Journal of Radioanalytical and Nuclear Chemistry

, Volume 296, Issue 3, pp 1275–1286

Separation and recovery of Cm from Cm–Pu mixed oxide samples containing Am impurity

  • Hirokazu Hayashi
  • Hiromichi Hagiya
  • Seong-Yun Kim
  • Yasuji Morita
  • Mitsuo Akabori
  • Kazuo Minato
Article

DOI: 10.1007/s10967-012-2304-y

Cite this article as:
Hayashi, H., Hagiya, H., Kim, SY. et al. J Radioanal Nucl Chem (2013) 296: 1275. doi:10.1007/s10967-012-2304-y

Abstract

Curium was separated and recovered as an oxalate from a Cm–Pu mixed oxide which had been a 244Cm oxide sample prepared more than 40 years ago and the ratio of 244Cm to 240Pu was estimated to 0.2:0.8. Radiochemical analyses of the solution prepared by dissolving the Cm–Pu mixed oxide in nitric acid revealed that the oxide contained about 1 at% of 243Am impurity. To obtain high purity curium solution, plutonium and americium were removed from the solution by an anion exchange method and by chromatographic separation using tertiary pyridine resin embedded in silica beads with nitric acid/methanol mixed solution, respectively. Curium oxalate, a precursor compound of curium oxide, was prepared from the purified curium solution. 11.9 mg of Cm oxalate having some amounts of impurities, which are 243Am (5.4 at%) and 240Pu (0.3 at%) was obtained without Am removal procedure. Meanwhile, 12.0 mg of Cm oxalate (99.8 at% over actinides) was obtained with the procedure including Am removals. Both of the obtained Cm oxalate sample were supplied for the syntheses and measurements of the thermochemical properties of curium compounds.

Keywords

244Cm Anion-exchange separation Chromatography Tertiary pyridine resin Cm/Am separation Oxalate 

Introduction

Measurements of the thermochemical properties of the curium compounds are necessary to understand the behavior of curium in nuclear fuel cycles. However, it is hard to obtain curium samples for measurements of the thermochemical properties. For such measurements, samples of mg or larger scale with high purity are necessary. On the other hand, in general, the handling quantity of curium is limited not to exceed mg-scale by the regulation depending on the shielding ability and other conditions of the facilities in Japan. Furthermore, in case of 244Cm, the most available isotope of curium, the half life of which is 18.1 years [1], 1 % of 244Cm decayed into 240Pu in three months. Therefore, measurements of the thermochemical properties of the samples of more than 99 % in purity can be carried out in a few months after preparation of the sample. For such purposes, we need to adopt a method suitable for separation and recovery of mg-scale sample and for supply of the sample immediately.

Procedures for separation and recovery of curium up to kg-scale from the irradiated fuels and targets, or high-level wastes have been summarized in literature [2, 3, 4]. For example, in the production of 244Cm at Savannah River site, Cm was separated and recovered in shielded hot cells after successive irradiations of 239Pu. The separation procedure includes recovery of Pu by solvent extraction with TBP in n-dodecane from nitric acid solution, removals of rare earths by solvent extraction, and those of Am by precipitation of K5AmO2(CO3)3 in hydrochloric solution. After the purification, curium oxalate, a precursor compound of curium oxide, was precipitated and recovered [4, 5].

In case of separation and recovery of mg-scale Cm from an aged 244Cm samples, reducing the loss of samples in the procedure is necessary. Glove boxes should be used for handling such small amounts of sample rather than the hot cells. An anion-exchange technique has advantages such as simplicity of handling over the solvent extraction to remove 240Pu from the solution [6]. On the other hand, nitric acid solution is preferable for the procedure in the glove boxes because it is less corrosive than other acid solutions such as hydrochloric acid. The oxalate precipitation method [2, 3, 4, 5, 7] is preferable for recovery of Cm as a solid compound, though the method of recovery from the solution containing mg-scale Cm with low concentration should be certified to reduce the loss of Cm in the recovery procedure.

In the present study, we demonstrated the separation and recovery of Cm sample in solid form from an aged 244Cm oxide sample to establish the procedure for purification and recovery of Cm sample with a high recovery ratio for the thermochemical properties measurements. The aged 244Cm oxide sample used in this study had been prepared more than 40 years ago and was considered as a 244Cm–240Pu mixed oxide as a result of the decay of 244Cm in the oxide. The purification of Cm sample was by removals of Pu, which was the main element of the oxide. Other impurities which had been found in the composition analyses of the starting material were removed after the removals of Pu.

Removals of Pu from the nitric acid solution of Cm–Pu mixed oxide

After the dissolution of the Cm–Pu mixed oxide into nitric acid solution and analyses of the contents in the solution, removals of Pu from the solution by using anion exchange resin column were carried out.

