Journal of Radioanalytical and Nuclear Chemistry

, Volume 298, Issue 2, pp 1309–1314

Separation of bulk Y from 89Y(n,p) produced 89Sr by extraction chromatography using TBP coated XAD-4 resin


  • Debasish Saha
    • Fuel Chemistry Group, Chemistry GroupIndira Gandhi Centre for Atomic Research
  • E. Senthil Vadivu
    • Fuel Chemistry Group, Chemistry GroupIndira Gandhi Centre for Atomic Research
  • R. Kumar
    • Fuel Chemistry Group, Chemistry GroupIndira Gandhi Centre for Atomic Research
    • Fuel Chemistry Group, Chemistry GroupIndira Gandhi Centre for Atomic Research

DOI: 10.1007/s10967-013-2514-y

Cite this article as:
Saha, D., Senthil Vadivu, E., Kumar, R. et al. J Radioanal Nucl Chem (2013) 298: 1309. doi:10.1007/s10967-013-2514-y


89Sr was produced using the 89Y(n, p) 89Sr reaction by irradiating yttria target in the fast breeder test reactor (FBTR). An analytical scale procedure was standardized for the removal of the bulk target yttrium by solvent extraction using the tri-n-butyl phosphate (TBP). The final purification of 89Sr source was carried out by ion exchange chromatography. However, extraction chromatography is preferred to solvent extraction due to its low waste generation and ease of operation. This paper reports the separation of Sr from bulk Y and other radioactive impurities produced during irradiation by extraction chromatography using TBP coated XAD-4 resin. Initially the separation procedure was standardized using 85Sr and 88Y tracers. The 89Sr in the dissolver solution of the irradiated yttria target was separated under the same standardized conditions. The study established the separation of Sr from Y in the dissolver solution of the irradiated target yttria by the TBP coated XAD-4 column. However the evaluation of the column after every use for about three separation studies exhibited the reduction in its breakthrough capacity for yttrium.


TBP/XAD-489SrExtraction chromatographyYttriumYttrium separationFast breeder test reactor (FBTR)Bone pain palliationBreakthrough capacity



Distribution coefficient


Initial activity in solution


Final activity in solution


Volume of the solution


89Sr is used for the treatment of bone pain palliation in metastatic bone cancer patients. It is produced in fast reactors using the 89Y(n,p) 89Sr reaction. Feasibility studies for the production of this radionuclide using the fast breeder test reactor (FBTR) at Kalpakkam are in progress. Standardization of analytical scale procedure for the chemical processing of the irradiated yttria target involved solvent extraction technique using tri-n-butyl phosphate (TBP) for separation of the bulk yttrium and purification of the 89Sr by ion exchange chromatography using DOWEX resin [1, 2]. A major drawback in this production route is the production of 88Y due to the 89Y(n, 2n) reaction which results in high dose and consequently use of shielded facilities for the chemical processing. The extraction chromatographic mode of separation offers a number of advantages [39] over solvent extraction mode of separation. It is the most effective way of achieving efficient separation whenever multi-stage extractions are required. It may be appropriate in this application where about three extractions by TBP are needed for the maximum removal of the bulk yttrium. Relatively minimum extractant can be used for a separation by this mode that helps in reducing the generation of secondary radioactive wastes. In addition, column operations are easier to handle in shielded facilities. In view of these advantages, removal of the bulk yttrium by TBP was investigated by the extraction chromatographic technique in this study.

Louis et al. and Kimura et al. [6, 10] reported the separation of lanthanides and actinides with TBP/XAD-4 resin. They have studied the distribution ratio of Eu(III) and Am(III) as a function of various loadings of TBP, equilibration time, storage time and also studied the effect of external salting out agent, and the various other factors affecting the resolution of the mixture of Eu, Am and U, Th [6, 7, 1012]. Aardaneh et al. [13] separated carrier free 88Y from bulk Sr, which is produced by proton-induced nuclear reaction i.e. 88Sr (p, n) 88Y. Conc. HNO3 was used by these authors but they were unable to obtain complete separation of Sr and Y. They recovered only about 90 % of the 88Y produced with the rest being lost in the Sr fractions. In the present study, XAD-4 [14, 15] was selected as the inert support. The extractant TBP was coated onto the inert matrix XAD-4 to study the separation behavior of target yttrium in the dissolver solution of irradiated yttria pellets by extraction chromatography mode [12].

The study was carried out in order to (i) evaluate the efficiency of separation for the bulk Y target material irradiated in FBTR from 89Sr and other radionuclides and also (ii) assess its potential application for the substitution of solvent extraction mode of separation of Y using TBP in the dissolver solution.


