Introduction

The need for efficient and selective ion exchangers exists in the treatment of radioactive waste effluents. The inorganic ion-exchangers serve as useful materials compared to organic resins because of their high radiation, chemical and thermal stability as well as compatibility with immobilization matrices. Also the more uniform ion exchange sites achievable in crystalline inorganic materials can lead to remarkable selectivity [1]. The benefits from using highly selective inorganic exchangers are not only the significant reductions in the liquid waste volumes and radioactive discharges into the environment, but also considerable reductions in the costs of waste management and processes for disposal. The presence of 137Cs (t 1/2 = 30.1 years) and 90Sr (t 1/2 = 28.5 years) in the waste solution poses a lot of problem due to long half lives and high fission yields. Extensive literature is available on the treatment of radioactive wastes using inorganic ion-exchangers for the removal of the above radionuclides and a comparison of the inorganic ion exchangers for the removal of 137Cs and 90Sr has been reported [2–6]. However the pH of the waste solution and the presence of high dissolved solids often is a limiting factor in the use of ion exchangers available today.

The titanosilicates are a new class of inorganic ion exchangers which have shown promise in the fine tuning of ion exchange properties. A large number of titanium silicates are known to exist among the minerals in nature. The pure crystalline silicotitanates (CST) and niobium substituted crystalline silicotitanates (Nb-CST) with a sitinakite topology have frame work structures that enclose tunnels or cavities which are highly selective to 90Sr and 137Cs respectively over a broad pH range and in the presence of Na+ [7]. The frame work of CST is made up of (TiO4) cubane like clusters that are linked by SiO4 units which results in a 3D structure with tunnels along c-axis which are of an ideal size for selective exchange of Cs+ and Sr2+ thus making these exchangers one of the most promising materials for their removal. The origin of selectivity appears to arise from the higher coordination number of cesium and strontium due to various hydration sites in the tunnel. In addition to the above characteristics the radioactive ions trapped inorganic material can be directly immobilized to a ceramic waste form for final disposal [8–10].

We report here the synthesis, characterization and comparative ion-exchange behaviour of CST and Nb-CST for uptake of 137Cs and 90Sr from an actual radioactive waste solution from pretreatment tank (PTT) at Centralised Waste Management Facility (CWMF), Kalpakkam. The distribution coefficients (K d) which indicate the selectivity towards the uptake of Cs and Sr and the decontamination factors (DF) after 24 h of equilibration for the above-said ion exchangers were determined. Kinetics of the uptake of the above-said radionuclides from PTT sample was also investigated. The ion exchange capacity of both CST and Nb-CST towards the uptake 137Cs and 90Sr was also estimated.

Experimental

Synthesis of CST and Nb-CST

The hydrothermal synthesis of CST with composition (58.5 mmol Ti, 6 mmol Si, and 620 mmol Na) was carried out by addition of chemicals in the order as mentioned below [11]. A total of 6.66 mL of TiCl4 was mixed with 25 mL of ultra pure water and 40 mL of 30% hydrogen peroxide and stirred vigorously. To the stirring mixture 40 mL of 10 M NaOH was added followed by 150 mL of ultra pure water and a solution of 6 mmol of SiO2 in 1 M NaOH. Finally to the clear solution 0.12 mol of NaOH was added. The mixture was sealed in an autoclave and kept at 200 °C for 10 days. After cooling to room temperature the solid product was separated by filtration, washed with 1 M NaOH and ethanol and dried for 24 h at 60 °C at air.

In the case of hydrothermal synthesis of Nb-CST with composition (12 mmol Ti, 16 mmol Si, 4 mmol Nb and 165 mmol Na) 3.43 g of titanium isopropoxide, 3.33 g of tetraethylorthosilicate and 0.54 g Nb2O5 were added to 50 mL aqueous NaOH (6.6 g) solution in an autoclave, stirred for half an hour and then placed in an oven for 200 °C for 3 days. The resulting product was collected by filtration. Since there is possibility of formation of a crystalline by-product with the major Nb-CST product formed the former was removed by a two-step process: first the filtered product was exposed to 1 M aqueous HCl wash for 3 h at room temperature to amorphize the by product and secondly the resulting solid was then exposed to 1 M NaOH wash for 3 h at 40 °C to dissolve the resulting amorphous material. The chemicals and the reagents used were of analytical grade.

Characterisation of CST and Nb-CST

Powder X-ray diffraction (XRD) pattern was collected on a STOE diffractometer using Cu Kα radiation between 2θ = 5 and 90º at a step size of 0.1º and a step time of 5 s. Scanning electron microscope (SEM)/energy dispersive spectroscopy (EDS) data was collected on a Philips XL-30 system. Surface area of the ion exchangers were determined using Smartsorb 90/91 using single point measurement technique. The DTA/TGA experiments were performed on Rheometric STA 1500 DTA/TGA instrument. Samples 10–20 mg were heated in air.