244Cm-240Pu mixed oxide

Black 244Cm–240Pu mixed oxide powder used in this study was originally 244Cm oxide sample prepared in Oak Ridge National Laboratory (ORNL) 40 years ago. About 20 % of 244Cm is considered to remain and the rest of 244Cm decayed into 240Pu. Table 1 shows the initial composition data (mass assay data of December 1, 1969) provided from ORNL, the decayed composition data measured by α spectrometry in 1999 [8], and those in 2009 (40 years later) calculated from the initial composition regarding the decay of the nuclides.
Table 1

The composition of the 244Cm oxide sample at 1969 (provided by ORNL), 1999 [8], and 2009 (calculated)

Nucleus

t1/2

Composition over actinides (at%)

 

(years)

At 1969

At 1999

At 2009

 

[1]

(measured)

(measured [8])

(calculated)

242Cm

0.446

0.0426

 

0

243Cm

29.1

0.0238

 

0.0090

244Cm

18.10

95.08

35.8

20.57

245Cm

8,500

0.729

 

0.727

246Cm

4,730

4.02

3.9

4.00

247Cm

1.56 × 107

0.0659

 

0.0659

248Cm

3.40 × 105

0.0381

 

0.0381

238Pu

87.7

 

0.1

0.0312

239Pu

24,110

  

0.0148

240Pu

6,563

 

60.2

74.31

241Pu

14.35

  

0.001

242Pu

3.733 × 105

  

0.0235

243Pu

5.658 × 10−4

  

0

244Pu

8.08 × 107

  

0

236U

2.342 × 107

  

0.1954

234U

2.455 × 105

  

0.0114

241Am

432.2

  

0.0013

243Am

7,370

  

0

Preliminary experiments of Pu removal using 0.8 mg of Cm–Pu mixed oxide

Experimental

To confirm the condition of the dissolution of the Cm–Pu mixed oxide and the removal of Pu using an anion-exchange resin, preliminary experiments were carried out by using a small amount of the sample. About 0.8 mg of the Cm–Pu mixed oxide was dissolved in nitric acid–H2O2–water system by heating with a hot plate and stirring with a Teflon-coated stirrer. The solution was dried and dissolved in 10 mL of 8 M nitric acid, dried again and dissolved in 4 mL of 8 M nitric acid. After filtration, 10 mL of Cm–Pu solution was prepared by adding 8 M nitric acid solution. 5 mL of the solution was used for the experiment. After adding 1 mL of H2O2 solution (30 wt%) to the Cm–Pu solution and being kept in the air for 4 days to adjust the valence state of Pu as Pu(IV), followed by adding 1.5 mL of concentrated nitric acid to adjust the acidity to 8 M, the solution was passed through the column of polypropyrene (Muromac column M; 2.5 mL in volume) having the anion-exchange resin (Bio-Rad AG1 1-X4 200–400 mesh) to remove Pu(IV) ions in the feed solution [6]. The resin was conditioned by passing 8 M nitric acid in the column in advance. 25 mL of 8 M nitric acid was also passed through the column to elute Cm ions and obtain “Cm fraction solution.” 10 mL of 0.5 M nitric acid followed by 10 mL of 0.1 M nitric acid were used to elude the absorbed Pu ions from the resin to obtain “Pu fraction solution.” All the elution was by gravity. The conditions of the experiments were shown in Table 2.
Table 2

The conditions and the results of the Pu removals experiments

 

Preliminary experiment

Run 1

Run 2

HNO3 solution dissolving the oxide

8 M 5 mL

9 M 4.2 mL

12.6 M 2 mL

 244Cm in the solution

35.3 μg

9.79 mg

9.18 mg

 240Pu in the solution

115 μg

40.0 mg

36.5 mg

Pu removal (1) with the anion exchange resin column of 2.5 mL in volume

 Feed solution

8 M 7.5 mL

6.6 M 11.5 mL

8.4 M 6 mL

 Elute for Cm fraction (1)

8 M 25 mL

6.6 M 30 mL

8 M 12 mL

 Elute for Pu fraction (1)

0.5 M 10 mL

0.5 M 10 mL

0.5 M 10 mL

 

0.1 M 10 mL

0.1 M 10 mL

0.1 M 15 mL

 Decontamination factor of 240Pu

297

36.4

12.8

 Recovery ratio of 244Cm

96.2 %

98.6 %

96.3 %

 Recovery ratio of 240Pu

69.3 %

87.2 %

70.0 %

Pu removal (2) with the anion exchange resin column of 1.0 mL in volume

 Feed solution

 

6.6 M 49.95 mL

7.6 M 19 mL

 Elute for Cm fraction (2)

 

6.6 M 20 mL

8 M 9 mL

 Elute for Pu fraction (2)

 

0.5 M 10 mL

0.5 M 5 mL

  

0.1 M 10 mL

0.1 M 9 mL

 Decontamination factor of 240Pu

 

1.99

724

 Recovery ratio of 244Cm

 

96.9 %

98.3 %

 Recovery ratio of 240Pu

 

2.0 %

10.9 %

α spectrometer (ORTEC Octete plus, BU-020-450-AS) and γ spectrometer with a low-energy germanium detector (ORTEC Lo-AX -51370/20-P) were used for radioactivity measurements. Each solution prepared for the analyses was diluted by 0.1 M or 1 M nitric acid to be analyzed by α spectrometry and γ spectrometry. The sample for α spectrometry (0.025 or 0.05 mL) were dropped on a counting disk, dried and glazed by using an induction heating apparatus before the measurement. In order to determine the concentration of Pu, some of the samples were analyzed after the extraction of Pu by using 30 vol% TBP/n-dodecane solution.

