Method for separation of 90Y from 90Sr

Abstract

Inorganic ion exchange materials for the separation of 90Y from 90Sr include chabazite, clinoptilolite, potassium pharmacosiderite, sodium titanosilicate and sodium nonatitanate. These materials are suitable for making a 90Y generator that contains 90Sr immobilized on an ion exchange column of the materials. The materials have a very high selectivity for 90Sr, a very low selectivity for 90Y, good radiation and thermal stability, low toxicity, fast reaction kinetics, and can be readily and reproducibly synthesized. A method is thus provided for eluting 90Y from the ion exchange material with an eluant solution. The eluant solution is preferably aqueous, preferably has a pH greater than about 5, and preferably includes a chelating agent. Preferred chelating agents include gluconic acid, oxalic acid, iminodiacetic acid, nitrilotriacetic acid, citric acid, and combinations thereof.

Description

BACKGROUND OF THE INVENTION

[0002] 1. Field of the Invention

[0003] This invention relates to methods, apparatus and compositions for separating yttrium-90 from strontium-90.

[0004] 2. Background of the Related Art

[0005] The use of radioactive isotopes as diagnostic, imaging and therapeutic agents is a relatively new area of medicine that has flourished in the last fifty years. A number of radioisotopes, primarily beta emitting radionuclides, are finding use in the in vitro treatment of cancers to destroy or sterilize cancer cells. The treatment is administered in a series of cycles to avoid radiotoxicity to other areas of the body, particularly the kidneys and bone marrow. The isotopes of interest are commonly attached to monoclonal antibodies or polypeptides specific for the cancer cells to be treated, thus delivering a dose of radiation directly to a tumor. This technique is termed radioimmunotherapy (RIT) and is increasingly being used to complement existing surgical techniques and chemotherapy.

[0006] In order to fuel the current research in the use of radionuclides to treat cancers, it is essential that new isotope production methods be developed to increase the availability and decrease the cost of radioisotopes. For medicinal applications, the radioisotope supplied needs to be radiochemically pure to prevent the accidental introduction of unwanted additional radionuclides into a patient, and, preferably, be carrier free. A fundamental aspect of increasing the availability of radioisotopes to medical personnel is the development of new, inexpensive, radiolytically stable materials to allow the necessary separations to be achieved.

[0007] 90 Y is a high-energy beta emitter that is finding use in the treatment of certain forms of cancer. 90Y decays by pure beta emission, with a half-life (T½) of 64 hours, to stable 90Zr. The energetic beta particles (2.3 MeV) can penetrate an average of 0.5 cm in human tissue, with a maximum penetration of up to 1 cm. Consequently, they are useful in the treatment of cancerous tumors like those found in Hodgkin's disease, where tumors are typically between 1 and 5 cm in diameter. The 90Y can be successfully attached to an antibody or peptide fragment, which will then transport the 90Y to the targeted tumor.

[0008] In order to use 90Y in the treatment of cancers, it is necessary to obtain a very pure source of the isotope that is free from the parent 90Sr. This is essential because 90Sr has a 28 year half-life and is likely to accumulate in the bone if inadvertently introduced into the body. The maximum tolerable amount of 90Sr fixed in the bone is only 2 μCi and consequently great care needs to be performed to achieve the necessary Sr/Y separation to ensure minimal introduction of 90Sr into the body during the 90Y radiotherapy.

[0009] 90 Y is the daughter product of 90Sr, an abundant fission product of 235U, found in nuclear wastes resulting from the reprocessing of spent commercial nuclear fuel and in the separation of 239Pu for weapons manufacture. 90Sr has a half-life of approximately 28 years. The radioactive decay scheme is outlined in Equation 1 below.

90 Sr(β−)→90Y(β−)→90Zr  (1)

[0010] In order to obtain a supply of 90Y, it is first necessary to separate 90Sr from other isotopes in the nuclear waste. This can readily be achieved using selective precipitation, ion exchange or solvent extraction techniques to produce a crude 90Sr ‘cow’ for use as a source of 90Y. 90Y can also be produced by the neutron irradiation of 89Y oxide, Y2O3, for a period ranging from several days to a week, but this is expensive and the 90Y product contains large amounts of inactive 89Y making it unsuitable for medicinal applications.

[0011] There are a number of methods described in the literature for the separation of the 90Y daughter from the parent 90Sr, including solvent extraction, ion exchange, precipitation and chromatographic procedures. Of these methods, ion exchange techniques have probably received the most attention. However, all of the current methods suffer from drawbacks. For instance, in some separation procedures, the 90Sr is held onto an organic cation exchange resin and the 90Y is eluted using an aqueous complexant solution, such as EDTA, oxalate, lactate, citrate etc. Consequently, the purified 90Y is generated as a complex that is not suitable for the direct labeling of antibodies and requires further processing. Organic ion exchange resins are also prone to radiation damage resulting in a decrease in capacity and the potential release of toxic organic molecules into the 90Y stream as the resin decomposes. Consequently, there is a need for new material and methods to produce pure 90Y.

[0012] The method disclosed by Bray and Webster in U.S. Pat. No. 5,512,256 uses a solvent extraction process to separate 90Y from 90Sr. A 0.3M solution of di(2-ethylhexyl)phosphoric acid (HDEHP) in n-dodecane is used to extract 90Y from a solution of 90Sr/90Y in 0.3M nitric acid. The HDEHP selectively extracts the 90Y into the organic phase and residual 90Sr can be removed by further washing the organic fraction with fresh 0.3M nitric acid. Although this method is very effective at separating 90Y from 90Sr, multiple steps are required and the recovery of both the 90Sr cow and 90Y fractions requires multiple washing and stripping phases. This produces waste organic and aqueous streams that need to be treated and disposed of safely. There will also be some radiolysis of both the organic complexant and the solvents that will limit their useful life and also may cause the release of unwanted organic species into solution. This is the primary method utilized to produce 90Y in the USA today.

[0013] In U.S. Pat. No. 5,368,736, Horwitz and Dietz use a multiple step chromatographic process to separate 90Sr from 90Y. The 90Sr stock solution in 3M nitric acid is passed through three strontium selective chromatographic ion exchange columns in series so that the solution exiting the third column contains essentially only 90Y, the 90Sr being retained on the columns. This raw 90Y solution is the passed through a rare earth selective column that selectively extracts the 90Y. The purified 90Y can then be eluted off the column. However, the chromatographic columns contain organic resins that are susceptible to radiation damage and may leach undesirable radiolysis fragments into the purified 90Y stream. Radiation damage is kept to a minimum by loading and then eluting the radioactivity from the columns, but this method also requires the use of a dedicated hot cell facility, necessitating shipment of the purified 90Y to the end user.