Ion exchange properties

All ion exchange experiments were performed using the actual radioactive waste solution from pretreatment tank at CWMF by the batch method. The characteristics of the waste are given in Table 1. The major cations in the waste solution were calcium, sodium and the anions were chloride and nitrate contributing to the total dissolved solids.

Table 1 Characteristics of pretreatment tank (PTT) waste
Full size table

Kinetic studies and determination of decontamination factor (DF)/distribution coefficients (K d)

About 1 g of the as-synthesised sample of CST/Nb-CST was equilibrated with 100 mL of the actual radioactive waste solution (initial activity of 137Cs −204 Bq/mL and 90Sr −37 Bq/mL) from pretreatment tank at CWMF at ambient temperature. Aliquots of the solution were taken at regular time intervals till 24 h viz., instantaneous, 0.5, 1, 1.5, 2, 2.5 and 24 h for estimation of radioactivity. The samples were centrifuged at 6,000 rpm and filtered. The radioactive cesium and strontium in the supernatant were estimated using gamma counter and Cerenkov radiation technique respectively.

The distribution coefficient (K d) was calculated using Eq. 1

$$ K_{\text{d}} \, ({\text{mL/g) }} = {\frac{{{\text{Initial\, activity}} - {\text{Final\,activity}}}}{\text{Final activity}}} \times \frac{V}{m} $$
(1)

Where V (in mL) is the volume of the solution and m (in g) is the mass of the ion exchanger respectively.

The percentage removal (%) was calculated using Eq. 2

$$ \% {\text{ Removal }} = {\frac{{{\text{Initial\,activity}} - {\text{Final\,activity}}}}{\text{Initial activity}}} \times 100 $$
(2)
$$ {\text{Decontamination factor }}\left( {\text{DF}} \right) \, = {\frac{\text{Initial\,activity}}{\text{Final activity}}} $$
(3)

Estimation of ion-exchange capacity

The ion exchange capacity of CST and Nb-CST towards the uptake of both cesium and strontium by each of them was determined by the method described below. In this experiment 1 g of the exchanger was equilibrated at ambient temperature with 100 mL of solution containing 10−1 M of Cs and Sr ion spiked by 137Cs and 90Sr tracer. After 24 h of equilibration the radioactivity of the supernatant was measured as described above. From the initial and final activity the amount of cesium and strontium exchanged was calculated.

Results and discussion

Synthesis and characterisation

The CST synthesized was light brown in colour and the yield was around 6–8 g in each run. The Nb-CST was a white microcrystalline powder of yield around 2.0–2.3 g. The powder patterns in Figs. 1 and 2 depict primary X-ray diffraction line of CST and Nb-CST at 2θ = 9º and at 11.2º respectively [12]. The typical surface area for the samples was in the range of 65–70 m2/g which is in conformity with the literature [9].

Fig. 1
figure 1

X-ray powder pattern of CST

Full size image
Fig. 2
figure 2

X-ray powder pattern of Nb-CST

Full size image

The SEM image of CST depicted in Fig. 3 exhibit large quantities of crystallites with cubic morphology of size 2.4–42 μm2 comprised primarily of Ti, Na, O and with small amount of Si as revealed by an EDS analysis and rod shaped morphology of size 0.3–0.9 μm2 comprised of primarily Ti, small amounts of Na, O and traces of Si as revealed by EDS analysis. The SEM image of Nb-CST depicted in Fig. 4 exhibit large quantities of crystallites with cubic morphology of size 2.9–18.1 μm2 comprising of primarily Na and O and equal amounts of Ti and Nb and small amount of Si as revealed by an EDS analysis and small quantities of rod shaped crystallites of size 2.9–17 μm2 comprising of primarily Na and O with small amounts of Ti and Nb with traces of Si as revealed by EDS analysis [9, 12].

Fig. 3
figure 3

SEM image of CST

Full size image
Fig. 4
figure 4

SEM image of Nb-CST

Full size image

From Fig. 5 which depicts the DTA/TGA curve for CST it can be observed that there is majority weight loss (~18wt%) by 150 °C with ~10% up to 1,000 °C. The endotherm centered 100 °C is accompanied by a weight loss in TGA and can be attributed to loss of water. The endotherm centered at 950 °C indicates a phase change as there is no weight loss in TGA. From Fig. 6 which depicts the DTA/TGA curve for Nb-CST there is majority weight loss (~17%) by 150 °C and ~11% up to 1000 °C. The endotherm centered around 100 °C is accompanied by a weight loss in TGA and can be attributed to loss of water.

