Determination of Lanthanides , Thorium , Uranium and Plutonium in Irradiated ( Th , Pu ) O 2 by Liquid Chromatography Using α-Hydroxyiso Butyric Acid ( α-HIBA )

An HPLC method is presented for the separation and determination of lanthanides (Lns), thorium (Th), uranium (U) and plutonium (Pu) from irradiated (Th, Pu)O2. Individual separation of Lns, Th, U and Pu is a challenging task because of 1) lanthanides having similar physical and chemical properties, 2) presence of complex matrix like irradiated fuel and 3) the co-existence of multiple oxidation states of Pu. Different procedures were developed for separation of individual lanthanides and actinides. The individual lanthanides were separated on a dynamically modified reversed phase (RP) column using n-octane sulfonic acid as an ion interaction reagent and employing dual gradient (pH and concentration) of α-hydroxyisobutyric acid (HIBA). In order to improve the precision on the determination of Lns, terbium (Tb) was used as an internal standard. The method was validated employing simulated high level liquid waste. Concentrations of lanthanides viz. lanthanum (La) and neodymium (Nd) in the dissolver solution were determined based on their peak areas. Th, U and Pu were separated on a RP column using mobile phase containing HIBA and methanol. Since Pu is prone to exist in multiple oxidation states, all the oxidation states were converted into Pu (IV) using H2O2 in 3 M HNO3. Under the optimized conditions, Pu(IV) eluted first followed by Th and U. The concentrations of Th, U and Pu were determined by standard addition method and were found to be 1.10 ± 0.02 mg/g, 5.3 ± 0.3 μg/g and 27 ± 1 μg/g, respectively, in the dissolver solution of irradiated fuel. These values were in good agreement with the concentration of Th determined by biamperometry and those of U and Pu by isotope dilution thermal ionization mass spectrometry.


Introduction
The Indian nuclear program envisages the effective utilization of thorium (Th) as a fertile material for the sustained production of electricity in the country [1].Unlike the natural uranium (U) based fuel which contains 235 U as the fissile material, thoria based fuel initially requires the addition of fissile materials like 233 U, 235 U and 239 Pu from outside.Mixture of (Th, 233 U) O 2 and (Th, 239 Pu) O 2 was proposed as fuel for advanced heavy water reactor (AHWR) to make system self-sustaining in 233 U [2,3].In order to assess the performance of (Th, Pu) O 2 MOX fuel, prior to its introduction in commercial reactors, (Th, 4% Pu)O 2 fuel clusters were irradiated in the Pressurized Water Loop (PWL) of CIRUS reactor, BARC.The implementation of mixed oxide based fuel cycle requires development of methodologies for evaluating the performance of the fuel in the reactor.Burn-up is an important parameter for nuclear fuel development, fuel management and fuel performance analysis [4,5].Burn-up is defined as the atom percent fission of heavy element (mass ≥ 225) during its life time in the reactor.Radiochemical and chemical analysis methods are generally used for burn-up determination of irradiated nuclear fuels [6].The destructive method involved dissolution of fuels followed by individual separation and determination of fission products and heavy elements.Isotope dilution-thermal ionization mass spectrometry (ID-TIMS) is an established method for the determination of burn-up [5,7,8].
Development of new methodologies for the separation and determination of Lns, Th, U and Pu is essential for burn-up determination of thorium based nuclear fuel [9].Different methods based on solvent extraction, ion exchange, precipitation, liquid chromatography, etc. have been reported for the separation of Lns, Th, U and Pu in various matrices [10,11].Among these methods, high performance liquid chromatography (HPLC) is a fast and highly efficient technique and has been applied for the separation of lanthanides and actinides in nuclear fuel samples and geological samples [12][13][14][15][16]. Due to the fast separation and less amount of the sample handling in HPLC, it minimizes the exposure to highly radioactive samples and is, therefore, attractive for the separation of fission products.Cassidy et al. studied the separation of lanthanides fission products using a dynamic ion exchange column from different fuel samples like UO 2 and (Th, U)O 2 fuels [17,18].However, in presence of bulk of actinides, the individual separation of lanthanides was not reported.Lanthanides were separated in a group by conventional ion exchange method and subsequently injected into HPLC after removal of Th and U.This offline separation requires multiple monitoring elements and there are chances of loss of analytes.The reported method gave good reproducibility among different runs for the same aliquot but results were inconsistent for different aliquots of same fuel.Sivaraman et al. carried out extensive studies on the separation of lanthanide fission products for fast breeder test reactor fuels [19][20][21].They determined the stability constant of actinide-HIBA complexes under different chromatographic conditions.However, (Th, Pu)O 2 is a unique fuel tested for AHWR reactor and liquid chromatographic method for separation of Lns from this matrix is not reported in literature.Various RP based methods have been reported for the separation of Th, U and Pu using hydroxyl carboxylic acid like HIBA and mandelic acid [22][23][24].In most of the reported method, the lanthanides, U and Th are present in comparable amounts.Studies were carried out previously in our laboratory using ion interaction reagent (IIR) on the reversed phase (RP) column for the separation of lanthanides from Th and U [25].Under the optimized chromatographic conditions using dual gradient of pH and concentration of mobile phase, lanthanides were separated out from bulk of Th and U from simulated irradiated thoria sample.However, the separation and determination of Lns, Th, U and Pu from irradiated (Th, Pu)O 2 fuel employing HPLC have not been reported so far.
Present paper deals with the individual separation of Lns, Th, U and Pu in an irradiated (Th, Pu)O 2 fuel.Handling of (Th, Pu)O 2 fuel samples is more challenging because of the radiation dose associated with the dissolver solution, presence of large amount of Th which is vulnerable to hydrolysis and multiple oxidation states of Pu under the chromatographic conditions [26].Since the chromatographic behavior of lanthanides is different than tetravalent and hexavalent actinides, two different separation procedures are developed.The challenges of the present studies are: 1) individual separation of lanthanide fission products in presence bulk of Th and interference of trivalent actinides in Lns separation, 2) difficulties in the quantification of Pu because of its multiple oxidation states, and 3) individual separation of Th(IV), Pu(IV) and U(VI).