[0014] Huntley's U.S. Pat. No. 5,494,647 discloses an ion exchange process for separating 90Y from 90Sr using CHELEX-100® (Bio-Rad Laboratories, Richmond, Calif.), a chelating ion exchange resin. CHELEX-100® is an organic ion exchange resin that consists of iminodiacetic acid groups mounted on a polystyrene/divinyl benzene substrate. The method is designed for use with environmental samples only containing trace amounts of 90Sr, and it is disclosed that the method does not work effectively at high strontium concentrations. The organic resin would also be susceptible to radiation damage and it is doubtful that the method would be able to produce the level of 90Y purity required for medicinal applications.

[0015] Therefore, there is a need for improved methods, apparatus, and compositions for separating yttrium-90 from strontium-90. It would be desirable if the compositions were highly radiation resistant, thermally stable, chemically stable, and non-toxic. It would be even more desirable if the compositions and methods provided very high affinities for strontium-90 and very low affinities for yttrium-90.

SUMMARY OF THE INVENTION

[0016] The present invention provides a process for separating strontium-90, comprising the adsorption of strontium-90 onto an inorganic ion exchange material from a solution containing a source of strontium-90. The solution is preferably neutral or near neutral. The process may entail selecting the inorganic ion exchange material from chabazite, clinoptilolite, pharmacosiderite, titanosilicate, nonatitanate, and combinations thereof.

[0017] In one embodiment, the inorganic ion exchange material is sodium nonatitanate prepared by reacting titanium isopropoxide and aqueous sodium hydroxide at a temperature between 100° C. and 250° C. for a period between 12 hours and 2 weeks. Optionally, the inorganic ion exchange material is sodium titanosilicate prepared by hydrothermally heating a titanium silicate gel in NaOH. Said titanium silicate gel may be hydrothermally heated in 6M NaOH at 170° C. for 2 days.

[0018] In another embodiment, the inorganic ion exchange material is a titanosilicate having the general formula:

M3H(AO)4(BO4)3.xH2O

[0019] where: M is a cation selected from H, K, Na, Rb, Cs and mixtures thereof;

[0020] A is selected from Ti and Ge; and

[0021] B is selected from Si and Ge; and

[0022] x is a value between 4 and 6.

[0023] A further embodiment of the invention provides a yttrium-90 generator prepared according to the aforementioned process. This generator may comprise inorganic ion exchange material selected from clinoptilolite, chabazite, pharmacosiderite, titanosilicate, sodium nonatitanate, or other inorganic compounds with a high affinity for strontium, and combinations thereof. The inorganic ion exchange material may be formed into pellets having a diameter between 0.2 and 0.5 mm. Optionally, the pellets comprise polyacrylonitrile, or another polymer, as a binder. Alternatively, the pellets may comprise amorphous titanium dioxide, clay, amorphous silica, amorphous zirconia, or another inorganic oxide as the binder.

[0024] An additional embodiment provides a process for separating yttrium-90 from strontium-90, comprising preparing a solution of strontium-90 then adsorbing strontium-90 from the solution onto an inorganic ion exchange material, and eluting yttrium-90 from the inorganic ion exchange material with an aqueous solution. The process may further comprise the step of allowing yttrium-90 to grow into the inorganic ion exchange material. These steps may be repeated. Optionally, the inorganic ion exchange material is selected from clinoptilolite, chabazite, pharmacosiderite, titanosilicate, nonatitanate, or other inorganic compounds with a high affinity for strontium, and combinations thereof. Preferably, the yttrium-90 is eluted with a solution including a chelating agent.

BRIEF DESCRIPTION OF THE DRAWINGS

[0025] So that the above recited features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments thereof that are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered limiting of its scope, for the invention may admit to other equally effective embodiments.

[0026] FIG. 1 is a diagram illustrating the structure of clinoptilolite illustrating the regular channels within the structure that give rise to ion sieving properties.

[0027] FIG. 2 is a diagram illustrating the layered structure of sodium nonatitanate, Na4Ti9O20.xH2O. Sodium ions (solid circles) and water molecules are located between layers of TiO6 octahedra.

[0028] FIG. 3 is a diagram illustrating the structure of the Cs-exchanged form of the titanosilicate, NaTS.

[0029] FIG. 4 is a diagram illustrating the structure of the potassium form of the pharmacosiderite titanosilicate, KTS-Ph.

[0030] FIG. 5 is a schematic representation of the structure of synthetic chabazite, showing the exchangeable cation sites located on the insides of the large tubes

[0031] FIG. 6 is a graph of x-ray diffraction patterns showing the effect of hydrothermal treatment on the crystallinity of sodium nonatitanate, including: (A) TA-A-18, no treatment, (B) TA-A-19, 21 hr. at 170° C., and (C) TA-A-17, 7 days at 170° C.

[0032] FIG. 7 shows the formula and structures of five suitable complexants.

[0033] FIG. 8 is a three-dimensional bar chart of the ratio of distribution coefficients for strontium and yttrium for nine ion exchange materials and five complexants tested.

[0034] FIG. 9 is a schematic diagram of a column filled with an ion exchange material, such as sodium titanosilicate.

[0035] FIG. 10 is a three dimensional bar chart of the yttrium yield from chabazite (AW-500) with a series of eluants. The two rows of columns represent the results for the two columns being tested.

[0036] FIG. 11 is a three dimensional bar chart of the yttrium yield from sodium titanosilicate (TA-A-13) with a series of eluants.

[0037] FIG. 12 is a three dimensional bar chart of the yttrium yield from clinoptilolite with a series of eluants.

[0038] FIG. 13 is a three dimensional bar chart of the strontium lost from clinoptilolite with a series of eluants.

[0039] FIG. 14 is a three dimensional bar chart of the yttrium yield from potassium pharmacosiderite (TA-A-2) with a series of eluants.

[0040] FIG. 15 is a three dimensional bar chart of the strontium lost from potassium pharmacosiderite (TA-A-2) with a series of eluants.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0041] This present invention provides inorganic ion exchange materials to separate 90Y from 90Sr in neutral to alkaline media. These inorganic ion exchange materials include chabazite (such as sodium chabazite), clinoptilolite (such as sodium clinoptilolite), pharmacosiderite (such as potassium pharmacosiderite (KTS-Ph)), titanosilicate (such as sodium titanosilicate (NaTS)), and nonatitanate (such as sodium nonatitanate (NaTi)). All of the ion exchangers are purely inorganic, highly radiation resistant, thermally stable, chemically stable, and non-toxic. Because of this stability, no release of toxic organic fragments is experienced, no reduction in ion exchange capacity occurs, and large levels of activity may be loaded onto a generator. Stability of the generator is essential because it has been estimated that the radiation dose in a 90Y generator containing 30 Ci of 90Sr/90Y is as high as 109 rad/day. A comparison of the characteristics of organic ion exchange resins and inorganic ion exchangers is given in Table 1.