Results and discussion

Figure 1 shows the α spectra of (a) the Cm–Pu solution in which the Cm–Pu mixed oxide sample was dissolved, (b) Cm fraction solution, and (c) Pu fraction solution. The spectra of Cm–Pu solution have the peaks assigned to 240Pu, 246Cm, 238Pu, and 244Cm, respectively, as reported for the sample of the same batch [8]. The spectra of Cm fraction solution have the peaks assigned to 246Cm and 244Cm; the peaks which can be assigned to 243Am are also observed. The spectra of Pu fraction solution have the peaks assigned to 240Pu, 238Pu and 244Cm. Table 3 includes the radioactivity of each nuclide in the solutions derived from the α spectra.
Fig. 1

The α spectra of a the Cm–Pu solution in which Cm–Pu mixed oxide was dissolved, b the Cm fraction solution, and c the Pu fraction solution in the preliminary experiments on Pu removals

Table 3

The radioactivity (MBq) and the weight (μg) of the nuclides contained in each solution derived from the α spectra and γ spectra and their atomic ratio over actinides in the preliminary experiments of Pu removals

 

244Cm

245Cm

246Cm

238Pu

240Pu

241Am

243Am

Cm–Pu solution(5 mL)

 Radioactivity (MBq)

106 ± 4

0.00850 ± 0.00076

0.0341 ± 0.0046

0.139a

0.963 ± 0.107

0.00353 ± 0.00024

0.0127 ± 0.0001

 Weight (μg)

35.3

1.33

3.00

0.219

115

0.0278

1.72

 Atomic ratio

0.222

0.009

0.019

0.001

0.737

0.000

0.011

Cm fraction (50 mL)

 Radioactivity (MBq)

102 ± 6

0.00908 ± 0.00020

0.0316 ± 0.0067

0.000391a

0.00324 ± 0.0038

0.00335 ± 0.00031

0.0128 ± 0.0001

 Weight (μg)

34.1

1.42

2.78

0.000616

0.387

0.0264

1.73

 Atomic ratio

0.843

0.035

0.068

0.000

0.010

0.001

0.043

Pu fraction (25 mL)

 Radioactivity (MBq)

0.0463 ± 0.0060

ND

ND

0.0805 ± 0.010

0.667 ± 0.002

ND

ND

 Weight (μg)

0.0155

ND

ND

0.127

79.6

ND

ND

 Atomic ratio

0.000

ND

ND

0.002

0.998

ND

ND

aRadioactivity of 238Pu was calculated from that of 240Pu using their ratio in Pu fraction solution

Figure 2 shows the γ spectra of (a) the Cm–Pu solution in which the Cm–Pu mixed oxide sample was dissolved, (b) Cm fraction solution, (c) Pu fraction solution, and (d) Pu fraction solution measured 55 days after the separation. The spectra of Cm–Pu solution have not only the peaks assigned to 240Pu, 244Cm and 245Cm, but also 243Am and 239Np as listed in Table 3. Existence of the peaks assigned to 243Am and 239Np (the daughter nuclide of 243Am) indicates the existence of 243Am as an impurity in Cm–Pu solution sample. The spectra of Cm fraction solution have the peaks assigned to 244Cm, 245Cm, 241Am, 243Am, and 239Np, and those of Pu fraction solution have the peaks assigned to 240Pu and 239Np. The peaks assigned to 239Np (t1/2 = 2.356 day) in Pu fraction solution were not found in the spectra obtained 55 days after the separation. This result indicates the absence of 243Am, the parent nuclide of 239Np in Pu fraction solution.
Fig. 2

The γ spectra of a the Cm–Pu solution in which Cm–Pu mixed oxide was dissolved, b the Cm fraction solution, and c the Pu fraction solution d Pu fraction solution measured 55 days after the separation in the preliminary experiments on Pu removals

The radioactivity, the weight, and the atomic ratio over the actinides in each solution derived from the α spectra and γ spectra are listed in Table 3. The atomic ratio over the actinides in the Cm–Pu solution fairly agrees with those calculated from the initial composition regarding the decay of the nuclides shown in Table 1, except the existence of 243Am. The atomic ratio of 243Am over the actinides in Cm–Pu solution is 1.1 at%. This result indicates the existence of 243Am in the Cm–Pu oxide sample as an impurity. As in Table 2, the recovery yield of 244Cm and the decontamination factor of 240Pu were calculated to be 96.2 % and 297, respectively for the removal of Pu ions using the anion exchange resin. On the other hand, the recovery yield of 240Pu in the Pu fraction solution was 69.3 %. 243Am was not removed by the procedure; it might be due to the fact that Am exists as Am3+ in the nitric acid solution as Cm3+ does [6]. The obtained Cm fraction solution has impurities of 243Am (4.3 at%) and 240Pu (1.0 at%).

The results of the preliminary experiment suggest that the procedure is suitable for the dissolution of Cm–Pu mixed oxides into nitric acid solution and the removal of Pu from the Cm–Pu solution.

Pu removal experiments using 50 mg of Cm–Pu mixed oxide

Experimental

Two sets of the experiments were carried out with using 50 mg of the Cm–Pu mixed oxide in which the amount of 244Cm did not exceed its maximum handling amount (10 mg) defined for our glove boxes by the regulation. Procedures of the dissolution of the Cm–Pu mixed oxide and the removal of plutonium from the solution were similar to the preliminary experiment, though the Pu removal procedures were repeated twice for each run to reduce the plutonium contents in the solution. The first and the second Pu removal procedures were with the column of 2.5 mL in volume (Muromac column M), and that of 1.0 mL in volume (Muromac column S), respectively. In run 1, 6.6 M nitric acid solution was used to reduce the aqueous wastes which were produced when the oxalates were precipitated as described later. The amounts of 6.6 M or 8 M nitric acid solution, 0.5 M nitric acid, and 0.1 M nitric acid solutions used were listed in Table 2. Adjustment of the valence state of Pu as Pu(IV) by adding H2O2 solution was carried out before the first Pu removals in both runs and the second Pu removal in run 2, but not in run 1. Composition analyses of the samples were carried out as described earlier.

