Preparation and characterization of sol-gel derived (Th_xCe_1−x)O₂ microspheres

1 Introduction

Different combinations of fissile and fertile materials can be used in various reactor systems. In self-sustaining equilibrium thorium cycles, no external fissile material is required at equilibrium. Such cycles can start with thorium and any suitable externally supplied fissile material (²³⁵U, ²³⁹Pu) [1], [2].

Thorium fuels and fuel cycles are particularly relevant to countries having large thorium deposits but very limited uranium reserves for their long-term nuclear power program. The feasibility of thorium utilization in high temperature gas cooled reactors (HTGR), light water reactors (LWR), pressurized heavy water reactors (PHWRs), liquid metal cooled fast breeder reactors (LMFBR) and molten salt breeder reactors (MSBR) were demonstrated. These activities have been well documented in several extensive reviews and conference proceedings published by US Atomic Energy Commission [3], US Department of Energy [4], [5], KFA, Germany [6] and IAEA [7]. More recently, the proceedings of IAEA meetings on thorium fuel utilization: options and trends has summarized the activities and coordinated research projects (CRP) of IAEA and the status of thorium fuel cycle option, including accelerator driven systems, in member states [8].

Thorium is three to four times more abundant than uranium. Thorium fuels, therefore, complement uranium fuels and ensure long-term sustainability of nuclear power. The interest in ThO₂–PuO₂ fuel is a long-term one as starter fuel for a ThO₂–²³³UO₂ fuel cycle. The absorption cross-section of ²³²Th (7.4 barns) for thermal neutrons is nearly three times that of ²³⁸U (2.7 barns). Hence, a higher conversion (to ²³³U) is possible with ²³²Th than with ²³⁸U (to ²³⁹Pu). Thus, thorium is a better ‘fertile’ material than ²³⁸U in thermal reactors. Thorium dioxide is chemically more stable and exhibits higher radiation resistance than uranium dioxide. The fission product release rate for ThO₂-based fuels is one order of magnitude lower than that of UO₂. ThO₂ has favorable thermo-physical properties because of the higher thermal conductivity and lower co-efficient of thermal expansion compared to UO₂. Thus, ThO₂-based fuels are expected to have better in-pile performance than UO₂ and UO₂-based mixed oxide [9].

Work with plutonium involves specific precautions and necessitates complex equipment. On the other hand, most of the countries are deprived of plutonium. For these inconveniences, plutonium is simulated by cerium in the maintained studies on the (Th, Pu)O₂ fuel production in many laboratories around the world. It is well known that ThO₂, UO₂, PuO₂ and CeO₂ have the same crystallographic structure (CaF₂ type, cubic). Therefore, mixed in any ratio, they form solid solutions [10].

Because of wide applications in the production of good quality ceramics around the world, sol-gel processes have been extensively studied for the production of thorium based nuclear fuel microspheres since 1960s. Various studies on thorium fuel cycle and sol-gel processes have been reported in different symposium proceedings [11], [12], [13]. Although the main aim of the sol-gel processes is to produce uniform and dense microspheres of thorium and uranium available to use in high temperature reactors (HTRs), sol-gel microspheres can be used also as feed material for pellet type nuclear fuels. The use of microspheres as a feed material for the pellet type fuels provides important production advantages, including minimal generation of dust and production of highly flowable materials. These properties are very important for large-scale production of oxide powders that are subsequently pelletized. Mixed-oxide microspheres produced via sol-gel process have a generally homogeneous composition, resulting from co-precipitation of heavy metals. In this case, the solid solution formation of mixed oxides is easily attainable [14], [15], [16].

In the present study preparation conditions of (Th_xCe_1−x)O₂ microspheres via sol-gel process were investigated and the obtained mixed oxides were characterized.

2 Experimental

Analytical grade Th(NO₃)₄·5H₂O and (NH₄)₂[Ce(NO₃)₆] (Merck) were used as initial materials. All reagents (ammonia, methyl isobutyl ketone) used in the experiments were purchased from Merck and were AnalaR grade. Ultrapure water (resistivity 18.2 MΩ·cm, TOC level 1–5 ppb) was prepared by the Millipore model water purification system including Elix and Mili-Q, and used in the experiments.

In the present study (Th_xCe_1−x)O₂ gel microspheres were prepared according to the process flowsheet given in Figure 1.

Figure 1:

Process flowsheet for synthesis and characterization of (Th_xCe_1−x)O₂ gel microspheres.

2.1 Sol preparation

Th(NO₃)₄ and (NH₄)₂[Ce(NO₃)₆] mixed nitrate solutions were prepared at different combinations to obtain final mixed oxide ratios of (Th_xCe_1−x)O₂ (with x=0.50, 0.75 and 0.95 mol %). In all experiments, the total metal concentration was taken as 1.6 mol/L prior to neutralization. By following the partial neutralization with ammonia, the total metal concentration became finally ~1.0 mol/L. Source sols were neutralized up to about 85% of partial neutralization by using 8 M NH₄OH. The sol preparation system is shown in Figure 2.

Figure 2:

Sol preparation apparatus.

The sols were prepared in a production vessel heated to (90±1)°C on a temperature controlled magnetic stirrer. A titroprocessor (Metrohm 686) was used for ammonia injection under pH control. For pH monitoring, a very small part of sol was cooled to 25°C and circulated from the production vessel to the pH monitoring section and then returned.

The addition of ammonia was continued when the monitored pH was lower than a preset value (pH=3.15), thus the pH increases with ammonia injection and decreases with conversion into colloids. The neutralization was advanced stepwise.