TABLE 1
Comparison of organic ion exchange resins
and inorganic ion exchange materials
Property	Organic Resins	Inorganic Ion Exchangers
Thermal Stability	Low	High
Ion Selectivity	Low to Moderate	Moderate to High
Radiation Stability	Low	High
Physical Form	Beads, Granules, etc.	Usually powders
Cost	Moderate	Variable

[0042] Clinoptilolite is a naturally occurring zeolite and has the ideal formula Na6Al6Si30O7224 H2O. It is a member of the Heulandite group of zeolites and has a layered porous structure, which is depicted in FIG. 1. The ion selectivity arises due to the regular pores and channels within the zeolite structure which give rise to an ion sieving effect. The sodium ions are readily exchangeable for other cations and, in nature, K+, Mg2+ and Ca2+ are generally also found on the exchange sites along with minor quantities of a range of other cations. Clinoptilolite has a particularly high selectivity towards Cs+ and Sr2+ ions. The cation exchange capacity of clinoptilolite is, however, fairly low and is only approximately 2.2 meq/g for the fully sodium-exchanged form.

[0043] Sodium nonatitanate, Na4Ti9O20xH2O, is synthesized by hydrothermally treating a titanium salt, such as titanium isopropoxide, in strong base at a temperature of between 150° C. and 250° C. The reaction is outlined in Equation 2.

9 Ti(OC3H7)4+4 NaOH(aq)→Na4Ti9O20.xH2O+9 C3H7OH  (2)

[0044] The resulting sodium nonatitanate material is only poorly crystalline and its precise structure has not been determined. However, experiments have suggested that it consists of layers of TiO6 octahedra separated by water molecules and sodium cations, as shown in FIG. 2. The sodium cations are weakly held and readily exchanged for strontium and other cations. The layers of TiO6 octahedra are generally separated by a space of 10 Å, though this distance can vary according to the amount of water intercalated between the layers. The nonatitanate has a high selectivity for strontium at pH greater than 7, but a negligible selectivity in more acidic conditions. Thus, strontium absorbed onto the material can be readily stripped with dilute mineral acid allowing the ion exchanger to be reused. The titanate also has good thermal, radiolytic and chemical stability and is expected to have a low toxicity. The theoretical cation exchange capacity (CEC) is 4.74 meq/g, which compares favorably with organic ion exchange resins.

[0045] Sodium titanosilicate (NaTS) has the ideal formula Na2Ti2O3SiO4.2H2O. This material can be synthesized in a crystalline form that has allowed its structure to be determined using X-Ray powder methods. The titanosilicate was found to have a tetragonal unit cell with a=b=7.8082(2) Å and c=11.9735(4) Å. Edge-sharing TiO6 clusters reside in all eight corners of the unit cell and silicate tetrahedra are located midway between the clusters and link them together. This arrangement produces tunnels parallel to the c axis where the exchangeable sodium ions and the water molecules reside. The remaining sodium ions are located in the framework, bonded by silicate oxygens and are thus not exchangeable. The structure of this ion exchange material is illustrated in FIG. 3.

[0046] Due to steric repulsions and space limitations, some of the sodium ions in the tunnels of the sodium titanosilicate are replaced by protons leading to an actual formula of Na1.64H0.36Ti2O3SiO4.1.84H2O. This exchanger was synthesized by hydrothermally heating a titanium silicate gel of appropriate stoichiometry of four moles of titanium for each mole of silicon in 6M NaOH at 170° C. for 2 days. This material has been shown to have a high selectivity for Cs+ ions in both acid and alkaline pH and a high selectivity for strontium in alkaline media. Strontium is readily removed by washing with dilute acid.

[0047] The second class of titanosilicate materials has the crystal structure of the natural mineral pharmacosiderite. Pharmacosiderite has the ideal formula KFe4(AsO4)3(OH)4 and crystallizes in the cubic system. Titanosilicates with the general formula M3H(AO)4(BO4)3.4-6H2O (M=H, K, Na, etc.; A=Ti, Ge; B=Si, Ge) were prepared using hydrothermal techniques. A homogenous gel of appropriate stoichiometry of four moles of titanium for each three moles of silicon was hydrothermally treated in an excess of either KOH or CsOH at 200° C. for 1 to 3 days. Sodium and proton forms were then prepared by exhaustively ion exchanging the material with either NaCl or HCl. The most studied material of these is the potassium pharmacosiderite, K3H(TiO)4(SiO4)3.4H2O (KTS) in which a=b=c=7.7644(3)Å. Each unit cell consists of clusters of four titania octahedra linked to each other by silicate groups as shown in FIG. 4. This produces a series of intersecting tunnels parallel to the a, b and c axes with the exchangeable ions residing close to the face-centers of the unit cell. Pharmacosiderites have shown very high affinities towards strontium ions in alkaline solutions.

[0048] Chabazite is a well-characterized synthetic zeolite having exchangeable cation sites located on the insides of the large tubes. FIG. 5 is a schematic representation of the structure of synthetic chabazite.

90 Y/90Sr Separation Process

[0049] One embodiment of the present invention provides a separation method that comprises the following steps:

[0050] (1) A source of 90Sr is loaded onto the inorganic ion exchange material from a solution of dilute sodium salt, the loaded exchanger is slurried into a column, and the column is washed with fresh sodium nitrate solution to remove any residual radioactivity not bound to the exchanger.

[0051] (2) 90Y is allowed to grow into the column.

[0052] (3) 90Y is eluted using a dilute solution of a sodium salt or similar eluant at a pH of 7 or greater. 90Sr is strongly held by the ion exchanger and remains on the column.

[0053] (4) Optionally, the eluted 90Y is then passed through a small, secondary microcolumn of the ion exchange material to remove any residual traces of 90Sr that may have been eluted from the primary ion exchange column. This small column, anticipated to be similar in size to a syringe filter, may be considered to be regarded as disposable and a fresh column is used for each 90Y elution.

[0054] Thus, 90Y is obtained as Y3+ ions (carrier-free) in a sodium salt solution and the 90Y is in a suitable form for attaching to monoclonal antibodies or for other processing. The separation process is rapid and simple requiring a minimum of steps or chemical additives. The proposed method is also amenable to the production of a 90Y generator that can be ‘milked’ at the point of use to produce 90Y on demand. Only the initial loading of the generator needs to be performed in a hot cell. Since neither the parent or daughter isotopes is a gamma emitter, shielding should not prove problematical allowing high 90Sr activities to be loaded onto a generator. However, some lead shielding will likely be necessary due to the Bremsstrahlung radiation produced by large quantities of 90Y.

[0055] The 90Y product is also unlikely to contain any contaminants, such as Fe3+, that could compete with Y3+ during the synthesis of the 90Y-labelled antibody. Cationic impurities will either be strongly held onto the ion exchange material and not eluted during the 90Y milking, or will be poorly absorbed onto the ion exchanger during the initial 90Sr loading and eluted when the column is washed prior to the first 90Y milking.

