Journal of Radioanalytical and Nuclear Chemistry

, Volume 322, Issue 3, pp 1273–1277 | Cite as

Improvement in flow-sheet of extraction chromatography for trivalent minor actinides recovery

  • Sou WatanabeEmail author
  • Tatsuya Senzaki
  • Atsuhiro Shibata
  • Kazunori Nomura
  • Masayuki Takeuchi
  • Kiyoharu Nakatani
  • Haruaki Matsuura
  • Yusuke Horiuchi
  • Tsuyoshi Arai


Extraction chromatography flow-sheet employing octyl(phenyl)-N,N-diisobutylcarbonoylmethylphosphine oxide and bis(2-ethylhexyl) hydrogen phosphate extractants for trivalent minor actinide recovery was modified to improve column separation performance. Excellent trivalent minor actinides recovery performance was obtained by column separation experiments on nitric acid solution containing the trivalent minor actinides and representative fission product elements, i.e. recovery yields > 93% with sufficient decontamination factors against the fission products. Those are the best performance which we have ever obtained by experiments inside hot cell.


Extraction chromatography Trivalent minor actinides Reprocessing CMPO HDEHP 


Partitioning and transmutation of minor actinides (MA: Np, Am, Cm) in spent nuclear fuel is essential strategy for reduction in volume and radio toxicity of nuclear wastes [1]. Japan Atomic Energy Agency has been proposing NEXT process (Fig. 1 [2]) for reprocessing of spent fast reactor fuels. In this process, Np can be recovered with U and Pu by solvent extraction with modified PUREX process [3]. Trivalent minor actinides (MA(III): Am and Cm) have to be recovered from raffinate of the solvent extraction process in order to achieve the all minor actinides partitioning. Extraction chromatography technology was employed as the selective MA(III) recovery in the NEXT process, and systematic development of this technology has been started in 2006 [4].
Fig. 1

Flow sheet of MA(III) recovery by extraction chromatography

The extraction chromatography utilizes packed columns with porous silica based adsorbents, and MA(III) is possible to be separated from other fission products through adsorption/desorption reactions. Fundamental studies to optimize structure of the adsorbents [5, 6, 7], optimization in process flow for efficient MA(III) recovery [8, 9, 10] and development of engineering scale devices [11, 12, 13] have been carried out to adopt this technology for future reprocessing plants. One of the difficulties for the implementation is efficient MA(III) separation performance from lanthanides (Ln) which show similar chemical behavior with MA(III), and design of a promising flow-sheet is still under development.

One of the promising flow sheets which consists of two steps column operation is shown in Fig. 1. In the first column of this flow-sheet, octyl(phenyl)-N,N-diisobutylcarbonoylmethylphosphine oxide (CMPO) [14] impregnated adsorbent (referred as CMPO/SiO2-P) is used for decontamination of light lanthanides (lLn: La–Nd) and representative fission products (FP: Cs, Sr, Mo, Zr and Platinum group elements) in high level liquid waste (HLLW) generated by the previous step in the NEXT process i.e. solvent extraction for U/Pu/Np co-recovery. MA(III) and all lanthanides in nitric acid medium are loaded into the CMPO/SiO2-P, and then MA(III) and heavy lanthanides (hLn: Sm–Gd, Y) are targeted to be eluted faster than lLn into diethylenetriaminepentaacetic acid (DTPA) solution utilizing the differences on affinity of those elements to DTPA. This concept is initially designed for solvent extraction process named as SETFICS which has developed in Japan Atomic Energy Agency [15]. Bis(2-ethylhexyl) hydrogen phosphate (HDEHP) [16] impregnated adsorbent (referred as HDEHP/SiO2-P) is used in the 2nd column, and MA(III) + hLn product of the 1st column is supplied into the column after adjusting acidity into 0.3 M by adding HNO3. MA(III) and hLn loaded in the 2nd column are separately eluted into nitric acid eluent with different retention times. The MA(III)/hLn separation can be achieved owing to differences on affinity of those elements to HDEHP [17]. One of the notable advantages of this flow sheet is that MA(III) is separated from other fission products and recovered in nitric acid medium without containing any complex reagents by simple column operations. Therefore, practical process which does not requires complicated apparatus and additional procedures for purification of MA(III) product is expected to be achieved.

