1. Introduction
Technetium-99m (
99mTc) is used in 85% of nuclear medicine diagnostic imaging procedures worldwide [
1,
2,
3,
4]; almost all
99mTc used for this purpose is obtained from the radioactive decay of molybdenum-99 (
99Mo), which is produced by processing irradiated highly enriched uranium (HEU) targets [
5]. The use of HEU raises concerns from a proliferation perspective. Furthermore, the reactors used for the production of
99Mo are susceptible to uncertainty due to planned and unplanned shutdowns and decommission [
6]. The increasing demand for
99mTc has prompted efforts to solve the supply shortages. Under such circumstances, several alternative methods have been proposed, such as reactor-based solutions, accelerator-based solutions, and the conversion of HEU to LEU (low-enriched uranium) [
7,
8,
9]. The use of HEU also has a problem with plutonium generation, which makes its use difficult for signatory countries of the NPT, such as Japan.
98Mo used to produce
99Mo by the (n,γ) reaction was proposed as a promising solution, which has inspired further development, with the advantages of radioactive waste, cost reduction, and non-proliferation [
10]. The neutron activation of
98Mo by the reaction (
98Mo(n,γ)
99Mo) has a small activation cross-section (0.13 b for thermal neutrons) compared with the neutron fission production of
99Mo (584 b for thermal neutrons) [
11]. From this perspective, it is required to increase the yield of
99Mo, and a variety of research has been carried out towards this objective (e.g., increasing the enrichment of
98Mo, and increasing the density of the pellet) [
12,
13].
The common practice is to use MoO
3 targets of natural abundance, which is economically preferred [
8]. However, the Mo/Tc ratio in MoO
3 is usually high, which makes it difficult to use conventional generators to extract
99mTc. To overcome this issue, the
99Mo is separated from the MoO
3 target by dissolving it in an alkali solution and passing it through an adsorbent material in a column [
14]. Because of the advantage of increased
99Mo production yield, the use of
98Mo target material with enrichment of >95% has also drawn wide attention [
5]. Meanwhile, the relatively high cost of highly enriched
98Mo requires more effort to recover the unutilized
98Mo of the target. The crystalline density of MoO
3 is about 4.7 g/cm
3; MoO
3 has limited solubility in water (1 g/L at 20 °C) and can be easily dissolved in sodium hydroxide (NaOH). The obtained solution can be further processed by separation techniques (e.g., column chromatography and solvent extraction). Mo metal is another typically used target, which can be dissolved in alkaline hydrogen peroxide (H
2O
2) or electrochemically. However, another oxide, molybdenum dioxide (MoO
2), has been comparatively much less explored as a target.
MoO
2 is almost insoluble in water [
15,
16] and contains more
98Mo per volume compared with MoO
3. This insolubility provides the possibility of processing the irradiated target by water, rather than dissolving the entire quantity of molybdenum oxide in the alkali solution. This technique proposes a porous structure and selectively dissolves only the
99mTc located on the surface. The solution can be further handled, and the undissolved target can be recovered. This process takes advantage of the large surface and the short diffusion distance from the inside of the solid to the surface, which allows for efficient extraction of the
99mTc. On the other hand, MoO
2 is less soluble than MoO
3, which means that it is more likely to be adsorbed onto the column and less likely to be eluted with the Tc-99m. This can reduce the Mo/Tc ratio in the generator and increase the yield and quality of the Tc-99m, which is of benefit to the separation operation of molybdenum–technetium and the recycling of the molybdenum target.
Porous oxide materials have garnered significant attention in various scientific and industrial applications due to their properties, such as high surface area and a tunable pore size [
17]. One widely explored method for the fabrication of these materials involves the use of pore formers [
18]. Pore formers, also known as sacrificial templates, porogens, or space holders, are materials intentionally added during the fabrication of porous materials to create voids or pores within the final structure [
19]. These pores can range from nanometers to micrometers in size and play a crucial role in determining the material’s properties and applications. The advantages of using space holders in porous materials are manifold. They allow precise control over pore size, shape, and distribution, enabling tailored materials for specific applications like catalysis adsorption and drug delivery [
20]. Furthermore, these methods can enhance the overall surface area and porosity of the oxide, which can improve the material’s reactivity and performance.
In this study, potassium molybdate (K2MoO4) is used as the space holder in the pore-forming process. The selection of K2MoO4 as a porogen is driven by its exceptional attributes, which make it a compelling choice for creating well-defined pores in MoO2 pellets, all while preserving their structural integrity and chemical composition. One of its standout features is its relatively high melting point, which is compatible with MoO2, the core material of interest. Moreover, the introduction of only the element potassium as a new component brings a distinctive aspect to the composition, contributing to the material’s unique properties. Additionally, there is a significant contrast in solubility in water when compared to MoO2, further enhancing its effectiveness as a porogen in the pore-forming process. These combined characteristics make K2MoO4 a standout candidate for our study, ensuring precise control over pore formation while maintaining the integrity of the resulting oxide material.
Hence, we propose to develop a porous MoO2 pellet target as a complement to the (n,γ) reaction target selection.
In our previous work, we conducted exploratory experiments to select the appropriate pore former and determine the sintering conditions. Building on this foundation, the current study focuses on the utilization of MoO
2 powder mixed with potassium molybdate (K
2MoO
4), which offers several advantageous properties, as mentioned above. Notably, the melting point of MoO
2 is 1100 °C [
21], while K
2MoO
4 exhibits a lower melting point of 919 °C. Furthermore, K
2MoO
4 demonstrates excellent solubility, which is 164.5 g/100 g H
2O at 25 °C.
The fabrication involved the process of sintering, followed by the formation of pores through dissolving the pore formers. Subsequently, we subjected the porous pellets to thorough characterization.