[0056] All three of these structures feature anionically charged frameworks with open channels to accommodate exchangeable cations. Six additional ion exchange materials were synthesized: potassium pharmacosiderite, sodium titanosilicate, (where structure is in FIG. 3), sodium nonatitanate (with three degrees of crystallinity), and sodium clinoptilolite.

[0057] All of the compounds listed above have been shown to have a strong affinity for strontium, a key factor for use as an yttrium generator. It is also important that the yttrium daughter is not retained by the ion exchange material. Fortunately Sr2+ and Y3+ have significantly different properties, with the smaller, more highly charged Y3+ surrounding itself with a larger shell of water molecules. This makes it easier for an ion exchange material with a three dimensional structure to discriminate between the two.

[0058] Yttrium is easily precipitated under conditions of low acidity (pH>3 or 4). Most ion exchange materials retain strontium poorly at low pH, so the column must be operated under neutral or basic conditions. Keeping yttrium in solution under these conditions requires the use of a complexant.

[0059] Five compounds known to chelate yttrium were tested to determine the efficacy for keeping yttrium in solution under non-acidic conditions. These compounds, all of which contain at least two carboxylic acid functionalities, are listed in Table 2.

TABLE 2
Compounds tested for solubilizing yttrium
Gluconic Acid
Oxalic Acid
Iminodiacetic Acid
Nitrilotriacetic Acid (NTA)
Citric Acid

[0060] All of the compounds in Table 2 were found to be effective for complexing yttrium to varying degrees. When tested for strontium they were found to be far less effective, with the strontium remaining largely uncomplexed. Adding a complexant was found to improve the separation between the two elements, with the solubility of the yttrium improved under both neutral and basic conditions leading to the strontium remaining sequestered on the ion exchanger while the yttrium remains in solution.

[0061] While this type of basic data is useful for characterizing the materials, it is not sufficient to demonstrate an effective isotope generator. This can be demonstrated best by loading a column with the ion exchange material, loading the ion exchange material with 90Sr, and demonstrating the elution of pure 90Y. In a series of column tests (described in the examples) the four inorganic ion exchange materials described here showing the greatest difference in their affinities for yttrium and strontium ions (sodium titanosilicate, chabazite, clinoptilolite, and potassium pharmacosiderite) were found to be useful for generating 90Y solutions from a 90Sr source. Some of these compounds (chabazite and sodium titanosilicate) were found to be sufficiently effective to serve as generators for 90Y suitable for medical use.

[0062] There are two criteria required of an ion exchange material for use in a 90Y generator. Those criteria are that the material has a much higher affinity for strontium than it does for yttrium and that the material is stable when exposed to intense radiation. All of the materials tested were selected so that they would meet the latter requirement. This is the specific advantage of inorganic ion exchange materials over organic ones.

[0063] All of the ion exchange materials tested also meet the first of these requirements to some degree. Two of the materials, chabazite and sodium titanosilicate, are extremely effective and deliver Sr-free yttrium suitable for use in the preparation of radiopharmaceuticals. Both of those compounds reliably deliver the 90Y produced by the decay of 90Sr isolated within their structures when eluted with a solution containing a complexant. Both elute a solution completely free of 90Sr, a requirement for any medical application of 90Y. The other ion exchangers tested are also useful. Although they are not presently believed to be as effective, they still have utility.

EXAMPLE 1

Synthesis of Potassium Pharmacosiderite

[0064] The starting material for potassium pharmacosiderite was 20 g of silica gel (Aldrich Chemical Company, Milwaukee, Wis., Grade 923) which was initially combined with 80 mL of 10 M KOH. In a separate beaker, 62 mL of titanium isopropoxide (97%, Aldrich), 40 mL of 30% H2O2, 200 mL of deionized water and 60 mL of 10M KOH were added together with vigorous stirring. The contents of the two beakers were then combined, stirred to homogenize the mixture, placed in a 1 L Teflon-lined hydrothermal vessel, and heated at 190° C. for 4 days. The product was centrifuged, washed with absolute ethanol, dried overnight at 60° C. then ground to a fine powder. The structure of the product was confirmed by x-ray powder diffraction (XRD). This material was labeled TA-A-2.

EXAMPLE 2

Synthesis of Sodium Titanosilicate

[0065] To synthesize sodium titanosilicate 33.3 g of tetraethoxyorthosilicate (98%, Aldrich) was combined with 45.6 g of titanium isopropoxide (97%, Aldrich) and added to 260 mL of 6.32 M NaOH. The resultant white gel was stirred well to ensure homogeneity, placed in a 1 L Teflon-lined hydrothermal vessel, and heated at 170° C. for 2 days. The product was centrifuged, washed with absolute ethanol, dried overnight at 60° C. then ground to a fine powder. The structure of the product was confirmed by XRD. This material, the structure of which is shown in FIG. 3, was labeled TA-A-13.

EXAMPLE 3

Synthesis of Sodium Nonatitanate

[0066] Three versions of sodium nonatitanate were prepared. Each of the three started with 77.5 g of titanium isopropoxide (97%, Aldrich) which was placed in a round-bottomed Teflon flask. While stirring constantly, 84.35 g of a 50 wt % solution of NaOH was added resulting in a white gelatinous precipitate. 60 mL of deionized water was added and the mixture was stirred for one hour and then heated at approximately 108° C. for an additional 3 hours with a water cooled reflux condenser used to prevent excessive loss of water. At this stage the raw sodium nonatitanate is ready for further processing.

EXAMPLE 4

Highly Crystalline Sodium Nonatitanate

[0067] One batch of material produced as described in example 3 was then transferred to a 1 L Teflon-lined hydrothermal vessel using 90 mL of deionized water and heated at 170° C. for 7 days. The product was centrifuged, washed with absolute ethanol, dried overnight at 60° C. then ground to a fine powder and designated TA-A-17.

EXAMPLE 5

Poorly Crystalline Sodium Nonatitanate

[0068] A second batch of material produced as described in example 3, designated TA-A-18, received no hydrothermal treatment. It was centrifuged, washed with absolute ethanol, dried overnight at 60° C. and then ground to a fine powder.

EXAMPLE 6

Partially Crystalline Sodium Nonatitanate

[0069] A third batch of material produced as described in example 3, designated TA-A-19, was transferred to a 1 L Teflon-lined hydrothermal vessel using 90 mL of deionized water and heated at 170° C. for 21 hours. The product was centrifuged, washed with absolute ethanol, dried overnight at 60° C., and then ground to a fine powder.

EXAMPLE 7

X-ray Characterization of Products

[0070] All of these materials produced in examples 4, 5 and 6 were characterized by x-ray diffraction to confirm their identity and purity. The diffraction patterns in FIG. 6 show how the crystallinity of sodium nonatitanate improves with hydrothermal treatment. It should be noted that all three patterns were normalized so that the highest peak has the same value. In actual fact, the material treated for a week is a far more efficient diffracter than either of the others.