Results and discussion

α spectra and γ spectra of the obtained solutions were similar to the corresponding spectra of the preliminary experiments. The weight, and the atomic ratio over the actinides in each solution obtained in run 1 and run 2 derived from the α spectra and γ spectra are listed in Tables 4 and 5, respectively. The recovery yield of 244Cm and the decontamination factor of 240Pu of each step were listed in Table 2, with the recovery yield of 240Pu in the Pu fraction solution. The decontamination factors of 240Pu of the first Pu removal procedures were lower than that in the preliminary experiment; it could be due to the fact that concentration of Pu in the solution was more than 100 times higher than that of the preliminary experiment. The decontamination factor of 240Pu of the second Pu removal in run 1 was lower than those of other Pu removal procedures. It could be due to the fact that H2O2 was not used to adjust the valence of Pu before the second Pu removal procedure; therefore, plutonium ions which were not in Pu(IV) state existed [9] and were not adsorbed to the resin. Therefore, the concentration of Pu in the Cm solution obtained in run 1 is higher than that obtained in run 2.
Table 4

The weight (mg) and the atomic ratio of the nuclides contained in each solution in Pu removal (run 1) derived from the α spectra and γ spectra

 

244Cm

245Cm

246Cm

238Pu

240Pu

241Am

243Am

Cm–Pu solution (10 mL)

 Weight (mg)

9.79

0.399

0.505a

0.0653

40.0

0.00157

0.626

 Atomic ratio

0.188

0.008

0.010

0.001

0.781

0.000

0.012

Cm fraction (1) (50 mL)

 Weight (mg)

9.16

0.359

0.473a

0.00203

1.04

0.00756

0.567

 Atomic ratio

0.788

0.031

0.040

0.000

0.091

0.001

0.049

Pu fraction (1) (25 mL)

 Weight (mg)

0.0141

ND

ND

0.0537

27.7

ND

0.00210

 Atomic ratio

0.001

ND

ND

0.002

0.998

ND

0.000

Cm fraction (2–1) (50 mL)c

 Weight (mg)

8.56

0.362

0.441

0.000986b

0.508

0.00653

0.563

 Atomic ratio

0.819

0.035

0.042

0.000

0.050

0.001

0.054

Cm fraction (2–2) (20 mL)c

 Weight (mg)

0.307

0.0142

0.0158

0.0000318b

0.0164

0.000263

0.0209

 Atomic ratio

0.819

0.038

0.042

0.000

0.044

0.001

0.056

Pu fraction (2) (25 mL)

 Weight (mg)

0.00966

ND

ND

0.0000412b

0.0212

ND

ND

 Atomic ratio

0.309

ND

ND

0.001

0.690

ND

ND

aThe weights of 246Cm were calculated from that of 244Cm using their ratio in Cm fraction solution

bThe weights of 238Pu were calculated from that of 240Pu using their ratio in Pu fraction solution (1)

cThe curium fraction solutions were collected in 2 portions

Table 5

The weight (mg) and the atomic ratio of the nuclides contained in each solution in Pu removal (run 2) derived from the α spectra and γ spectra

 

244Cm

245Cm

246Cm

238Pu

240Pu

241Am

243Am

Cm–Pu-Am solution

 Weight (mg)

9.183

0.389

0.483a

0.00677

36.476

0.0069

0.566

 Atomic ratio

0.192

0.008

0.010

0.002

0.776

0.000

0.012

Cm–Am fraction (1)

 Weight (mg)

8.813

0.356

0.464a

0.00528b

2.845

0.0061

0.540

 Atomic ratio

0.674

0.027

0.035

0.000

0.221

0.001

0.042

Pu fraction(1)

 Weight (mg)

ND

ND

ND

0.00452

25.524

ND

ND

 Atomic ratio

ND

ND

ND

0.002

0.998

ND

ND

Cm–Am fraction (2)

 Weight (mg)

8.663

0.406

0.456a

0.000b

0.004

0.0066

0.572

 Atomic ratio

0.857

0.040

0.045

0.000

0.000

0.001

0.057

Pu fraction (2)

 Weight (mg)

0.024

0.000

0.001a

0.00536

2.774

0.0001

0.002

 Atomic ratio

0.009

0.000

0.000

0.002

0.989

0.000

0.001

aThe weights of 246Cm were calculated from that of 244Cm using their ratio in Cm fraction solution of run 1

bThe weights of 238Pu were calculated from that of 240Pu using their average ratio in Pu fraction solution (1) and Pu fraction solution (2)

The Cm solutions obtained by the Pu removal procedure in run 1 and run 2 have impurities of 243Am (5.4 at%) + 240Pu (4.9 at%), and 243Am (5.7 at%) + 240Pu (0.04 at%), respectively.

Removals of Am impurity from the solution after removals of Pu

In the previous sections, we found that 243Am impurity (ca 1 at%) existed in the aged 244Cm oxide sample, and the concentration of 243Am became about 5 at% by removing Pu. To obtain the Cm sample of higher purity, removing 243Am impurity from the sample is necessary.