2. Materials and Methods
2.1. Manufacture of Porous MoO2 Pellet
The molybdenum(Ⅳ) oxide (Strem Chemicals, 99%) powders were mixed with potassium molybdate (FUJIFILM Wako Pure Chemical Industries, Ltd., Osaka, Japan); zinc stearate (FUJIFILM Wako Pure Chemical Corporation, 12.0~14.5%) was added as a lubricant for sintering, which decomposed after the sintering temperature reached 300~400 °C. The mixed compounds were weighted (Azpro Analytical Balance 120g ASR124/E, Minimum measurement: 0.0001 g) to reach the expected ratio (70% MoO2 and 30% K2MoO4), and weighted powders were ground using the agate mortar and pestle since the fine powders can benefit the sinter. Furthermore, the ground powders were filled into the pressing die (Φ7 mm, δ1 mm), by which the pellet form was obtained. The dark and dry pellet was sintered using a tube furnace (TMF-300N/-500N/-700N, AS ONE Corporation, Osaka, Japan).
In addition, by taking advantage of the extreme insolubility of molybdenum(Ⅳ) oxide in water, a compound that is easily soluble in water can be taken for consideration as a pore-forming compound. The selection of the pore former requires a higher solubility in water and is unreactive with MoO2, while potassium has an extremely high solubility in water, which is 164.5 g in 100 g water at 25 °C. Simultaneously, the pore former also requires a high melting point, so that they can be sintered under high temperatures. The melting point of MoO2 is 1100 °C and 919 °C for K2MoO4. The sinter was carried out under 900 °C for 6 h, which is below the melting point of the smaller pellet composition.
The sintered pellet was eluted with 100 mL ultrapure water, by which the K2MoO4 can be easily dissolved, and the pores were formed. The eluted pellet was dried under a vacuum.
The manufacturing process of porous MoO2 pellets is as follows:
Mixing of MoO2 and pore former powders.
Grinding to fine powders.
Pressing to form a pellet.
Sintering.
Eluting and drying.
In the first two stages, MoO2 was combined with K2MoO4 powders, and the mixture was finely ground before being compressed to form the desired structure. These initial steps were crucial to enhance the sintering process.
For the sintering process itself, the temperature profile was carefully controlled. Initially, the temperature was set to 400 °C and maintained for 1 h to facilitate the decomposition of zinc stearate, a component within the mixture. Following this, the temperature was raised to 900 °C and held for 6 h. This stage promoted grain growth and facilitated the binding of particles to one another.
In the final step, due to the disparity in solubility between MoO2 and K2MoO4, K2MoO4 was selectively dissolved into water, leaving behind pores within the MoO2 pellet.
In these steps, precise control over the pore system can be achieved by adjusting the grinding time, altering the ratio of mixtures, or fine-tuning the sintering conditions. This level of control allows us to tailor the resulting pore structure to meet the specific requirements and desired characteristics for our application.
2.2. Characterization of Porous MoO2 Pellet
A micrometer, a caliper, and an electric balance were used to physically measure the obtained pellet. The X-ray diffraction (XRD) apparatus (RINT-2500HF+/PC, Rigaku Corporation) and scanning electron microscope–energy dispersive spectroscopy (SEM-EDS, JEOL JCM-6000Plus/EDS, Melville, NY, USA) were utilized for characterization.
4. Discussion
Throughout the experiment, extensive efforts were devoted to identifying suitable pore formers and optimizing sintering conditions. Ultimately, the production of a porous MoO2 target was successfully accomplished by sintering at 900 °C using a pore former, followed by elution with ultra-pure water. This sintering temperature aligns closely with the ideal sintering temperature for MoO2, which reaches 81.8% of the melting point of MoO2. In the pellet fabrication process, MoO2 and K2MoO4 are added in a ratio of 7:3. When considering the relative density, the calculated quantities for producing a pellet with a 7 mm diameter and 1 mm thickness were 0.1220 g of molybdenum dioxide and 0.0354 g of potassium molybdate. The resulting sintered target exhibited dimensions of 7.0 mm in diameter and 1.1 mm in thickness. The weight of the target obtained after the elution process with water to form pores and drying in a vacuum was 0.1214 g. Remarkably, this weight closely matches that of the initially added molybdenum dioxide. This outcome demonstrates that during the pore formation process, potassium molybdate primarily dissolves in ultra-pure water, while molybdenum dioxide remains in the pellet due to its extremely insolubility in water.
Our findings were further validated through X-ray diffraction (XRD) and scanning electron microscope (SEM) analyses. The XRD results revealed that the peak positions of the obtained sample are well matched to those of the molybdenum dioxide PDF card, and that only a small part of the peaks corresponded to the potassium molybdate PDF card; these peaks exhibited low intensity, which means potassium molybdate was well removed. SEM imaging effectively demonstrated significant structural alterations in the pellet both before and after the elution process. After the elution, pores became conspicuously apparent, thereby affirming their formation because of the dissolution of potassium molybdate. These pores play a critical role in expanding the material’s surface area and reducing diffusion distances. However, it is worth noting that an excessive formation of pores carries the risk of compromising the material’s structural integrity.
The utilization of porous MoO2 pellets presents several notable advantages, as evidenced by the experimental results. These advantages encompass enhanced selective dissolution, improved generator yield, efficient technetium extraction, production stability, and target recyclability. These benefits collectively address supply challenges, elevate the quality of 99mTc production, and provide a promising solution for stable and efficient production in nuclear medicine applications. The MoO2 targets with well-formed pores achieved through this method have great potential in molybdenum–technetium separation and target recycling for the stable and efficient production of 99mTc.