EXAMPLE 8

Synthesis of Sodium Clinoptilolite

[0071] The sodium clinoptilolite used here was produced by processing a natural material. Purchased raw clinoptilolite was wet sieved (40-60 mesh) and dried overnight at 100° C. A 100 g portion was then contacted with 500 mL 0.1 M NaCl for 45 minutes with gentle agitation. The supernatant was decanted and discarded and the 0.1 M NaCl wash repeated twice. Na+ clinoptilolite was then washed two times with 500 mL Nanopure water with a contact time of 5 minutes and subsequently dried overnight at 100° C. Samples used for screening experiments were ground to a fine powder.

EXAMPLE 9

Preliminary 90Y/90Sr Separations

[0072] An experiment was performed to assess the feasibility of using the inorganic ion exchange materials for separation of 90Y from 90Sr. Three materials were evaluated, namely KTS-Ph, NaTi and clinoptilolite. The KTS-Ph and NaTi had previously been formed into pellets 0.2-0.5 mm in diameter using polyacrylonitrile as a binder. The clinoptilolite was supplied by BNFL Plc. of England and was already in granules suitable for column use. Approximately 1 ml of each material was slurried into a column and 25 mL of a 0.05M NaOH/0.05M NaNO3 solution containing 0.1 mCi of 90Sr passed through the column over a period of approximately 5 minutes. The liquid exiting the column was collected in 5 mL fractions and counted using liquid scintillation counting (LSC). The samples were counted again at a later date and the decrease in total counts recorded. The LSC spectra did not initially suggest the presence of 90Sr (only 90Y) in any of the samples analyzed, indicating the absorption of nearly 100% of the 90Sr by the ion exchangers.

[0073] After allowing the samples to decay, a small amount of 90Sr first became visible in the spectra from the NaTi samples after 605 hours (9.45 half lives of 90Y), by which time the 90Y had decayed to less than 0.2% of it's initial activity. Although it was not possible to quantify the results with any certainty due to the contribution of 90Y produced from the residual 90Sr decay, an activity separation factor of 90Sr/90Y of about 1000 seems likely.

[0074] Liquid scintillation counting does not allow the simultaneous determination of 90Sr and 90Y since the spectra produced by the beta emissions from the two nuclides overlap significantly. Consequently, a solution containing the two isotopes produces a ‘two humped’ spectrum with 90Y at the higher energy end. By counting the samples containing the partially purified 90Y at different time intervals and knowing the half life of 90Y, it is therefore possible to qualitatively note the appearance of 90Sr as the contribution due to 90Y decreases with time. Initially, the peak due to 90Y swamps any minor peak corresponding to 90Sr, but as time progresses and the 90Y decays, the 90Sr component becomes more significant and can be discerned on the scintillation spectrum.

[0075] The pharmacosiderite performed even better than the NaTi with definite 90Sr only being visible after 972 hours or over 15 90Y half-lives, by which time the 90Y had decayed to less than 0.003% of its initial activity. Thus, a 90Sr/90Y separation factor of much greater than 1000 was achieved.

[0076] The clinoptilolite also showed no evidence of 90Sr after 264 hours (about 4 90Y half lives), but these experiments were terminated prior to the appearance of 90Sr.

[0077] Experiments were not performed to study the elution characteristics of the ion exchangers, but it is clear that each of the materials has a high affinity for 90Sr and a low affinity for 90Y in dilute sodium nitrate solutions making them suitable candidates for use in a 90Sr/90Y generator system.

[0078] Although these preliminary experiments did not measure a specific 90Sr/90Y separation factor, the results showed a very high 90Sr/90Y separation suggesting that optimization of the loading and eluting steps of the exchanger or the use of larger ion exchange beds (or multiple beds) will allow the required separation factors of 106 or greater to be readily achieved.

[0079] The ion exchange capacity of the ion exchangers is variable, with clinoptilolite having the lowest capacity. Assuming a capacity of 2 meq/g, a maximum of 1 mmol of Sr2+ can be loaded per gram of ion exchange material. The specific activity of 90Sr is 50 Ci/g and, therefore, 1 mmol equates to 0.09 g. Thus, 9 Ci of 90Sr can be loaded onto 1 g of ion exchange material. This means that ion exchanger consumption will be minimal and thus will constitute a very minor proportion of the generator costs. Using an estimated cost of the $1,000 per kilogram for the ion exchange material, the cost of the ion exchanger will be only $1 per gram, and thus will constitute a negligible part of the total generator cost.

EXAMPLE 10

Evaluation of Ion Exchange Materials

[0080] A total of eight potentially useful ion exchange materials and one potential binder were identified. These are listed in Table 3, along with the abbreviation used in subsequent tables and figures. The first step is to evaluate the affinity of the selected ion exchange materials for strontium as a function of salt concentration.

TABLE 3
Inorganic Ion Exchange Materials Evaluated in Batch Tests
Material                         Abbreviation
Chabazite (commercial product)   AW-500
Potassium pharmacosiderite (synthesized) TA-A-2
Sodium titanosilicate (synthesized) TA-A-13
Sodium nonatitanate (synthesized and hydrothermally TA-A-17
treated for seven days)
Sodium nonatitanate (synthesized with no hydrothermal TA-A-18
treatment)
Sodium nonatitanate (synthesized and hydrothermally TA-A-19
treated for 21 hours)
Sodium nonatitanate (commercial product) Honeywell
Titanium-based binder material (synthesized through Hydro TiO2
the hydrolysis of Ti(i-OPr)4)
Sodium clinoptilolite (commercial product exchanged Clino Na+
into sodium form)

[0081] Samples were evaluated using a simple batch technique to allow the rapid screening of a large number of materials with multiple complexants. Blanks were run for each matrix to check for any loss of strontium/yttrium during filtration or absorption of strontium/yttrium onto the scintillation vials. In all solutions evaluated, strontium absorption in the experimental blanks was negligible.

[0082] In each case 0.05 g of ion exchange materials was contacted with 10 ml of a solution, spiked with either 89Sr or 88Y, in a capped scintillation vial. (Solutions spiked with 88Y were filtered immediately before use to remove precipitated yttrium. Experience with 89Sr has shown that no precipitation occurs and they were not filtered before use.) The mixtures were shaken for 6 hours, filtered through a 0.2 μm syringe filter and the residual activity determined using liquid scintillation counting (LSC). Distribution coefficients (Kd values) were then determined according to the following equation:

Kd=((Ai−Af)/Af)×(v/m)  (2)

[0083] where: Ai is the initial activity in solution (counts per minute/mL)

[0084] Af is final activity in solution (counts per minute/mL)

[0085] v is the volume of the solution (mL), and,

[0086] m is the mass of exchanger (g)

[0087] The final pH of the solution was also noted. Six hours was chosen to allow equilibrium to be reached for each of the ion exchange materials. This period is more than adequate. Previous research has shown that the kinetics for reactions of this type are very rapid and that they generally proceed to >>95% completion in less than five minutes. All experiments were performed in duplicate, and, if significant variations between duplicate samples occurred, the experiments were repeated until good agreements on the Kd values were obtained.