Instrumentation
The HPLC system consisted of an L-2130 (Elite La-Chrom, Hitachi) low-pressure quaternary gradient pump and an L-2450 (Elite LaChrom) diode array detector.C 18 monolith RP column (100 mm × 4.6 mm, Chromolith, Merck) and C 18 particulate RP column (150 mm × 4.6 mm, 5 µm, Supelcosil) were used as the stationary phases.Solutions were injected into the column using a Rheodyne injector (Model 9725i) with a 100 µL loop.The eluted species were monitored after post-column reaction with a metallochromic reagent, which was added using a reciprocating pump (Eldex Laboratories Inc.) into a low dead volume-mixing tee (Valco).The signal from the detector was processed by EZChrom software package.

Reagents
All solutions were prepared using deionised water from Milli-Q system (Millipore) and were filtered through a 0.45 µm membrane filter (Millipore) prior to using.α-HIBA (Lancaster) was used as an eluent.Sodium n-octane sulphonate monohydrate (Fluka) and tetrabutylammonium iodide (Sigma-Aldrich) were used as the ion interaction reagents (IIR).ICP-standards of the lanthanides, U and Th (Inorganic Venture) were used after appropriate dilutions with Milli-Q water and mobile phase.HNO 3 and NH 4 OH (Merck) were used for adjusting the pH of the mobile phase.H 2 O 2 (Merck) in combination with HNO 3 was used as a redox reagent to bring Pu into the IV oxidation state.For ion exchange separation, Dowex 1 × 8, 200 -400 mesh size (Sigma-Aldrich) and Bio-Rad AG 1 × 2, 200 -400 mesh size (Bio-Rad) were used as stationary phases.Arsenazo (III) (Fluka) was used as the post-column metallochromic reagent (PCR).The arsenazo complexes of Lns, Th, U and Pu were monitored at 653 nm.Irradiated (Th, Pu)O 2 sample received from Post-Irradiated Examination Division (PIED), BARC was used for the method development.NIST-SRM-950a U 3 O 8 , K 4 Pu(SO 4 ) 4 and enriched isotope of 142 Nd were used as spikes for ID-TIMS analysis of U, Pu and Nd, respectively.