Reagents and equipments

All chemicals used were of analytical grade unless otherwise mentioned. The inert matrix used in this study, Amberlite XAD-4 (20–60 mesh size) was obtained from M/S Supelco, USA. A measured quantity of XAD-4 was equilibrated with distilled water in a platform shaker for an hour. This process was repeated thrice. The resultant XAD-4 was washed with isopropanol thrice using the same procedure and finally washed with acetone. After removing acetone, the XAD-4 was dried in an oven at 110 °C for 24 h. The final material obtained after wash was 60 % of its initial weight. Tri-n-butyl phosphate (98 %) was obtained from Alfa Aesar, Karlsruhe, Germany. Nitric Acid was obtained from Loba Chemie, India. 88Y and 85Sr tracers were procured from CERCA LEA, France. Distilled water was used for all the experiments. Radiochemical assay was carried out by high resolution γ-spectrometry using HPGe detector.

Impregnation of TBP on XAD-4

Impregnation of TBP on XAD-4 resin was carried out by taking XAD-4: TBP in the ratio of 2:3 (w/v) [15] using the process described elsewhere [16]. In a typical experiment, 10 mL of water soaked XAD-4 (10 g) and 15 mL of TBP was taken in three glass bottles (i) without any diluent (ii) with methanol and (iii) n-Hexane diluent. The mixture was equilibrated at room temperature for 24 h using a platform shaker of PISCES make. The residual mass was dried in an oven at 110 °C for 24 h to obtain finally the free-flowing TBP impregnated XAD-4 resin. The complete loading of the TBP onto the resin was confirmed from the observation that the weight of the completely dried TBP loaded XAD-4 resin was equal to the sum of the weights of TBP and XAD-4 resin taken initially.

Determination of distribution ratio for Sr and Y

Typically, about 50 mg of dry TBP/XAD-4 resin was taken in 10 mL of different acidities i.e. pH 3, 1 M, 3 M, 6 M, 9 M, 12 M, 14 M and 16 M of nitric acid and known amount of 85Sr and 88Y tracers added. The mixtures were equilibrated using a rotary shaker (4 h and 50 rpm speed). A few minutes were allowed to enable the resin to settle down after shaking. 5 mL of the aqueous phase was sampled out for the assay by high resolution γ-spectrometry using HPGe detector.

Standardization of the separation procedure using 85Sr and 88Y tracers

1.6 g of the XAD-4 resin coated with undiluted TBP was taken in a glass column [6 mm (ID) × 190 mm (height)]. The column was washed with nitric acid of pH 3 solution and subsequently conditioned with conc. HNO3. The tracer solutions of 85Sr and 88Y in conc. HNO3 were added to the column. The washing of Sr and elution of Y were carried out using conc. HNO3 and pH 3 nitric acid solutions respectively. Sr in the column was washed using 20 column volumes i.e. up to about 70 mL of conc. HNO3 and the loaded Y in the column was eluted using an equal amount of pH 3 nitric acid solution. The flow rate in the range of 0.5–1 mL/min and the sample fractions collected every time of the volume of 5 mL were maintained during the elution. The constant sample size of 5 mL was preferred over the constant period of sample collections. This is to maintain the uniformity in the sample size despite any possible small variation in the flow rate during the elution. These samples were subsequently analyzed by gamma spectrometry. In order to evaluate the reproducibility of the separation with repeated usage of the column, the experiment was repeated three times under similar conditions. The time intervals between these trials were around 12 h.

Breakthrough capacity measurement

The breakthrough capacity for Y was measured using conc. HNO3 in the same column with the same flow rates as described earlier. Yttrium stock solution of 0.157 mg/mL was prepared in conc. HNO3 with 88Y tracer added to it. The solution was passed through the column at a flow rate of ~1 mL/min. 5 mL fractions were collected and assayed by gamma spectrometry. To investigate the performance of the resin after repeated use, experiments were conducted using freshly prepared resin as well as resin that had been used for three separations.

Purification of Sr from bulk Y in the dissolver solution

Yttria powder was pelletized and sintered using the sintering aid, ZnO. 77 g of the sintered pellets were irradiated in an SS tube in FBTR for 73 days. After the irradiation, these pellets were transferred to the hot cell facility at Chemistry Group. An irradiated yttria pellet of about 1 g was removed from this sample and dissolved in concentrated nitric acid i.e. ~69 wt% under reflux condition in a fume hood. However for the investigation on the separation behavior of Y and Sr in the present study using TBP coated XAD-4 resin by extraction chromatographic separation, a sample of the dissolver solution was used.