An experimental trial of the flow-sheet has been carried out in the previous study [10], and improvements in flow-sheet were shown to be necessary to achieve target MA(III) recovery performance i.e. MA(III) recovery yields > 99% and decontamination factors (DF) of fission products > 102. Tailings of the MA(III) elution curves overlapped on those of lLn or hLn for CMPO/SiO2-P or HDEHP/SiO2-P columns, respectively, and appropriate process or operational conditions to enhance the separation performance should be discussed. Fundamental studies for the improvement of the performance were systematically carried out, and the modification of the flow-sheet was carried out focusing on acidities of eluents. In this study, MA(III) separation performance of the modified flow-sheet was evaluated though hot experiments on MA(III) containing solution.


Porous silica particles with average diameters of 50 μm and average pore diameters of 50 nm were prepared by sol–gel method. Those particles were coated by styrene divinyl benzene copolymer, and CMPO or HDEHP was impregnated into the polymer. Preparation of the adsorbent was carried out based on an article [18]. Degree of cross linkage of the polymer was 15%, and amount of the extractant impregnated was 33 wt% of the total adsorbent.

Column separation experiments were carried out in a hot cell of Chemical Processing Facility (CPF) of Nuclear Fuel Cycle Engineering Laboratories in Japan Atomic Energy Agency. Photo of experimental setup designed for operation through remote handling devices is shown in Fig. 2, where it is consisted of a glass column with 1 cm diameter and 32.5 cm height (1 Bed Volume (BV) = 25.5 cm3), a pump, a pressure gauge and panels with valves for switching solutions for the mobile phase. The adsorbent was packed in the glass column, and temperature of the column was controlled by supplying temperature controlled water into water jacket surrounding the column.
Fig. 2

Experimental setup of the extraction chromatography experiments in hot cell

Flow-sheets of the 2 steps column operation before and after the modification are shown in Table 1. For both columns, flow rate of the mobile phase was decreased to enhance the separation performance. pH of DTPA solution for the eluent of the 1st column was changed from 3 to 2, where separation of MA(III) from lLn is expected to be improved. Acidity of HNO3 solution for the eluent of the 2nd column was also changed from 1 to 0.8 M for better separation of MA(III) from hLn.
Table 1

Experimental conditions of the column separation operations


Original flow [10]

Modified flow

Flow rate

4 cm/min

2 cm/min



HLLW ([H+] = 5 M)

HLLW ([H+] = 5 M)



5 M HNO3: 5BV

5 M HNO3: 5BV

T = 323 K


50 mM DTPA

50 mM DTPA


(pH = 3): 10 BV

(pH = 2): 10 BV


T = 298 K


MA(III) + hLn(III) product of CMPO column ([H+] = 0.3 M)

MA(III) + hLn(III) product of CMPO column ([H+] = 0.3 M)


0.3 M HNO3: 3 BV

0.3 M HNO3: 3 BV


1 M HNO3: 10 BV

0.8 M HNO3: 10 BV

The feed solution, a wash solution and an eluent were sequentially supplied into the column, and effluent was fractionally collected at every 1 BV. Composition of the feed solution for the 1st column is shown in Table 2, where several fission products were added in the feed solution in order to evaluate separation performance of MA(III). Concentrations of radioactive elements in the fractions were analyzed by alpha and gamma measurements, and Nd in the fraction was analyzed by ICP-AES measurements. Decontamination factor (DF) of fission products were calculated by the following equation,
$${\text{DF}} = \frac{{C_{\text{M,F}} V_{\text{F}} }}{{\sum\nolimits_{i = m}^{n} {C_{\text{M,i}} V_{\text{i}} } }}$$
where CM,i and, CM,F are the concentrations of metal M in i th fraction and feed solution, respectively. Vi and VF are volumes of effluent in i th fraction and the feed solution, respectively.
Table 2