Procedure
Appropriate quantities of α-HIBA and sodium n-octane sulfonate were dissolved in water and made to solutions with concentrations 0.5 M and 0.1 M, respectively.α-HIBA was adjusted to the desired pH using high-purity NH 4 OH and HNO 3 .Mobile phase flow-rate of 1.0 mL min −1 was used.The PCR solution [1.5 × 10 −4 M Arsenazo (III) and 0.01 M urea in 0.1 M HNO 3 ] was delivered at a flow rate of 0.3 mLmin −1 .The lanthanides, Th, U and Pu solution of appropriate concentrations were prepared after dilution with mobile phase.Concentration of Pu was determined by biamperometry titration [27].Dissolution of irradiated fuel sample was carried out in a facility housed in a shielded glove box.Aliquots from the irradiated fuel dissolver solution were transferred to a shielded fume-hood for further experiments.
For the determination of Lns, about 0.5 g of dissolved sample solution was taken and 2.5 ppm of Tb was added as an internal standard.This solution was directly injected into HPLC through 100 µL injection port.The mobile phase containing α-HIBA of pH 6.5 was changed from 0.05 M to 0.15 M in 30 min; whereas α-HIBA of pH 3.5 was changed from 0.15 M to 0.3 M in a time interval of 30 to 40 min.A C 18 RP column (250 mm × 4.6 mm) was used for the individual separation of Lns.For validation of method using ID-TIMS, all the aliquots were subjected to necessary chemical treatments to ensure depolymerisation and proper isotopic homogenization.The spiked and unspiked aliquots were used for separation by anion exchange using Dowex 1 × 8, 200 -400 mesh resin in 9 M HCl medium [8].The effluent containing Th and fission product fraction was collected.Pu and U fractions were sequentially eluted from the column using 0.1 M hydroxylamine hydrochloride in 5M HCl and 0.5 M HNO 3, respectively.The fraction containing Th and fission products was subjected to a second stage anion-exchange separation using Dowex 1 × 8200 -400 mesh resin in 7 M HNO 3 medium.The non-retained fission product fraction was collected and subjected to a third stage separation using Bio-Rad AG 1 × 2200 -400 in a mixture containing HNO 3 and MeOH to separate Nd fraction [8].[21].With α-HIBA, Lns and Pu(III) form mainly cationic complexes which are different from those formed by Th, U, Pu(VI) and Pu(IV) [25].Thus Lns-HIBA complexes can be separated on dynamically modified RP column using an IIR which is sorbed unto the column converting hydrophobic surface into the charged surface, for ion exchange separation of lanthanides [28].However, Th, U and Pu(IV) get sorbed onto the RP column by hydrophobic interaction, when α-HIBA is used as a complexing reagent.In presence of large amounts of IIR, the actinides were found to elute in between the lanthanides which makes lanthanides determination difficult.Hence, at a given concentration of IIR, the surface of RP column remains partly ionic and hydrophobic in nature.The separation of lanthanides from Th, U and Pu was performed using dual gradient (pH and Concentration) of α-HIBA.At higher pH, tetravalent and hexavalent actinides are retained better than lanthanides on dynamically modified reversed phase column leading to the elution and separation of lanthanides prior to actinides.Due to the complex nature of irradiated (Th, Pu) O 2 fuel matrix, two different separation procedures were developed as presented in Scheme 1.In the first stage, Lns were separated and determined on dynamically modified column whereas the separation and determination of Th, U and Pu was achieved using RP column.

Determination of Lanthanides in Irradiated
(Th, Pu) O 2 Sample by HPLC A C 18 reversed-phase column, dynamically modified with n-octane sulphonate, was used as the stationary phase for separation of lanthanides from irradiated (Th, Pu)O 2 fuel.It was reported that with increase in the concentration of IIR, the retention of Lns increases whereas the retention of Th and U decreases [28].Th being the matrix element in the present case, the concentration of IIR selected was 5 mM to separate Lns without affecting the Th holding capacity.Effect of pH showed that at lower pH, higher fraction of HIBA remains in the uninonized form, resulting in faster elution of U and Th whereas, at pH ≥ 5.0, Th and U showed stronger retention than lanthanides and eluted after Lns.With the increase in concentration of HIBA at pH 4.0, Lns, Th and U showed decrease in retention.Dual gradient (pH and concentration) of HIBA was used for the separation of Lns, Th, U and Pu.Pu was found to elute as multiple peaks in chromatographic run due to the presence of its For Th determination, a portion dilute to 200 times and separate on RP column as per the conditions given in multiple oxidation states.Hence, dissolver solution was evaporated to near dryness and treated with H 2 O 2 in 3M HNO 3 to convert all the Pu into (IV), and Pu was re-dissolved in 0.1 M HNO 3 to maintain in single oxidation state.Lns were separated by using the gradient condition.Initially, HIBA of pH 6.5 was used for the separation of lanthanides followed by elution of actinides with HIBA of pH 3.0.Under these chromatographic conditions, Pu(IV) elutes along with Th and U and hence does not cause any interference to the lanthanide peaks.Henceforth Th, U and Pu are in the oxidation states, Th(IV), UO 2 2+ and Pu(IV), respectively.The response of the Lns with post-column regent was found to vary with the mobile phase concentration and pH.Therefore, terbium was used as an internal standard.The advantages of internal standard approach are: 1) calibration plot for a wide concentration range is not required.Single injection of the sample is sufficient, and 2) matrix effects do not influence the results since the standard is introduced into the sample.Relative response factors (RRF) for the individual Lns pairs were calculated using the simulated samples.For a pair of Tb and La, the RRF can be calculated as where, C and A represent concentration and absorbance (peak area) of the lanthanides, respectively.The RRFs of the different pair of La/Tb, Pr/Tb, Nd/Tb and Sm/Tb were found to be constant and are presented in Table 1.Validation of lanthanides determination was carried out by employing a simulated high level liquid waste solution.Concentrations of lanthanide fission  products (La, Ce, Pr, Nd and Sm) were determined based on their peak area employing terbium as an internal standard.Figure 1 shows the chromatogram of the separated lanthanides from the irradiated sample.Concentrations of La, Pr, Nd and Smin the dissolver solution of irradiated (Th,Pu)O 2 sample were determined based on the internal standard approach and were found to be 0.6 ± 0.05 μg/g; 0.4 ± 0.01 μg/g; 1.8 ± 0.1 µg /g and 0.5 ± 0.02 μg/g, respectively.