2.2 Gelation

Gelation was carried out in the apparatus shown in Figure 3. The gelation system consists of a stainless steel nozzle attached to a vibration system allowing passage of the sol to the inner part of the gelation column, where it dispersed into droplets. The sol drop formation system was horizontally placed at the top of the gelation column. A peristaltic pump was used for introduction of sols to gelation system. The gelation column includes two phases. The upper phase is hexone presaturated with ammonia to provide good sphere formation and serves as the prehardening medium to obtain microspheres with the desired properties. The lower phase where the microspheres are gelated completely is 8 M NH₄OH. The resulting gel microspheres are collected in a reservoir connected to the bottom of the column and aged in 8 M NH₄OH for 48 h.

Figure 3:

Microsphere production apparatus.

The aged gel microspheres were placed in a washing column and washed by diluted ammonia (1% v/v) until complete removal of NH₄NO₃ formed from the coprecipitation reaction of thorium and cerium. This avoids problems that can arise in the further heat treatment steps. For complete removal of NH₄NO₃, the quantity of washing solution used was 15-fold in excess of volume of microspheres. The washed microspheres were dried at 80°C for 72 h.

3 Results and discussion

In the present study, it was seen that low partial neutralization ratios result in unsuitable gel formation and gel-sphere failure was mostly inevitable. For obtaining good quality sol with high thorium and cerium content, it is necessary to keep the limit of partial neutralization attainable without precipitation. In the experiments, for all solutions prepared at three different Th/Ce ratios, more than 85% partial neutralization was easily attainable at a neutralization rate at 0.3%·min⁻¹. At higher neutralization rates, the pH increase was too rapid, the pH values uncontrollably surpassed the setting value (pH=3.15) and unpeptizable precipitations occurred.

For sol drops formation, hexone saturated with ammonia was available to obtain suitable microspheres. NH₄OH solubility in hexone being above the level of 0.3 mol·L⁻¹, provides necessary pre-hardening to drops before complete gelation.

Synthesized (Th_0.50Ce_0.50)O₂, (Th_0.75Ce_0.25)O₂ and (Th_0.95Ce_0.05)O₂ microspheres were calcined at 450°C, 800°C and 1150°C for 2 h and XRD spectra were taken for each ratio at corresponding temperatures (Figures 4–6 ). For the each Th/Ce ratio it is clearly seen that, intensities of peaks increase and the peaks become sharper with the increasing calcination temperatures. This suggests that larger crystals occur by the aggregation of smaller crystals as the temperature increases.

Figure 4:

XRD spectra of (Th_0.50Ce_0.50)O₂.

Figure 5:

XRD spectra of (Th_0.75Ce_0.25)O₂.

Figure 6:

XRD spectra of (Th_0.95Ce_0.05)O₂.

Bragg angles θ_B, full width at half maximum (FWHM), distance between the layers (d), crystallite size (t) and unit cell size (a), are given in Tables 1–3, respectively, for three different ratios of mixed oxides calcined at previously indicated temperatures.

For (Th_0.50Ce_0.50)O₂, the increase in crystallite size with increasing temperature is clearly seen. The crystallite size being 124.8 Å for 450°C, it was found as 416.1 Å for 1150°C. Unit cell size at 1150°C was calculated as 5.4989 Å. For (Th_0.75Ce_0.25)O₂ the crystallite size being 103.9 Å for 450°C, it was found as 356.3 Å for 800°C and no increase was observed above this temperature. Unit cell size at 1150°C was calculated as 5.5740 Å. For (Th_0.95Ce_0.05)O₂, the crystallite size being 249.4 Å for 450°C, it was found as 415.6 Å for 1150°C. Unit cell size at 1150°C was calculated as 5.5913 Å.

It is clearly seen that unit cell sizes of the (Th_xCe_1−x)O₂ microspheres increase with the increasing thorium content. As it is well known, ionic radius of thorium and cerium are 0.98 Å and 0.92 Å, respectively [18] and that the increase in the unit cell size with thorium content is a result of difference in ionic radius.

Thermal analysis curves taken both in O₂ and N₂ atmosphere of each synthesized form (Th_0.50Ce_0.50)O₂, (Th_0.75Ce_0.25)O₂ and (Th_0.95Ce_0.05)O₂ are given in Figures 7–9 , respectively. The endothermic peak appeared between 1050 and 1100°C for all mixed oxide ratios in oxidative atmosphere, so that it can be attributed to initial step of sintering. In contrast to the oxidative atmosphere, no endothermic peaks were observed at above indicated temperatures. This shows that the sintering process is easier under oxidative atmosphere.

Figure 7:

TG/DTA graphs of (Th_0.50Ce_0.50)O₂ spheres (a) N₂ (b) O₂.

Figure 8:

TG/DTA graphs of (Th_0.75Ce_0.25)O₂ spheres (a) N₂ (b) O₂.

Figure 9:

TG/DTA graphs of (Th_0.95Ce_0.05)O₂ spheres (a) N₂ (b) O₂.

BET specific surface area and BJH porosity analysis results are given in Table 4. By the rise in temperature from 80°C to 1150°C, BJH pore diameter was increased. BJH pore volume and BET specific surface area values decreased.

As a result, in the present study crack-free (Th_xCe_1−x)O₂ microspheres at three different ratios (x=0.50, 0.75 and 0.95) with high homogeneity were prepared via sol-gel technique and were characterized. The use of sol-gel derived microspheres as press feed material possesses important advantages on the classical powder techniques. Their use as a press feed material will provide, undoubtedly, high flowability and low dust generation for following operational steps, resulting in minimal exposure of personal dose and simplification of supplementary precautions.