, March, 2001.
3. R.J. Taylor, I. May, A.L. Wallwork, I.S. Dennis, N.J. Hill, B.Ya. Galkin, B.Ya. Zilberman, and Yu.S. Fedorov, "The Applications of Formo- and Aceto-Hydroxamic Acids in Nuclear Fuel Processing", J. Alloys & Comp., 271-273, 534-537 (1998).
4. May, R.J. Taylor, and G. Brown, "The Formation of Hydrophilic Np(IV) Complexes and Their Potential Application in Nuclear Fuel Reprocessing", J. Alloys & Comp., 271-273, 650-653 (1998).
5. R. J. Taylor and I. May, The Reduction of Actinide Ions by Hydroxamic Acids, Czech. J. Phys.,49, 617-621 (1999).
6. P. Alimarin, F. P. Sudakov, and B. G. Golovkin, Use of n-Benzoyl-n-Phenylhydroxylamine in Analytical Chemistry, Russ. Chem. Rev., 31, 466-474 (1962).
7. B. Chatterjee, Donor Properties of Hydroxamic Acids, Coord. Chem. Rev., 26, 281-303 (1978)
8. G. Petrich and Z.Kolarik, The 1981 PUREX Distribution Data Index, KfK 3080, Institute for Hot Chemistry, Karlsruhe, Germany, January, 1981.
9. G. F. Vandegrift, Personal Communication, November 2000.

Figure 1: Hydrolysis of AHA in Nitric Acid Solutions

Figure 2: Pu Distribution from Nitric Acid

Figure 3: Pu Distribution in 30 Vol% TBP from 1.7M HNO₃ with AHA

Figure 4: Effect of AHA on Pu Distribution in
30 Vol% TBP from NaNO₃ Solutions Containing 0.5M HNO₃

Figure 5: Distribution of Np(IV) in 30 Vol% TBP from AHA/HNO₃ Solutions

Figure 6: Distribution of Np(IV) in 30 Vol% TBP in FHA/HNO₃ Solutions⁴

Figure 7: Distribution of Np(IV) in 30 Vol% TBP from AHA/FHA/HNO₃ Solutions

Figure 8: Distribution of Np(IV) in 30 Vol% TBP as a Function of AHA Concentration

Figure 9: Distribution of Np(IV) in 30 Vol% TBP from AHA/NaNO₃/HNO₃ Solutions