pH of Mobile Phase
Figure 2 shows the effect of pH of the mobile phase on the retention behavior of Th, U and Pu.At pH ≤ 2.5, Pu showed strong retention compared to Th which in turn showed stronger retention than U.This is due to the fact that at lower pH, Pu and Th form predominantly M(IBA) 4 type of species whereas U exists as cationic species.However, at pH ≥ 3.5, retention of U drastically increases compared to marginal increase in the retention of Th and Pu.This is because at higher pH, Th and Pu undergo hydrolysis and must be forming [Th(IBA) 4 (OH) n ] n− and [Pu(IBA) 4 (OH) n ] n− (where n = 1 or 2) species, respectively.Being the anionic species, Th and Pu complexes show relatively poor retention on the RP column.
In the case of U, the dominating species at higher pH are [UO 2 (IBA) 2 ] and [UO 2 (IBA) 3 ] − , which are sufficiently hydrophobic in nature and exhibit strong retention on RP column.With further increase in pH ≥ 5.0, U was not eluted till 60.0 mins whereas the retention time of Th and Pu was 13.3 and 9.9 mins, respectively.Finally, pH 4.3 of the mobile phase was chosen for studying the effect of concentration of HIBA.

Effect of Concentration of HIBA
Retention behavior of Th, U and Pu was studied as a function of concentration of α-HIBA at pH 4.3.As it is seen in Figure 3, the retention of Th, U and Pu were decreasing with the increase in concentration of HIBA.The decrease in the retention times of the actinides with increase in concentration of HIBA is attributed to the competition between the undissociated HIBA molecules and actinide-HIBA complex for the C 18 stationary phase.With the increase in concentration of HIBA used as an eluent, the number of undissociated HIBA molecules in the mobile phase increases and this results in faster dis- placement of actinide-HIBA complexes from the stationary phase.However, retention of U decreased drastically indicating the hydrophobic nature of U-HIBA complex.0.1 M of HIBA was chosen for separation studies.

Mechanism of Retention of Th, U and Pu
To study the mechanism of retention of Th, U and Pu on the RP column, n-octane sulphonic acid (IIR) was introduced in the mobile phase with the varying concentration.

Separation and Determination of Th, U and
The the retention of Th, U and Pu.Chromatographic conditions: same as Figure 4.
w The retention of U decreases drastically compared to that of Th and Pu which indicates relatively more hydrophobic nature of U-HIBA complex.The presence of MeOH also resulted in improving the peak shape and shortening the retention time.Since Th and Pu separation was getting affected at higher percentage of MeOH, it was proposed to use MeOH as a gradient for the separation of Th, U and Pu from the actual dissolver sample.