The mutual separation of Cm and Am is not only the technique to purify Am and Cm samples for basic scientific research but also one of the key technologies in the partitioning of the radioactive waste management [10]. Procedures for separation of Cm from Cm–Am mixture by solvent extraction, ion exchange, and precipitation methods have been summarized in literature [2, 3, 4, 5, 10]. The mutual separations of actinides attract a concern also in analytical chemistry. Recently, optimization of the condition for the separation of lanthanides, Am, and Cm of tracer amounts for preparation of the samples used for mass spectrometry using high performance liquid chromatography has been reported [11]. Separation of tracer amounts of 245Cm from 249Cf, the parent nuclide of 245Cm by cation exchange method using 2-hydroxy-2-methyl-propanoic acid has also been reported [12]. On the other hand, the mutual separations of trivalent actinides by the chromatographic method using the tertiary pyridine resin and the nitric acid/methanol mixed solvent system have been reported [13, 14, 15]. This method has some advantages such as the simplicity of the procedure, the mildness of the solvent, and avoidance of adding other reagents which must be removed with extra procedures; these advantages are adequate to our purpose. In addition, the optimum composition of the solvent for Am/Cm separation had been discussed [15].

In this section, removals of Am by the chromatographic method using the tertiary pyridine resin with the nitric acid/methanol mixed solvent system were demonstrated. Feasibility of Cm/Am separation by this method which is considered as a key technology in a partitioning of the radioactive waste [14] was also proved by using mg-scale minor actinides.

Preliminary experiments on separation of Cm and Am by using tertiary pyridine resin with μg-scale sample

Experimental

To confirm the condition of the chromatographic separation of Cm and Am in mg-scale by using tertiary pyridine resin, preliminary experiments were carried out by using a solution containing Cm and Am of μg-scale. The highly porous tertiary pyridine resin embedded in silica beads used in this study was produced by the Research Laboratory for Nuclear Reactors, Tokyo Institute of Technology [16]. The solution containing actinides was prepared in the previous section; the atomic ratio of the actinides was Cm:Pu:Am = 0.899:0.044:0.057. The compositions of the solvents used in this study were (1) 8 M HNO3 40 vol%–methanol 60 vol% and (2) 1 M HNO3 20 vol%–methanol 80 vol% solutions, which were recommended as the optimum conditions [14, 15]. The concentrations of the actinides in the solutions were 1.61 MBq/mL of 244Cm and 237 Bq/mL of 243Am for solvent (1) (3 mL), and 1.12 MBq/mL of 244Cm and 168 Bq/mL of 243Am for solvent (2) (2 mL), respectively. Each solution was fed into a column (1 cm in diameter and 10 cm in length) packed with the resin. The column was conditioned with the nitric acid/methanol mixed solution of the same composition before use. Elution was carried out by using the same solvent at the flow rate of 80 mL/h at room temperature. The effluent from the column was collected in fractions every 3 min. The concentrations of 244Cm and 243Am in the fractions were determined from the γ-ray spectra at 152.63 keV for 244Cm and those at 74.66 keV for 243Am.

Results and discussion

Figure 3 shows the chromatogram of the experiment using 8 M HNO3 40 vol%–methanol 60 vol% solution. The elusion peak top of 244Cm and that of 243Am were observed at the effluent volume around 25.5 and 48.9 mL, respectively. The order of the elusion was as reported in literature [13, 14, 15]. Figure 3 indicates that Cm was separated from Am although the tails of each peak were overlapped partly.
Fig. 3

The chromatogram of the preliminary experiment on separation of Cm and Am using 8 M HNO3 40 vol%–methanol 60 vol% solution

The distribution coefficient, the separation factor and the resolution parameter were used to evaluate the separation ability. The distribution coefficient (Kd) and the separation factor (S) are described with the equations,
$$ K_{\text{d}} = \left( {V_{\text{r}} - V_{ 0} } \right)/V_{\text{s}} $$
(1)
$$ S = K_{{{\text{d}},{\text{Am}}}} /K_{{{\text{d}},{\text{Cm}}}} $$
(2)
where Vr, V0, and Vs are the effluent volume at the elution peak, the dead volume of the column packed with resin, and the volume of the resin, respectively [10]. We estimated the apparent separation factor by neglecting the dead volume of the column, Sa = Vr,Am/Vr,Cm. The overlap at 10 % values was used as the resolution parameter according to Suzuki et al. [15],
$$ R = \left( {V_{{{\text{r}},{\text{ Am}}}} - \, V_{{{\text{r}},{\text{ Cm}}}} } \right)/\left( {{{\Gamma}}_{\text{Am}} + {{\Gamma}}_{\text{Cm}} } \right) $$
(3)
where ΓAm and ΓCm are the left side width and the right side width at 1/10 height of each peak maxima, respectively. The apparent separation factor and the resolution parameter of this system were 1.9 and 0.85, respectively. The apparent separation factor was smaller than the reported separation factor with a similar system (2.4) [15]. The difference can be due to the neglect of the dead volume of the column in this study. On the other hand, the resolution parameter was similar to that reported for the similar system (0.7 ± 0.3) [15].
Figure 4 shows the chromatogram of the experiment using 1 M HNO3 20 vol%–methanol 80 vol% solution. The elusion peak top of 244Cm and that of 243Am were observed at the effluent volume around 13.8 and 25.7 mL, respectively. The distribution coefficients which are approximately proportional to the elusion peak positions were about a half of those with 8 M HNO3 40 vol%–methanol 60 vol% solution. This tendency was similar to the reported result [15]. The apparent separation factor and the resolution parameter of this system were 1.9 and 0.73, respectively. The resolution parameter was smaller than that obtained with 8 M HNO3 40 vol%–methanol 60 vol% solution in this study, although the apparent separation factor was similar. The parameters indicating the ability of separation using 1 M HNO3 20 vol%–methanol 80 vol% solution in this study was not good as the reported result of the similar system (S = 2.52 ± 0.02, R = 1.44 ± 0.11) [15]. We selected to use 8 M HNO3 40 vol%–methanol 60 vol% solution for the Am removal experiment, because it indicated a better resolution parameter in the preliminary experiments.
Fig. 4