Composition of the feed solution for the first column




3.9 × 107 Bq/mL


6.3 × 105 Bq/mL


9.4 × 104 Bq/mL


1.1 × 104 Bq/mL


310 ppm


5 M

Results and discussion

CMPO/SiO2-P column

Elution curves obtained by CMPO/SiO2-P column operation is shown in Fig. 3, where C0 and C correspond to concentrations of metals in the feed solution and the effluent, respectively. 137Cs was discharged from the column with the feed solution and initial part of the wash solution, and which reasonably agrees with behavior observed in previous studies [8, 9, 10]. Other fission products such as Sr and Pd are expected to show similar elution curves since those elements have little affinity to CMPO. After supplying the DTPA eluent, MA(III) and Eu were eluted before Nd elution. This result reasonably agrees with those observed in the previous study [10]. Some fission product elements which show relatively smaller affinity than MA(III) to CMPO such as Sb and Ru are expected to be eluted before the MA(III) elution according to our previous data [8, 9]. Therefore, MA(III) and hLn are selectively recovered in an interim product if volume range for the product is appropriately selected. The volume range of the product should be taken from V = 12 BV according to beginning of the MA(III) elution. In order to achieve high decontamination factor against Nd, right border of the product has to be set before the beginning of the Nd elution. However, the right border is desirable to be located at rightward wherever possible in the respect of the recovery yields of MA(III). Therefore, performance of the 1st column inevitably depends on trade-off between decontamination factor of lLn and recovery yields of MA(III).
Fig. 3

Elution curves from CMPO/SiO2-P column

Recovery yields of MA(III) (RYMA(III)) and decontamination factor of Nd (DFNd) plotted as position of right border of the interim product are shown in Fig. 4, where left border of the product was fixed at V = 12 BV. RYMA(III) increased with the position of right border and reached more than 99% at V > 15 BV. DFNd was more than 103 at V < 13.5 BV and decreased to be lower than 5 at V > 15 BV. DFNd > 102 with RYMA(III) = 95% were achieved when the right border was set to be 14 BV, and those performance was improved by the modification in the flow-sheet (i.e. DFNd = 102 with RYMA(III) = 90% in the previous study). RYMA(III) can be increased up to 99% if DFNd = 17 is acceptable. In this study, volume range of 12 < V < 14 BV was employed for the interim product with prioritizing DFNd over RYMA(III), where RYAm, RYCm and DFNd were 94.6, 98.8% and 1.4 × 102, respectively.
Fig. 4

RYMA(III) and DFNd plotted as a function of volume range of the interim product

HDEHP/SiO2-P column

Elution curves obtained by HDEHP/SiO2-P column operation is shown in Fig. 5. MA(III) were immediately eluted from the column after supplying the HNO3 eluent and properly separated from Eu. MA(III)/hLn separation behavior was drastically improved by the modification of the flow-sheet. Since Nd eluted with MA(III), further decontamination of lLn by the 2nd column cannot be expected. MA(III) recovery performance of the 2nd column also has trade-off relation between RYMA(III) and DFEu. Range of the volume for the final product should begin with V = 5.5 BV, and sufficient RYMA(III) with large DFEu is expected to be achieved if right border of the volume range is appropriately determined.
Fig. 5

