A Mechanism-Based Synergistic Stabilization Strategy for Room-Temperature Internal Gelation Process Toward Scalable HTGR Fuel Kernel Preparation

Abstract

High-temperature gas-cooled reactors (HTGRs) employ spherical fuel elements containing thousands of tristructural-isotropic (TRISO) particles, each centered on a UO₂ fuel kernel. The internal gelation process is a key technology for preparing these UO₂ fuel kernels. However, its application is limited by the poor room-temperature stability of conventional broths and the inherent trade-off between broth stability and mechanical strength. In this work, a novel five-component broth system composed of ZrO(NO₃)₂, hexamethylenetetramine (HMTA), urea, acetylacetone (ACAC), and glucose was developed. The synergistic effects of ACAC and glucose on broth stability and gelation kinetics were systematically investigated. An optimal ACAC/glucose molar ratio of 1:1 and an ACAC/ZrO₂⁺ ratio of 1.5 yielded a zirconium broth stable for over 5 h at 25 °C. Yttrium-stabilized zirconia (YSZ) microspheres prepared under optimized conditions exhibited excellent sphericity (1.04 ± 0.01), high density (5.84 g/cm³), and a crushing strength of 8.0 kg sphere⁻¹. Importantly, this stabilization strategy was successfully extended to the uranium broth, increasing its room-temperature stability from minutes to 6 h. The results demonstrate that the synergistic stabilization strategy effectively decouples the trade-off between broth stability and mechanical strength during the internal gelation process, providing an energy-efficient, scalable route for the preparation of nuclear fuel microspheres.

1. Introduction

HTGR technology has undergone several development phases over the past decades. Early demonstration projects in Germany, including the AVR and THTR reactors, established the technical feasibility of pebble-bed HTGR systems. Japan’s High-Temperature Engineering Test Reactor (HTTR) further validated high-temperature operation and safety performance, while the United States advanced TRISO fuel qualification through the Advanced Gas Reactor (AGR) program. Although HTGRs’ development experienced periods of reduced momentum in the early 2000s, global interest has resurged in recent years due to enhanced safety requirements, decarbonization targets, and the need for high-temperature industrial heat.

China has taken a leading role in this renewed development by successfully commissioning the HTR-PM demonstration reactor. The HTR-PM requires the annual production of several hundred thousand spherical fuel elements, each containing approximately 12,000 TRISO particles, each centered on a UO₂ fuel kernel. This unprecedented production scale imposes stringent requirements on the stability, reproducibility, and process control of UO₂ fuel kernel preparation.

The internal gelation process has emerged as a critical method for preparing spherical ceramic nuclear microspheres for HTGRs, particularly UO₂ fuel kernels and inert matrices [1,2]. The internal gelation process relies on the decomposition of gelation agents, such as hexamethylenetetramine (HMTA), which gradually increases the pH of the broth, inducing metal ion hydrolysis and polymerization to form gel microspheres quickly [3,4,5]. Despite its advantages in controlling particle size and sphericity, the internal gelation process is hindered by the poor room-temperature stability of the broth, primarily due to the high reactivity of metal ions and the urea–formaldehyde condensation reaction, which often leads to premature solidification during droplet formation [6,7].

Zirconia (ZrO₂) is commonly used as a surrogate material for uranium dioxide (UO₂) due to its similar ionic radius and analogous preparation chemistry [8]. This surrogate approach enables detailed mechanistic studies and process optimization without the constraints of radioactive materials. In the conventional nitric-acid–urea–HMTA system, adjusting the NO₃⁻/ZrO²⁺ or urea/ZrO²⁺ ratio can improve broth stability. However, it often compromises microsphere quality, leading to deformation, reduced mechanical strength, and increased susceptibility to cracking during washing [9]. Although the stable formulations for 15 h at room temperature have been reported for zirconium broth, the resulting ceramic microspheres exhibit insufficient strength [6,10]. Moreover, the instability issue is even more pronounced in uranium broths, which typically require strict cooling to near 0 °C for handling, posing significant challenges for industrial scale-up [11,12]. The stability–performance trade-off in both zirconium and uranium systems poses a key challenge.