The chromatogram of the preliminary experiment on separation of Cm and Am using 1 M HNO3 20 vol%–methanol 80 vol% solution

Figure 5 shows the cumulative recovery ratio of 244Cm and that of 243Am calculated from the results of the preliminary experiment using 8 M HNO3 40 vol%–methanol 60 vol% solution. In case the recovery ratio of Cm is over 70 %, the higher the recovery ratio of 244Cm, the lower the purity of 244Cm, because 243Am was contaminated in the effluents as shown in Fig. 3. Figure 6 indicates the calculated atomic ratio of Cm and that of Am recovered from the solution containing Cm and Am (Cm:Am = 0.95:0.05) using the cumulative recovery ratios shown in Fig. 5. This calculation shows that Cm solution with the purity of 99.9, 99.8, and 98.1 % can be obtained by recovering 72.3, 91.4, and 98.1 % of Cm in the solution containing Cm and Am (Cm:Am = 0.95:0.05), respectively. In order to obtain highly pure Cm solution with high recovery ratio of Cm in mg-scale experiment, we decided to use a column which is longer than that used in the preliminary experiment. The longer column is expected to have better separation ability because the position of the elution peak is proportional to the length of the column meanwhile the half width of the elution curve is proportional to the square root of the length of the column if other conditions are identical, according to the plate theory of chromatography [17]. Figure 7 shows the simulated chromatograms, in which the peaks were expressed with Gaussian distribution, with varying the column length based on the plate theory described above. The simulation indicates that 244Cm of the purity higher than 99.9 % can be obtained with the recovering ratio 93.3, 99.8, and 100 % of 244Cm by using 10, 15, and 20 cm column, respectively. We decided to use the column of 20 cm in length which is considered to have enough ability of Cm/Am separation and is short enough to handle in the glove boxes easily.
Fig. 5

The cumulative recovery ratio of Cm and that of Am calculated from the result of the preliminary experiment using 8 M HNO3 40 vol%–methanol 60 vol% solution

Fig. 6

Calculated results of the atomic ratio of Cm and that of Am recovered from the solution containing Cm and Am (Cm:Am = 0.95:0.05) using the cumulative recovery ratio of Cm and that of Am in the preliminary experiment using 8 M HNO3 40 vol%–methanol 60 vol% solution

Fig. 7

Simulation of the chromatogram on separation of Cm and Am using 8 M HNO3 40 vol%–methanol 60 vol% solution with varying the column length based on the results of the preliminary experiment

Removals of Am from the solution containing mg-scale Cm

Experimental

The 8 M nitric acid solution containing Cm and Am obtained by run 2 of Pu removal experiment was used for the experiment on the removals of Am. The Cm–Am solution was dried and then dissolved into 2 mL of 8 M HNO3 40 vol%–methanol 60 vol% mixture. The removal of Am using a column (1 cm in diameter and 20 cm in length) packed with the tertiary pyridine resin and measurements of the concentrations of 244Cm and 243Am in the fractions were carried out with a similar manner described in the previous section.

The fractions having 244Cm with high concentration were gathered; small amounts of water which was used for rinsing each container having the fraction were also gathered together. The gathered solution was dried and dissolved in 2 mL of 4 M nitric acid solution to be used for recovering Cm as an oxalate. Composition analyses of the samples were carried out as described earlier.

Results and discussion

Figure 8 shows the chromatogram of the Cm/Am separation using 8 M HNO3 40 vol%–methanol 60 vol% solution evaluated from the γ spectra. The elusion peak top of 244Cm and that of 243Am were observed at the effluent volume around 33.8 and 80.8 mL, respectively. The peaks of the retention volume were smaller than those expected from the simulation shown in Fig. 7. Moreover, the peaks show conspicuous asymmetry which has the characteristics classified as an “adsorption peak.” The shape of the “adsorption peak” is distorted and the peak maximum moved to a position of lower retention relative to that expected from the plate theory [18]. High concentration of the solutes affects the adsorption isotherm because the solutes cover the surface of the resin and causes the screen effect [18]. The apparent separation factor and the resolution parameter of this system were calculated to be 2.4 and 1.3, respectively. They should be affected not only by varying the column length which was twice larger than that used in the preliminary experiment but also the asymmetry caused by high concentration of the solutes as described earlier. Therefore, these parameters cannot be compared with the results of the preliminary experiments by using the plate theory. Figure 9 shows the cumulative recovery ratio of Cm and that of Am calculated from the result of Cm/Am separation experiment. This calculation indicates that we can obtain the sample having 98 % of 244Cm and 1.3 % of 243Am which originally existed in the feed solution by gathering the fractions from 27.9 to 63.4 mL. It corresponds to that high purity 244Cm solution which contains only 0.07 at% of 243Am impurity can be obtained from the 244Cm solution containing 5 at% of 243Am with high recovering ratio (98 % of 244Cm). This result also shows the feasibility of this method for Cm/Am separation using mg-scale minor actinides although the peak asymmetry of the chromatography can occur when large amounts of actinides are used.
Fig. 8