Elution curves from HDEHP/SiO2-P column

RYMA(III) and DFEu plotted as position of right border of the final product are shown in Fig. 6, where left border of the product was fixed at V = 5.5 BV. RYMA(III) sharply increased with the volume and reached more than 98% at V = 7 BV, and DFEu rapidly decreased with the volume to be below 30 at V = 8 BV. The right border was decided to be V = 7.5 BV to achieve high RYMA(III) and DFEu simultaneously, where RYAm, RYCm and DFEu were 98.3, 98.4% and more than 103, respectively.
Fig. 6

RYMA(III) and DFEu plotted as a function of volume range of the final product

Recovery yields of MA(III) and decontamination factors of FP through the 2steps operation are summarized in Table 3. RY of Am and Cm were significantly improved from previous study (RYAm = 68.8% and RYCm = 62.9%) with satisfying target decontamination factors (more than 102 for each element). In this study, half-length columns of the engineering scale system were used due to operational limitation inside the hot cell of CPF, therefore the column performance is expected to be improved with the large scale device. Consequently, the modified flow-sheet must be the most practical one in the respects of the excellent MA(III) recovery performance and of the desirable composition of the final MA(III) product solution without any complexing reagents. Further optimization in operation conditions for practical use should be carried out using the engineering scale devices.
Table 3

Summary of column separation performance





















> 103

> 103


1.00 × 100

> 103

> 103


1.4 × 102

1.0 × 100

1.4 × 102


Optimization in operational conditions of extraction chromatography process for trivalent minor actinides (MA(III)) recovery using CMPO/SiO2-P and HDEHP/SiO2-P packed columns were carried out, and performance of the process was demonstrated by column separation experiments inside a hot cell. Modification in flow-rate of the mobile phase and compositions of eluents for both columns achieved significant improvements in MA(III) recovery performance. MA(III) were selectively recovered in nitric acid solution without containing complex reagent with 93.0 and 97.2% of recovery yields for Am and Cm, respectively, and decontamination factors of representative fission products were larger than target value i.e. 100. The flow-sheet developed by this study would be one of the most practical ones, and demonstration with engineering scale devices is necessary for the practical use.