[0088] Table 4 shows how the strontium distribution coefficients vary for eight ion exchange materials as the salt concentration is varied by three orders of magnitude. These results are presented here because this data was used in the design of the rest of the experiments. The Na-clinoptilolite shows the greatest variation with concentration, 3½ orders of magnitude. The strontium selectivity of both of the zeolites decreased significantly in higher ionic strength solutions, thus limiting their use to less saline solutions. (This is actually an advantage, because a requirement for high electrolyte content, e.g., high salinity, could add complexity to later steps in the preparation of the final pharmaceutical.) Most materials showed one order of magnitude of variation or less.

TABLE 4
Effect of NaCl Concentration on the Distribution
Coefficient for Strontium
Ion Exchange
Material	1 M NaCl	0.1 M NaCl	0.01 M NaCl	0.001 M NaCl
Clino Na+	8	124	3,260	36,900
AW-500	1,860	88,300	1,270,000	1,210,000
TA-A-13	556,000	273,000	119,000	42,900
TA-A-2	18,300	251,000	594,000	281,000
Honeywell	80,600	1,030,000	258,000	166,000
TA-A-18	1,530,000	2,570,000	739,000	372,000
TA-A-19	1,030,000	1,240,000	272,000	172,000
TA-A-17	167,000	834,000	264,000	90,400

EXAMPLE 11

Effectiveness of Complexing Agents

[0089] To evaluate the effectiveness of the complexant 0.45 μCi of 88Y (a β+ [positron] emitter with a 107 day half-life) was added to a series of solutions each 0.01 M in NaCl concentration and 0.001 M in one of the nine complexants. (A blank was also run.) The test was repeated six times with the pH adjusted to a different value each time. In each case the solution was stirred for an hour and filtered. The amount of yttrium remaining in solution was determined by LSC of the filtered solution. The effectiveness of the complexants was evaluated by comparing the amount of yttrium remaining in solution with the complexant with the amount remaining in solution without any complexant. Seven of the complexants were found to improve the solubility of yttrium and five of these were selected for further study. Although both EDTA and HEDTA were found to be quite effective in this test, they were eliminated from future testing because they also suppress the retention of strontium on the ion exchange material. The complete results appear in Table 5.

TABLE 5
Fraction of 88Y Remaining in Solution with each Complexant
	pH 5	pH 7	pH 9	pH 10	pH 12	pH 13
Blank	78%	64%	61%	46%	61%	34%
D-Gluconic Acid	84%	85%	81%	74%	89%	87%
Oxalic Acid	110%	111%	103%	93%	42%	16%
Glycolic Acid	66%	57%	56%	55%	79%	20%
Iminodiacetic Acid	94%	93%	71%	49%	63%	72%
Nitriloacetic Acid	97%	98%	85%	99%	107%	60%
Citric Acid	91%	98%	99%	96%	103%	34%
Acetic Acid	49%	25%	14%	19%	24%	28%
HEDTA	92%	104%	92%	86%	100%	89%
EDTA	91%	103%	84%	100%	100%	95%
Note, balues over 100% indicate the magnitude of the experimental error in these measurements.

EXAMPLE 12

Distribution Coefficients in the Presence of Complexing Agents

[0090] The results shown in Table 4 and Table 5 supply the basis for the testing carried out in this and subsequent examples. The formulae and structures of the complexants selected for the next series of experiments are shown in FIG. 6.

[0091] The strontium and yttrium selectivity of the chosen ion exchange materials were evaluated in six different solutions (0.01M NaCl, 0.01M NaCl/0.001M gluconic acid, 0.01M NaCl/0.001M oxalic acid, 0.01M NaCl/0.001M iminodiacetic acid, 0.01M NaCl/0.001M nitrilotriacetic acid, and 0.01M NaCl/0.001M citric acid) using radiotracer techniques. EDTA and HEDTA were not evaluated because previous studies had indicated that the ion exchange materials had a reduced affinity for strontium in EDTA and HEDTA solutions. The other complexants did not significantly decrease the strontium affinity.

[0092] The distribution coefficients for both yttrium and strontium were evaluated using the same batch technique with radiotracers as was described in example 10, above. The six hour equilibration time allowed in these experiments is clearly more than adequate. Previous research has shown that the kinetics for reactions of this type are very rapid and that they generally proceed to >>95% completion in less than five minutes.

[0093] Table 6 shows the results of the batch tests for strontium with each of the five selected complexants and with no complexant. In all cases the strontium was strongly absorbed onto the ion exchange material. The poorest performer was the material intended to serve as the binder.

[0094] Table 7shows the results of the batch tests for yttrium with each of the five selected complexants and with no complexant. In general, yttrium was not retained as effectively as strontium. This is the desired result.

[0095] Ideally we would like to have an ion exchange material with an exceptionally high affinity for strontium and no affinity for yttrium. Fortunately for ion exchange purposes strontium is coordinated by water much more loosely than yttrium. This means that in solution the more highly charged yttrium is effectively a much larger cation than the strontium, and has a greater tendency to be excluded from sites inside the cage or layer structure of the ion exchange materials.

TABLE 6
Distribution Coefficients (Kd) for Strontium
	Complexant
				Imino-	Nitrilo-
Ion		Gluconic	Oxalic	diacetic	triacetic	Citric
Exchangers	None	Acid	Acid	Acid	Acid	Acid
AW-500	20,750	14,072	14,544	15,289	6,447	13,236
TA-A-2	19,789	17,617	14,425	15,591	2,734	14,669
TA-A-13	16,746	11,431	13,809	14,839	10,859	14,899
TA-A-17	19,728	17,214	17,464	16,326	11,710	16,520
TA-A-18	20,761	18,430	14,859	16,146	16,995	16,451
TA-A-19	20,643	17,973	17,651	16,572	16,573	13,022
Honeywell	21,091	18,440	15,195	17,562	10,014	18,178
Hydro TiO2	21.40	825.1	782.4	66.10	169.0	1,213
Clino Na+	301.6	533.7	910.0	283.8	314.6	180.4

[0096]

TABLE 7
Distribution Coefficients (Kd) for Yttrium
	Complexant
				Imino-	Nitril-
Ion		Gluconic	Oxalic	diacetic	triacetic	Citric
Exchangers	None	Acid	Acid	Acid	Acid	Acid
AW-500	6,772	87.20	36,252	6,320	1.00	230.6
TA-A-2	11,209	1,471	31,141	9,554	2.25	29,897
TA-A-13	6,763	233.0	20,429	8,041	60.65	20,592
TA-A-17	25,616	1,691	67,241	15,678	7,408	73,261
TA-A-18	23,330	2,443	115,936	20,650	13,468	97,918
TA-A-19	31,606	4,367	93,683	22,510	13,528	40,210
Honeywell	21,279	8,223	68,311	20,673	361	40,000
Hydro TiO2	254,917	36,247	247,610	61,410	638	36,702
Clino Na+	10,755	111.0	42,944	204.5	0.35	38.80

[0097] The key to selecting a combination of ion exchange material and complexant is maximizing the ratio of the strontium Kd to the yttrium Kd. The complete set of ratios appears in Table. It is clear from this data that nitrilotriacetic acid (NTA) produces the highest ratio in every case where Sr is substantially more tightly bound than Y.