The chromatogram of the Cm/Am separation using 8 M HNO3 40 vol%–methanol 60 vol% solution

Fig. 9

The cumulative recovery ratio of Cm and that of Am calculated from the result of Cm/Am separation experiment using 8 M HNO3 40 vol%–methanol 60 vol% solution

Fractions from 27.9 to 63.4 mL obtained by the chromatographic separation were gathered. The weight and the atomic ratio over the actinides in the Cm solution obtained by gathering the fractions of Cm/Am separation, which were derived from the α spectra and γ spectra, are shown in Table 6. The purity of the obtained solution became high (99.5 at% over actinides), though the amount of Cm decreased in handling the samples in this procedure.
Table 6

The weight (mg) and the atomic ratio of the nuclides contained in the Cm solution after Cm/Am separation derived from the α spectra and γ spectra

 

244Cm

245Cm

246Cm

238Pu

240Pu

241Am

243Am

Cm solution after Cm/Am separation

 Weight (mg)

6.398

0.284

0.337a

0.00005b

0.027

0.0001

0.0058

 Atomic ratio

0.908

0.040

0.047

0.000

0.004

0.000

0.001

aRadioactivity of 246Cm were calculated from that of 244Cm using their ratio in Cm fraction solution of Pu removal run 1

bRadioactivity of 238Pu were calculated from that of 240Pu using their average ratio in Pu fraction solution (1) and Pu fraction solution (2) of Pu removal run 2

Recovery of Cm by oxalate precipitation method

An oxalate precipitation method was adopted to recover curium in a solid form. The conditions of the precipitation were set to reduce the loss of Cm in the solutions.

Experimental

Two sets of the experiments were carried out; their conditions are listed in Table 7. As in Table 7, (1) the solution obtained by run 1 of the Pu removal and (2) the solution purified with the Pu removal and Am removal were used for the experiments. Oxalate precipitates were obtained in the solution of HNO3-oxalic acid of which compositions were adjusted by adding 0.5 M oxalic acid and pure water; the latter was used only in run 1. The solution coexistent with the precipitates was separated and the precipitate was washed to decrease the amount of the nitric acid coexistent with the precipitate to avoid causing harms on the furnace when the oxalates are converted to oxides. The washing procedure was adding 5 mL of washing liquids in the container having the precipitate followed by removing the solution with a mechanical pipet after the precipitate was settled on the bottom of the container. After repeating the washing procedures using the liquids shown in Table 7, the precipitate was dried by heating with a hot plate. A part of the dried precipitate sample was dissolved in 1 M HNO3 to be used for the composition analyses. Composition analyses of the samples were carried out as described earlier.
Table 7

The condition and the results of the recovery of Cm by oxalate precipitation method

 

Run 1

Run 2

Cm solution

Pu removed

Pu removed, Am removed

 

6.6 M HNO3, 49.95 mL containing 9.3 mg of Cm

4 M HNO3, 2 mL containing 7.0 mg of Cm

Condition of precipitation

0.25 M HNO3–0.1 M oxalic acid, 1,330 mL

1 M HNO3–0.375 M oxalic acid, 8 mL

Solubility of Cm(III)[19]

0.8 mg/L

6 mg/L

Estimated amounts of Cm(III) in the solution

1.06 mg

0.048 mg

Measured amounts of Cm(III) in the solution

0.896 mg

0.041 mg

Ratio of Cm recovered in the precipitate

90.5 %

99.4 %

Washing liquids

Water, 3 times

Water/0.05 M oxalic acid, twice/water

The weight of the precipitates

11.9 mg

12.0 mg

Results and discussion

The weight, and the atomic ratio over the actinides in the solution coexistent with the precipitates and the solution dissolving the oxalate precipitate, derived from the α spectra and γ spectra for run 1 and run 2 are listed in Table 8 and Table 9, respectively. The amounts of Cm(III) in the solution coexistent with the precipitates calculated from the solubility data of Cm(III) in HNO3 - oxalic acid systems [19] shown in Table 7 are close to the measured total amount of Cm (the sum of 244Cm, 245Cm, and 246Cm) in the solution coexistent with the precipitates. The recovery yields of Cm from the Cm solutions into the oxalate for each run, derived from the amounts of Cm in each solution were 90.5 % and 99.4 %, respectively.
Table 8

The weight (mg) of the nuclides contained in the solution coexistent with the precipitate (1) and (2), and the solution in which Cm oxalate sample was dissolved derived from the α spectra and γ spectra in recovery of Cm experiment (run 1)

 

244Cm

245Cm

246Cm

238Pu

240Pu

241Am

243Am

The solution coexistent with the precipitate (1) (830 mL)a

 Weight (mg)

0.595

0.031

0.031

0.000045b

0.0231

0.00021

0.048

 Atomic ratio

0.8169

0.0423

0.0418

0.0001

0.0322

0.0003

0.0635

The solution coexistent with the precipitate (2) (500 mL)a

 Weight (mg)