  1. 1.
    International Atomic Energy Agency (2004) Technical reports series no. 435; implications of partitioning and transmutation in radioactive waste management. IAEA, ViennaGoogle Scholar
  2. 2.
    Funasaka H, Itoh M, Proceedings of Global 2007, September 9–13, 2007, Boise, Idaho, USA (2007)Google Scholar
  3. 3.
    Nakahara M, Sano Y, Koma Y, Kamiya M, Shibata A, Koizumi T, Koyama T (2007) Separation of actinide elements by solvent extraction using centrifugal contactors in the NEXT process. J Nucl Sci Technol 44(3):373–381CrossRefGoogle Scholar
  4. 4.
    Koma Y, Sano Y, Nomura K, Watanabe S, Matsumura T, Morita Y (2010) Development of the extraction chromatography system for separation of americium and curium. In: Proceedings of the OECD Nuclear Energy Agency 11th information exchange meeting on actinide and fission product partitioning and transmutation, San Francisco, USA, November 1–4, 2010, IV-4, OECD/NEAGoogle Scholar
  5. 5.
    Watanabe S, Arai T, Ogawa T, Takizawa M, Sano K, Nomura K, Koma Y (2012) Optimization composition of TODGA/SiO2-P adsorbent for extraction chromaotgraphy process. Proc Chem 7:411–417CrossRefGoogle Scholar
  6. 6.
    Watanabe S, Sano Y, Shiwaku H, Yaita T, Ohno S, Arai T, Matsuura H, Koka M, Satoh T (2017) Local structure and distribution of remaining elements inside extraction chromatography adsorbents. Nucl Instrum Methods Phys Res B 404:202–206CrossRefGoogle Scholar
  7. 7.
    Sano Y, Watanabe S, Matsuura H, Nagoshi K, Arai T (2017) Microanalysis of silica-based adsorbent for effective recovery of radioactive elements. J Nucl Sci Technol 54(10):1058–1064CrossRefGoogle Scholar
  8. 8.
    Watanabe S, Senzaki T, Shibata A, Nomura K, Koma Y, Nakajima Y (2011) MA recovery experiments from genuine HLLW by extraction chromatography. In: Proceedings of global 2011, Makuhari, Japan, December 11–16, 2011, Paper 387433 [CD-ROM]Google Scholar
  9. 9.
    Watanabe S, Nomura K, Kitawaki S, Shibata A, Kofuji H, Sano Y, Takeuchi M (2016) Flow-sheet study of MA recovery by extraction chromatography for SmART cycle project. Proc Chem 21:101–108CrossRefGoogle Scholar
  10. 10.
    Watanabe S, Sano Y, Kofuji H, Takeuchi M, Shibata A, Nomura K (2018) Am, Cm reocvery from genuine HLLW by extraction chromatography. J Radioanal Nucl Chem 316:1113–1117CrossRefGoogle Scholar
  11. 11.
    Watanabe S, Goto I, Sano Y, Koma Y (2010) Chromatography column system with controlled flow and temperature for engineering scale application. J Eng Gas Turbines Power 132(10):102903CrossRefGoogle Scholar
  12. 12.
    Watanabe S, Goto I, Nomura K, Sano Y, Koma Y (2011) Extraction chromatography experiments on repeated operation using engineering scale column system. Energy Proc 7:449–453CrossRefGoogle Scholar
  13. 13.
    Watanabe S, Sano Y, Nomura K, Koma Y, Okamoto Y (2015) Safety operation of chromatography column system with discharging hydrogen radiolytically generated. EPJ Nucl Sci Technol 1:9CrossRefGoogle Scholar
  14. 14.
    Horwitz EP, Kalina DG, Diamond H, Vandegrift GF, Schulz WW (1985) The TRUEX process—a process for the extraction of the transuranic elements from nitric acid wastes utilizing modified PUREX solvent. Solv Extr Ion Exch 3(1&2):75–109CrossRefGoogle Scholar
  15. 15.
    Koma Y, Watanabe M, Nemoto S, Tanaka Y (1998) A counter current experiment for the separation of trivalent actinides and lanthanides by the setfics process. Solv Extr Ion Exch 16(6):1357–1367CrossRefGoogle Scholar
  16. 16.
    Weaver B, Kappelmann FA (1964) TALSPEAK: a new method of separating americium and curium from the lanthanides by extraction from an aqueous solution of an aminopolyacetic acid complex with a monoacidic organophosphate or phosphonate. In: ORNL-3559Google Scholar
  17. 17.
    Savantesson I, Persson G, Hagstrӧm I, Liljenzin JO (1980) Distribution ratios and empirical equations for the extraction of elements in purex high level waste solution-II: HDEHP. J Inorg Nucl Chem 42:1037–1043CrossRefGoogle Scholar
  18. 18.
    Wei Y, Kumagai M, Takashima Y (2000) Studies on the separation of minor actinides from high-level wastes by extraction chromatography using novel silica-based extraction resins. Nucl Technol 132:413CrossRefGoogle Scholar

Copyright information

© Akadémiai Kiadó, Budapest, Hungary 2019

Authors and Affiliations

  • Sou Watanabe
    • 1
    Email author
  • Tatsuya Senzaki
    • 1
  • Atsuhiro Shibata
    • 1
  • Kazunori Nomura
    • 1
  • Masayuki Takeuchi
    • 1
  • Kiyoharu Nakatani
    • 2
  • Haruaki Matsuura
    • 3
  • Yusuke Horiuchi
    • 4
  • Tsuyoshi Arai
    • 4
  1. 1.Japan Atomic Energy AgencyTokai-muraJapan
  2. 2.University of TsukubaTsukubaJapan
  3. 3.Tokyo City UniversitySetagayakuJapan
  4. 4.Shibaura Institute of TechnologyKoutoukuJapan

Personalised recommendations