TABLE 8
Ratio of Distribution Coefficients (Kd) for Strontium and Yttrium
	Complexant
				Imino-	Nitrilo-
Ion		Gluconic	Oxalic	diacetic	tracetic	Citric
Exchangers	None	Acid	Acid	Acid	Acid	Acid
AW-500	3.06	161	0.40	2.42	6,447	57.4
TA-A-2	1.77	12.0	0.46	1.63	1,215	0.49
TA-A-13	2.48	49.0	0.68	1.85	179	0.72
TA-A-17	0.77	10.2	0.26	1.04	1.58	0.23
TA-A-18	0.89	7.54	0.13	0.78	1.26	0.17
TA-A-19	0.65	4.12	0.19	0.74	1.23	0.32
Honeywell	0.99	2.24	0.22	0.85	27.8	0.45
Hydro TiO2	0.00	0.02	0.00	0.00	0.26	0.03
Clino Na+	0.03	4.81	0.02	1.39	899	4.65

[0098] The data from Table 8 is illustrated graphically in FIG. 8. The data has been plotted on a log scale to render the relative size of the bars at Sr:Y ratios less than 100 (102) more apparent. (If plotted on a linear scale, only three bars are visible.) Bars extending upward from the central plane (at a ratio of 100, or 1) indicate combinations that favor the retention of strontium over yttrium. Bars extending downward from the plane indicate combinations favoring the retention of yttrium over strontium.

[0099] It is clear from these results that NTA, citric acid, and gluconic acid are the most effective complexants, improving the ion exchanger's preference for Sr over Y in most cases. As shown in both the table and the figure, most of the ion exchange materials had little preference for either ion in the absence of a complexant. (Oxalic acid causes the ion exchangers to prefer Y to Sr, possibly due to a precipitation reaction.) Based on these results four ion exchangers were selected for column testing, chabazite (AW-500), sodium titanosilicate (TA-A-13), clinoptilolite, and potassium pharmacosiderite (TA-A-2).

EXAMPLE 13

Preparation of Generator Columns

[0100] Materials produced as powders were pelletized using an inorganic binder and all materials sized with a 40-60 mesh portion selected for testing.

[0101] For each of the ion exchangers to be tested sufficient material to fill a 1 mL-bed volume of 40-60 mesh sized particles was slurried in deionized water and poured into ion exchange columns (internal diameter 0.7 cm). The column beds were then washed by passing 50 mL of deionized water through each to remove any remaining fines. To prevent the escape of any pelletized ion exchange material which might be released during the course of experiments due to mechanical degradation of the pelletized material 0.2 μm syringe filters were then attached to the effluent end of each column.

[0102] Two columns were prepared for three of the ion exchange the materials (sodium titanosilicate, chabazite, and clinoptilolite) with three columns prepared with the potassium pharmacosiderite.

[0103] Individual columns were loaded with 90Sr by passing a 20 mL 0.01 M NaOH solution spiked with 0.1 mCi 90Sr/0.05 mCi 85Sr through each at a flow rate of approximately 20 mL/hr. The flow rate through each column was maintained by a peristaltic pump that pulled solutions though columns and into collection vessels through {fraction (1/16)}″ i.d. Tygon tubing. Strontium uptake was found to be virtually quantitative in all cases. Each column was then washed by passing 20 mL 0.01 M NaCl through the column then bringing the liquid level inside the column down to the top of the ion exchange bed. Columns were then stored until their respective weekly elution.

[0104] FIG. 9 shows one of the ion exchange columns 10 having an effluent filter 12 and filled with pellets 14 of an ion exchange material, such as sodium titanosilicate.

EXAMPLE 14

Column Tests

[0105] Seven days after the initial loading, a weekly elution routine was established for each column and maintained until data collection was complete. For each elution, a 10 mL aliquot of chosen complexant/eluant (pH was varied throughout course of project) was passed through each column at a flow rate of 80 mL per hour. Portions of the resulting 10 mL eluate were analyzed by both liquid scintillation counting (1 mL) and gamma counting (9 mL) and the final pH of the solution recorded.

[0106] Following elution, each column was washed by passing 20 mL of 0.01 M NaCl (pH 6.1±0.05) through each at a flow rate of 80 mL per hour. The resultant 20 mL wash eluate was analyzed by liquid scintillation counting (1 mL) and gamma counting (19 mL) and the pH recorded. Columns were then stored until their next cycle of weekly elutions/washing were performed.

[0107] Each column was milked on a weekly basis and the ingrown 90Y eluted using a series of eluants.

EXAMPLE 15

Column Tests with Chabazite

[0108] The procedures described in example 14 were carried out using chabazite (AW-500). FIG. 10 shows the 90Y yields obtained with a series of three eluants, 0.001 M NTA in 0.01 M NaCl (for the first three weeks), 0.01 M NaCl alone (week four), and 0.001 M (weeks five and eight) and 0.003 M citric acid in 0.01 M NaCl (weeks six and seven).

[0109] It's clear that both NTA and citric acid are effective complexants for the selective elution of 90Y from a 90Sr cow. The results shown here were obtained by scintillation counting to measure the β-decay activity of the 90Y isotope. 90Sr is also a β emitter making it impossible to determine if trace amounts of 90Sr are being lost from the column. When 90Sr gets into the human body it rapidly deposits in bones and due to its long half-life, it is a major health risk that must be avoided, and can be avoided by insuring that a 90Y generator is sufficiently selective. By adding 85Sr, a γ-emitter, to the 90Sr loaded on the column, even a small loss of strontium during milkings can be measured using a gamma counter. Using this technique there was no evidence of any strontium loss from the AW-500.

EXAMPLE 16

Column Tests with Sodium Titanosilicate

[0110] The procedures described in example 14 were carried out using chabazite (AW-500). FIG. 11 shows the milking results using sodium titanosilicate carried out with the eluents initially in the same order as described in example 15. This ion exchanger performed well with NTA, but not with citrate. Because of its consistent poor performance with citrate as the complexant it was also tested with gluconic acid as the complexant. This complexant failed to produce an improvement. Like chabazite, there was no evidence of Sr leaching.