0.219

0.009

0.011

0.000027b

0.0139

0.00018

0.016

 Atomic ratio

0.8129

0.0346

0.0416

0.0001

0.0524

0.0007

0.0578

Cm oxalate(10.9 mg)

 Weight (mg)

8.257

0.362

0.481

0.000491b

0.0253

0.00013

0.526

 Atomic ratio

0.8561

0.0373

0.0493

0.0000

0.0026

0.0006

0.0545

a The precipitates were prepared by repeating the procedure, in which 1/3 quantity of the solutions were mixed to make the precipitates and the solution coexistent with the precipitates was removed after the precipitate was settled on the bottom of the container, 3 times. The solution removed after the first and the second mixing of the solutions and that after the third mixing were recovered separately; the solutions were named the solution coexistent with the precipitate (1) and (2)

b The weight of 238Pu were calculated from that of 240Pu using their ratio in Pu fraction solution (1) of Pu removal run 1

Table 9

The weight (mg) and the atomic ratio of the nuclides contained in the solution coexistent with the precipitate, and the solution in which Cm oxalate sample was dissolved derived from the α spectra and γ spectra in recovery of Cm experiment (run 2)

 

244Cm

245Cm

246Cm

238Pu

240Pu

241Am

243Am

The solution coexistent with the precipitate (8 mL)

 Weight (mg)

0.03764

0.00182

0.00198a

ND

ND

0.00000

0.00004

 Atomic ratio

0.908

0.044

0.047

ND

ND

0.000

0.001

Cm oxalate(12.0 mg)

 Weight (mg)

6.204

0.263

0.327a

0.00001b

0.0058

0.00079

0.00795

 Atomic ratio

0.912

0.039

0.048

0.000

0.001

0.000

0.001

aRadioactivity of 246Cm were calculated from that of 244Cm using their ratio in Cm fraction solution

bRadioactivity of 238Pu were calculated from that of 240Pu using their average ratio in Pu fraction solution (1) and Pu fraction solution (2) of Pu removal run 2

The Cm oxalate precipitate obtained in run 1 (11.9 mg) has impurities of 243Am (5.4 at%) and 240Pu (0.3 at%). Pu content in the oxalate was lower than that in the Cm solution. From the fact that the atomic ratio over the actinides of the solution coexistent with the precipitate was almost the same as that of the Cm solution, it was considered that a part of the Pu contained in the oxalate was removed during the washing procedure. On the other hand, 243Am impurity originally contained in the Cm–Pu oxide was not removed by this procedure. As shown in Table 9, the purity of Cm oxalate sample obtained in run 2 (12.0 mg) was 99.8 % and it has impurities of 243Am (0.1 at%) and 240Pu (0.1 at%). Both of the obtained Cm oxalate samples were supplied for the syntheses and measurements of the thermochemical properties of Cm compounds.

Conclusion

Curium was separated and recovered as an oxalate from the Cm–Pu mixed oxide which had been a 244Cm oxide sample prepared more than 40 years ago; the ratio of 244Cm to 240Pu was estimated to 0.2:0.8. Radiochemical analyses of the solution prepared by dissolving Cm–Pu mixed oxide in nitric acid revealed that the Cm–Pu mixed oxide contained about 1 at% of Am impurity. To obtain high purity curium solution, plutonium was removed by using an anion exchange resin column and americium was removed by the chromatographic separation using tertiary pyridine resin embedded in silica beads using nitric acid/methanol mixed solution. Curium oxalate, a precursor compound of curium oxide, was prepared from the purified curium solution. 11.9 mg of Cm oxalate having some amounts of impurities, which are 243Am (5.4 at%) and 240Pu (0.3 at%) was also obtained without Am removal procedure. Meanwhile, 12.0 mg of Cm oxalate sample (99.8 at% over actinides) having small amounts of impurities, which are 243Am (0.1 at%) and 240Pu (0.1 at%), was prepared with Am removal procedure. Both of the Cm oxalate sample were supplied for the syntheses and measurements of the thermochemical properties of curium compounds.

Acknowledgments

This study contains the result of “Basic actinide chemistry and physics research in close cooperation with hot laboratories” carried out under the Strategic Promotion Program for Basic Nuclear Research by the Ministry of Education, Culture, Sports, Science and Technology of Japan. The authors acknowledge Prof. T. Suzuki (Tokyo Institute of Technology, present affiliation is Nagaoka University of Technology) for providing the tertiary pyridine resin. The authors acknowledge Messrs. S. Tagami, A. Itoh, and H. Kato for their technical supports.

Copyright information

© Akadémiai Kiadó, Budapest, Hungary 2012

Authors and Affiliations

  • Hirokazu Hayashi
    • 1
  • Hiromichi Hagiya
    • 1
  • Seong-Yun Kim
    • 1
    • 2
  • Yasuji Morita
    • 1
    • 3
  • Mitsuo Akabori
    • 1
  • Kazuo Minato
    • 1
    • 3
  1. 1.Nuclear Science and Engineering DirectorateJapan Atomic Energy AgencyTokai-mura, Naka-gunJapan
  2. 2.Cyclotron and Radioisotope CenterTohoku UniversityAramaki Aoba-kuJapan
  3. 3.Tokai Research and Development CenterJapan Atomic Energy AgencyTokai-mura, Naka-gunJapan

Personalised recommendations