Pu from Irradiated (Th, Pu)O 2 by RP-HPLC
above optimized chromatographic conditions were Figure 5 shows the retention behavior of Th, U and Pu  Isotope dilution-thermal ionization m (ID-TIMS) methodology was employed for the concentration determination of U, Pu and Nd in the dissolver fuel samples.Chemical separation of U, Pu and Nd fractions in the dissolver solution was carried out to eliminate the potential isobaric interferences during the mass spectrometric analysis as well as to get good ion yield in TIMS.Nd, U and Pu fractions obtained from unspiked and spiked aliquots were used to determine isotopic composition and concentration by ID-TIMS, respectively.Concentrations of U, Pu and Nd in the sample were found to be 5.3 ± 0.3 % g/g, 26.8 ± 0.2% g/g and 1.8 ± 0.2% g/g, respectively.Concentration of Th in the dissolver solution, determined biampero metrically using EDTA as a titrant, was found to be 1.03 ± 0.2% mg/g [30].The concentrations of Nd, Th, U and Pu obtained were in good agreement with the concentrations determined by HPLC.

3. 1 .
Separation and Determination of Lns from Irradiated (Th, Pu)O 2 3.1.1.Chromatographic Behavior of Lanthanides, Th, U and Pu In solution, under the chromatographic conditions, lan-thanides exist in III oxidation state whereas Th and U exist in IV and VI states, respectively.The situation in the cases of Pu is different.In view of the closeness of redox potential values, Pu can exist in different oxidation state viz.III, IV and VI simultaneously.Retention behaviors of Lns and Pu(III) is different from Th(IV), U (VI), Pu(VI) and Pu(IV) when α-HIBA was used as an eluent on RP column

1
Dissolver solution of Irradiated (Th, Pu)O 2 Evaporate to dryness Redox treatment with 3M HNO 3 & 30% H 2 O 2 Treatment with concentrated HNO 3 Evaporate to near dryness Evaporate to near dryness Re-dissolve in 1 M HNO 3 and take aliquots Aliquot for Lns separation LC separation on a RP column as per the conditions given in Figure Aliquot for Th, U and Pu separation Dissolve in 0.5 M HIBA of pH 6.0 Separate U and Pu on RP column as per the conditions given in Table 2
Conditions for the Separation of Th, U and Pu U forms [UO 2 (IBA)] + , [UO 2 (IBA) 2 ] and [UO 2 (IBA) 3 ] − whereas Th and Pu form [M(IBA)] 3+ , [M(IBA) 2 ] 2+ , [M(IBA) 3 ] + and [M(IBA) 4 ] (M = Th or Pu) types of complexes with HIBA [21,29].Dominance of one species over the other depends upon the pH and concentration of HIBA which is responsible for the relative difference in the retention times of Th, U and Pu.Hence the effects of chromatographic conditions such as pH of the mobile phase, concentration of HIBA on the retention of Th, U and Pu was studied.

Figure 2 .
Figure 2. Effect of pH of the mobile phase on the retention of Th, U and Pu.Chromatographic conditions: 0.1 M of HIBA and Column: C 18 RP (100 mm × 4.6 mm, Chromolith).

Figure 3 .
Figure 3.Effect of concentration of α-HIBA in the mobile phase on the retention of Th, U and Pu.Chromatographic conditions: pH of mobile phase, 4.3 and other conditions are same as in Figure 2.

3 . 6 .
sequential elution of Pu, Th and U.In order Validation of HPLC Method ass spectrometric u to improve the resolution between Th and Pu, gradient condition, as given in Table2was employed.To improve the resolution as well as Th holding capacity, a larger stationary phase of dimension 250 mm × 4.6 mm was used.Aliquots of irradiated sample were evaporated to near dryness and were taken in the mobile phases containing pH 6.5 of 0.13 Mα-HIBA, and were injected through 100 µL injection port.As shown in Figure6, lanthanides and other trivalent species are eluted at the solvent front.Th, U and Pu are retained on the column because of the formation of hydrophobic species using α-HIBA as an eluent.Under the optimized conditions, the elution pattern was Pu (IV) followed by Th and U.This is due to the fact that at higher pH for Th and Pu(IV), M(HIBA) 4 type of species are most dominating whereas for U, [UO 2 (IBA) 3 ] − is the major species.However, the lower retention time of Pu and Th compared to U can be explained on the basis of strong hydrolysis nature of Th and Pu (IV) in the chromatographic conditions resulting in the formation of [M(IBA) 4 (OH) 2 ] 2− .The concentrations of Pu, Th and U were determined by standard addition method and were found to be 27 ± 1 µg/g, 1.10 ± 0.02 mg/g, and 5.3 ± 0.3 µg/g, respectively, in the dissolver fuel solution.

Figure 6 .
Figure 6.Direct injection of dissolver solution of irradiated