EXAMPLE 17

Column Tests with Clinoptilolite

[0111] The procedures described in example 14 were carried out using clinoptilolite. FIG. 12 shows the 90Y milking results for the columns loaded with clinoptilolite carried out with the eluents in the same order as described in example 15. The results look quite good, but there is a problem. FIG. 13 shows the results for leaching Sr from the cow. Although the levels of 85Sr detected are low, the release is unacceptable if clinoptilolite is to be used in a 90Y generator. This loss of strontium concurs with the data shown in Table, which demonstrates the low affinity of clinoptilolite for strontium in comparison with the other ion exchange materials evaluated.

EXAMPLE 18

Column Tests with Potassium Pharmacosiderite

[0112] The procedures described in example 14 were carried out using potassium pharmacosiderite. FIG. 14 shows the 90Y elution results for potassium pharmacosiderite carried out with the eluents initially in the same order as described in example 15. Like sodium titanosilicate, this material is an effective absorbent for 90Sr when NTA is used as the complexant, but less effective with citrate. Because of its consistent poor performance with citrate as the complexant, it was also tested with gluconic acid as the complexant. This complexant failed to produce an improvement. Unlike sodium titanosilicate, it also leaches a small amount of strontium with NTA, as shown in FIG. 15. Although the amount of strontium lost is an order of magnitude less than observed for clinoptilolite, it is still too great to allow the material to be used in a 90Y generator.

EXAMPLE 19

Stability of the Inorganic Ion Exchange Material

[0113] Examples 15 through 18 above are directed to demonstrating how well an inorganic ion exchange column performs for the primary task in a 90Y generator. It is also important to know how a column will mechanically degrade with use. To determine this two columns were fabricated, one with chabazite (AW-500) containing a quadruple loading (4 mL bed volume) and the other with sodium titanosilicate (standard 1 mL bed volume). They were each connected to a peristaltic pump and subjected to a flowing stream of 0.003 M citric acid in 0.01 M NaCl making a single pass through the bed at a flow rate of 80 mL/h for over 24 hr. Samples of the solution were collected and analyzed by atomic absorption spectroscopy (AA) for Si, Al, and Ti.

[0114] The results from the stability tests carried out on the unloaded columns appear in Table 9. The values given are the concentration of the metal ions observed in the complexant solution after passing through the column. The higher than expected levels of Al and Si are attributed to the clay binder used in the fabrication of the pellets. This is a feature that can be changed by starting with unpelletized chabazite and using a more stable binder. At the concentrations being evaluated here, AA is of marginal utility in measuring the concentrations of these ions and the results are only approximate. It is safe to say that there is little, if any, extraction of ions from the ion exchange material occurring.

TABLE 9
Ions in Solution
	Al	Si	Ti
	(ppm)	(ppm)	(ppm)
	Blank	n.d.	n.d.	n.d.
	AW-500	27	35	n.d.
	TA-A-13	n.d.	n.d.	n.d.
	Note: n.d. means “not detectable”

[0115] While the foregoing is directed to the preferred embodiment of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow. It will be understood from the foregoing description that various modifications and changes may be made in the preferred embodiment of the present invention without departing from its true spirit. It is intended that this description is for purposes of illustration only and should not be construed in a limiting sense. The scope of this invention should be limited only by the language of the following claims.

[0116] In the present specification “comprises” means “includes” and “comprising” means including” and these terms do not exclude the involvement other components or steps.

[0117] The features disclosed in the foregoing description, or the following claims, or the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for attaining the disclosed result, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.

Claims

1. A process, comprising: adsorbing strontium-90 onto an inorganic ion exchange material from an aqueous solution comprising a source of strontium-90; and eluting yttrium-90 from the inorganic ion exchange material with a solution having a pH greater than about 5 and including a chelating agent.

2. The process of claim 1, wherein the solution is aqueous.

3. The process of claim 1, wherein the chelating agent is selected from gluconic acid, oxalic acid, iminodiacetic acid, nitrilotriacetic acid, citric acid, and combinations thereof.

4. The process of claim 1, where the solution is neutral.

5. The process of claim 1 where the solution is alkaline.

6. The process of claim 1, wherein the inorganic ion exchange material is selected from clinoptilolite, chabazite, potassium titanosilicate pharmacosiderite, sodium titanosilicate, sodium nonatitanate, and combinations thereof.

7. The process of claim 6, wherein the aqueous solution includes a chelating agent selected from gluconic acid, oxalic acid, iminodiacetic acid, nitrilotriacetic acid, citric acid, and combinations thereof.

8. The process of claim 1, wherein the inorganic ion exchange material is sodium nonatitanate prepared by reacting titanium isopropoxide and aqueous sodium hydroxide at a temperature between 100° C. and 250° C. for a period between 12 hours and 2 weeks.

9. The process of claim 1, wherein the inorganic ion exchange material is sodium titanosilicate prepared by hydrothermally heating a titanium silicate gel in NaOH.

10. The process of claim 9, wherein the titanium silicate gel is hydrothermally heated in 6M NaOH at 170° C. for 2 days.

11. The process of claim 1, wherein the inorganic ion exchange material is a titanosilicate having the general formula:

12. The process of claim 1, further comprising forming the inorganic ion exchange material into pellets.

13. The process of claim 12, wherein the pellets are formed with an average diameter between 0.1 and 1.0 mm.

14. The process of claim 12, wherein the pellets are formed with an average diameter between 0.2 and 0.5 mm.

15. The process of claim 12, wherein the pellets comprise a polymeric binder.

16. The process of claim 15, wherein the polymeric binder comprises polyacrylonitrile as a binder.

17. The process of claim 12, wherein the pellets comprise an inorganic binder.

18. The process of claim 17, wherein the inorganic binder is amorphous.

19. The process of claim 17, wherein the pellets comprise an inorganic binder selected from amorphous titanium dioxide, amorphous silica, amorphous zirconium oxide, and combinations thereof.

20. A process, comprising: (a) preparing a first solution including strontium-90; (b) adsorbing strontium-90 from the first solution onto an inorganic ion exchange material; and (c) allowing yttrium-90 to grow into the inorganic ion exchange material; and (d) eluting yttrium-90 from the inorganic ion exchange material with a second solution including a chelating agent.

21. The process of claim 20, further comprising: repeating steps (c) and (d).

23. The process of claim 20, wherein the inorganic ion exchange material is selected from clinoptilolite, chabazite, potassium titanosilicate pharmacosiderite, sodium titanosilicate, sodium nonatitanate, and combinations thereof.

24. The process of claim 20, wherein the first solution has a pH greater than about 5.

25. The process of claim 20, wherein the second solution has a pH greater than about 5.

26. The process of claim 20, where the second solution is neutral.

27. The process of claim 20, where the second solution is alkaline.

28. The process of claim 20, wherein the second solution includes a chelating agent.

29. The process of claim 28, wherein the chelating agent is selected from gluconic acid, oxalic acid, iminodiacetic acid, nitrilotriacetic acid, citric acid, and combinations thereof.

30. The process of claim 20, further comprising: (e) washing the inorganic ion exchange material to remove poorly bound strontium-90 prior to eluting.