An irradiated Y2O3 pellet of about 1 g was dissolved in conc. HNO3 and used for this study. The solution contained 89Sr, 88Y and other radioactive impurities produced during the irradiation of the yttria pellet i.e. 58Co, 65Zn and 155Eu. In order to trace the Sr profile by gamma spectrometry, the tracer 85Sr was added externally as 89Sr is a pure beta emitter. 60Co was also added to the dissolver solution as the activity from 58Co produced in the irradiated pellet was too low to assay. The above solution was loaded onto the chromatographic column which had been earlier packed with TBP coated XAD-4 resin in water medium and conditioned using conc. HNO3. Sr and other radioisotopes had shown no affinity to the column in conc. HNO3 solution and were eluted in the initial fractions. The elution of Y was carried out using nitric acid solution of pH 3. The fractions of 5 mL size were collected at regular intervals during the elution and assayed subsequently by gamma spectrometry.

Results and discussion

Determination of distribution ratio for Sr and Y

The distribution ratio (D) was calculated based on the following equation-
$$ D = \left( {\frac{{C_{0} - C_{i} }}{{C_{i} }}} \right) \times \left( \frac{v}{w} \right) $$
where C0 and Ci are the initial and final total activity in the solution, v the volume of the solution (mL) and w, the weight of the resin taken (g) [17, 18].

The variation of D for yttrium with the acidity for the XAD-4 resin coated with TBP in different media such as no diluent, MeOH and n-hexane followed similar trend even though the difference in D is slightly large in conc. HNO3 medium for the resin coated with TBP in no diluent medium compared to the resin coated methanol and hexane media i.e. D values for the resin without diluent, methanol and hexane diluent in conc. HNO3 medium were 70, 35 and 25 respectively. However the resin used throughout this study is the one in which the extractant TBP was coated without diluent.

The D of Y and Sr as a function of acidity for the XAD-4 resin coated with TBP in water medium is shown (Fig. 1).
Fig. 1

Distribution ratio measurement at various [HNO3] with XAD-4 resin coated with TBP in water medium. (Experimental conditions: TBP coated resin 50 mg; volume of different acid solutions 10 mL, equilibration in rotospin, speed 50 rpm, time 4 h; assay gamma spectrometry)

The data were plotted from two trials with error bar and in case of close values, the height of the error bar was negligible. The low D of Y increases negligibly up to 10 M HNO3 and starts increasing slowly up to 14 M but increases sharply to reach its high value at conc. HNO3.

However D of Sr remains negligible up to 14 M HNO3 while it increases relatively only slightly at conc. HNO3.The trend suggests that (i) the loading of Y at higher acidity in the column results in the washing of Sr while Y is held up in the column and (ii) Y can be eluted at lower acidity subsequently. Since the D value of Sr at higher acidity is not zero, some mixing of Sr and Y may be observed during the separation.

Separation profiles of Sr and Y using their tracers

The solution containing yttrium and strontium tracers i.e. 88Y and 85Sr in conc. HNO3 was loaded onto the column and Sr was not held using conc. HNO3 while Y was held in the column. After complete washing of Sr, nitric acid of pH 3 was used to elute the loaded yttrium from the column. In this experiment, Sr was separated from yttrium completely (Fig. 2.). However this experiment was repeated another two trials under the same experimental conditions to evaluate the consistency of the separation efficiency. Figure 2 also shows mixing of Y in the Sr fractions in the second and third trials using the same column.
Fig. 2

Elution profile with TBP/XAD-4 resin for the separation of Sr and Y(3 consecutive runs using tracers). (Experimental conditions: TBP coated resin 1.6 g; column dimension: 6 mm (ID) × 190 mm (height); traces used 88Y and 85Sr; flow rate of ~1 mL/min; fractions collected 5 mL, assay gamma spectrometry)

The Y activity mixing in the Sr fraction increases sharply from 1st to 2nd run i.e. the 70 mL of the eluted Sr fraction in conc. HNO3 contained 0.56 % of the total Y in the sample in the 1st run which increased to 7.9 % in the second run.

Breakthrough capacity measurement

Breakthrough capacity was estimated under the similar experimental conditions for Y using its definition of the break point in the curve of elution volume of Y stock solution versus C/C0 at which C/C0 became approximately equal to 0.5 i.e. effluent concentration is about 50 % of influent concentration.