Fig. 5
figure 5

DTA/TGA analysis of CST

Full size image
Fig. 6
figure 6

DTA/TGA analysis of Nb-CST

Full size image

Ion exchange properties

Kinetic studies and determination of decontamination factor (DF)/distribution coefficients (K d)

The kinetics of uptake of cesium and strontium by CST and Nb-CST is given in Fig. 7. It is evident from the figure that the kinetics of removal of 90Sr from actual waste solution is the same for both CST and Nb-CST. There was 94% uptake of 90Sr by both CST and Nb-CST instantaneously. The ion exchange properties viz., DF and K d values of both CST and Nb-CST after 24 h of equilibration are depicted in Table 2. A DF value of 176 and 136 was obtained after 24 h of equilibration for uptake of 90Sr by CST and Nb-CST respectively. The K d value for 90Sr after 24 h of equilibration was 17,525 mL/g for CST whereas it was found to be 13,500 mL/g for Nb-CST. The similarities in the shape of the curves suggest that although Nb5+↔Ti4+ substitutions increased the tunnel capacity it did not affect the Sr diffusion rates [11]. Similar results were obtained by Oji et al., which depicted significant affinity and uptake (greater than 96%) for both cesium and strontium from high caustic nuclear waste simulants by CST loaded membranes [13].

Fig. 7
figure 7

Uptake of 137Cs and 90Sr by CST and Nb-CST

Full size image
Table 2 Ion exchange properties of CST and Nb-CST after 24 h of equilibration
Full size table

The uptake of 137Cs by Nb-CST is low till 2 h of equilibration unlike 90Sr but after 24 h of equilibration there was an increase in the K d from 9,982 to 35,490 mL/g which resulted in 99% uptake implying kinetics of 137Cs uptake by Nb-CST is slower than that of 90Sr. This is due to the larger size of Cs+ than Sr2+ its diffusion rate is expected to be lower than that of Sr2+ [11]. A DF of 40 and 355 was obtained for uptake of 137Cs by CST and Nb-CST after 24 h of equilibration. The uptake of 137Cs by pure CST is less than that of Nb-CST. From the above studies it can be concluded CST is selective to uptake of Sr and Nb-CST is selective to uptake of 137Cs. The selectivity of the above mentioned silicotitanate exchange materials may be due to chemistry and geometry of the tunnels and pores [14]. The selectivity is a result of physico-chemical and stereochemical factors such as the hydrated radii, hydration energy, valence of cations and space requirement in titanosilicate [15].

The cesium K d values varied from 15,000 to 1,24,000 mL/g and Sr K d values varied from 610 to 33,000 mL/g during the performance evaluation of 17 inorganic ion exchange materials (pharmacosiderite, sodium zirconium silicate, different forms of crystalline silicotitanate materials developed by Sandia National Laboratory, zeolite exchangers etc.) for removal of Cs and Sr from actual N-reactor storage basin water [16]. The K d values for the uptake of Cs and Sr by the crystalline silicotitanates synthesized at CWMF are in conformity with the values obtained in the literature.

Estimation of ion-exchange capacity

The ion exchange capacity of Nb-CST towards 90Sr and 137Cs was estimated to be 11.8 and 3.2 meq/g respectively whereas the ion exchange capacity of CST towards 90Sr and 137Cs was estimated to be 14.6 and 4.4 meq/g respectively.

The ion exchange capacity towards cesium for some of the commercial inorganic ion exchangers viz., zeolite (modernite), silicotitanate (CST UOP), hexacyanoferrate (Cs Treat) and laboratory synthesized sodium form of CST are ~2, ~2, 0.36, and 1.06 meq/g respectively whereas for Sr it is found to be in the range of 2–5 meq/g [15–20]. The ion exchange capacities for the CST and Nb-CST synthesized at CWMF have higher exchange capacities towards Cs and Sr thus proving to be potential ion exchangers for their application in column studies.

Based on the above results further studies are underway for the development of microspheres encapsulating both the ion exchangers, their characterization and also to explore their feasibility in continuous column operations for treatment of radioactive wastes.

Conclusion

A complete study on the synthesis, characterization and uptake of Cs and Sr from actual radioactive waste solution at CWMF by CST and Nb-CST has been carried out. The CST and Nb-CST ion exchangers are highly crystalline with bright cubic and rod like morphology with varying amounts of Ti, Si, Na, Nb and O. The kinetics of removal of 90Sr from actual waste solution is same for both CST and Nb-CST whereas the kinetics of 137Cs uptake by Nb-CST is slower than that of 90Sr which is due to the larger size of Cs+ than Sr2+ and hence its diffusion rate is expected to be lower than that of Sr2+. The ion exchange capacity of both the ion exchangers towards the uptake of 137Cs and 90Sr also enhances the evidence towards their selectivity. Hence CST and Nb-CST are promising ion exchangers for the selective removal of Sr and Cs respectively from radioactive waste solutions.