The terms C0 and C denote the influent and effluent concentrations of the yttrium stock solution. The 50 % breakthrough capacity of the fresh resin was found to be 6.4 mg of Y/g of TBP coated XAD-4 resin (Fig. 3a) while the same was found to be 1.8 mg of Y/g of TBP coated XAD-4 resin after its use for three separations (Fig. 3b) The degradation in the breakthrough capacity may be due to some leaching of the extractant TBP from the resin. However the application of this technique for the separation depends on the breakthrough capacity calculated up to the fraction where Y has just started appearing rather than 50 % breakthrough capacity because of the fact that the processing of the irradiated yttria needs the collection of pure fraction of Sr. 50 % breakthrough capacity is used here for just comparing the performance of the resin.
Fig. 3

a For fresh resin. b For the resin after three separations. Breakthrough capacity measurement for Y in the TBP coated XAD-4 column (Experimental conditions: TBP coated resin 1.6 g; column dimension: 6 mm (ID) × 190 mm (height); conc. of Y stock solution 0.157 mg/mL in conc. HNO3 with 88Y tracer; flow rate of ~1 mL/min; fractions collected 1 mL (up to initial 10 mL) thereafter 5 mL; assay gamma spectrometry. The activity of a sample on the y axis in the above graphs are plotted on the mid-point of the sample size on the x axis)

Usually the breakthrough capacity observed experimentally is less than the stoichiometric capacity depending on the various parameters such as size and uniformity of the particle, volume capacity, degree of cross linking of the polymer support, temperature, flow rate, concentration of counter ions in the feed, aspect ratio of the column etc. [19].

The breakthrough capacity reported here for the column is much less than the theoretical capacity of 65.2 mg of Y/g of TBP coated XAD-4 resin which was calculated based on the stoichiometric amount of Y corresponding to the amount of TBP coated in the one gram of the resin i.e. only about 10 % of the stoichiometric amount corresponding to the loaded TBP in the resin in which 1 mol of yttrium needs three moles of TBP.
$$ {\text{Y}}^{\text{III}} {\text{ + 3HNO}}_{ 3} {\text{ + 3TBP }} \rightleftharpoons {\text{Y}}^{\text{III}} \left( {{\text{NO}}_{ 3} } \right)_{ 3} \cdot 3 {\text{TBP}} $$

Further optimization of various parameters may be required to maximize the breakthrough capacity for the best utilization of the column.

Elution profile with the dissolver solution

The amount of Y in the dissolver solution used as the feed for the separation was taken such that it is much less than the breakthrough capacity. The dissolver solution contained the impurities of Co, Eu and Zn in addition to the target Y and the product Sr (Table 1).
Table 1

Radioisotopic impurities present in dissolver solution and their origin



Production route




89Y(n,2n) 88Y

Target material




Rare earth impurities from target material




Sintering aid




Activation product of SS

The Sr fraction of the dissolver solution is well separated from target Y even though mixed with the above-mentioned impurities i.e. 65Zn, 58Co and 155Eu (Fig. 4).
Fig. 4

The separation profiles of Sr and Y with actual dissolver solution. (Experimental conditions: TBP coated resin 1.6 g; column dimension: 6 mm (ID) × 190 mm (height); feed: input solution containing 85Sr and 60co tracers; flow rate of ~1 mL/min; fractions collected 5 mL, assay: gamma spectrometry using HPGe detector)

Further purification of Sr from these impurities needs to be carried out using ion exchange chromatography. When pure Y2O3 target pelletized and sintered without using a sintering aid is irradiated within a quartz sample tube enclosed inside a Stainless Steel (SS) irradiation capsule, these impurities may be avoided. In addition, if the breakthrough capacity of the column is enhanced by optimizing the various experimental parameters like column dimensions, flow rate, breakthrough capacity etc., the processing capacity of the column for the irradiated yttria can also be improved.

Further investigations for the improvement of column reproducibility can also facilitate the repeated use of the resin for processing higher amount of irradiated target.

Hence the technique can mature into a feasible method of substitution for the applications for the bulk removal of yttrium in the chemical processing of the irradiated yttria target only after improving the column efficiency and reproducibility.


The extractant TBP coated onto the inert matrix Amberlite XAD-4 was evaluated for its efficiency in separating Sr from bulk Y by extraction chromatographic mode. Repeated separation using the same TBP-coated Amberlite XAD-4 resin caused reduction in the breakthrough capacity of the resin which needs some investigations for the feasibility of any leaching. Further investigation is needed for an efficient adaptation of this technique for the bulk processing of the irradiated target of yttrium.

Copyright information

© Akadémiai Kiadó, Budapest, Hungary 2013