To address this limitation, a novel five-component broth system incorporating acetylacetone (ACAC) and glucose as stabilizing additives has been developed. ACAC, a chelating agent, could complex with metal ions (ZrO²⁺ or UO₂²⁺), thereby reducing the concentration of free ions available for premature precipitation. Glucose modifies the gelation kinetics synergistically. This study first elucidates the individual and combined effects of ACAC and glucose on the stability of zirconium surrogate broth and the properties of the final sintered YSZ microspheres. Based on the mechanism obtained from the zirconium system, this strategy is successfully applied to the uranium system, achieving a significant extension of the uranium broth’s stability at room temperature. By optimizing the formulation, long-term room-temperature stability of the uranium broth is achieved without compromising the quality of the microspheres, providing a practical approach for industrial-scale production.

2. Materials and Methods

2.1. Materials

Zirconyl nitrate (ZrO(NO₃)₂, ≥99.0%, Shanghai Aladdin Biochemical Technology Co., Ltd., Shanghai, China), yttrium nitrate hexahydrate (Y(NO₃)₃·6H₂O, ≥99.0%, Shanghai Maclean Biochemical Technology Co., Ltd., Shanghai, China), hexamethylenetetramine (HMTA, ≥99.0%, Sinopharm Chemical Reagent Co., Ltd., Beijing, China), urea (≥99.5%, Sinopharm Chemical Reagent Co., Ltd., Beijing, China), acetylacetone (ACAC, ≥99.0%, Sinopharm Chemical Reagent Co., Ltd., Beijing, China), and D-glucose (≥99.0%, Sinopharm Chemical Reagent Co., Ltd., Beijing, China) were used. Silicone oil (50 cSt, Sinopharm Chemical Reagent Beijing Co., Ltd., Beijing, China) was used as the dispersion medium.

2.2. Preparation of Stable Zirconium Broth

The overall preparation process is shown in Figure 1. Firstly, a mixed precursor solution containing ZrO²⁺ (1.60 mol/L) and Y³⁺ (0.14 mol/L) was prepared. To 47.90 mL of the precursor solution, 3.83 mL of concentrated nitric acid was added under continuous stirring for 30 min to form a homogeneous Zr/Y solution at 600 rpm. Separately, a specified amount of glucose was dissolved in deionized water at 60 °C, followed by the addition of a specified amount of ACAC. Ultrasonic treatment was then applied to obtain a clear ACAC/glucose solution. The ACAC/glucose solution was mixed with the Zr/Y solution to form Solution A.

Meanwhile, a solution was prepared by mixing HMTA (3.00 mol/L) and urea (2.625 mol/L), hereafter referred to as HMUR solution. While maintaining an HMTA/ZrO²⁺ molar ratio of 2:1, the HMUR solution was slowly and uniformly added to Solution A at room temperature under vigorous stirring at 600 rpm to prevent localized precipitation. The final stable zirconium broths were yielded, and the corresponding compositions were summarized in

Table 1. Stability times of zirconium broths with different formulations.

Samples	Zr/Y Solution (mL)	n(HMTA /ZrO2+)	n(ACAC /ZrO2+)	n(C6H12O6/ZrO2+)	n(ACAC/C6H12O6)	Stability Time of Zirconium Broths (min)
1#	10	2	0	1	0	0
2#	10	2	0.5	1	0.5	30
3#	10	2	1	1	1	180
4#	10	2	1.5	1	1.5	180
5#	10	2	2	1	2	180
6#	10	2	1	0	-	30
7#	10	2	1	0.5	0.5	50
8#	10	2	1	1	1	180
9#	10	2	1	1.5	1.5	180
10#	10	2	1	2	2	180
11#	10	2	0.5	0.5	1	30
12#	10	2	0.75	0.75	1	90
13#	10	2	1	1	1	180
14#	10	2	1.25	1.25	1	220
15#	10	2	1.5	1.5	1	300
16#	10	2	1.75	1.75	1	320
17#	10	2	2	2	1	330

.

2.3. Preparation of Gel Microspheres

The broth was placed in a high-pressure syringe (SS-100 mL, Ditron Electronic Co., Ltd., Baoding City, China) with a 27 G needle and an inner diameter of 0.21 mm. The syringe was pushed by a high-pressure syringe pump (PHD ULTRA, Harvard pump, Harvard Apparatus, Holliston, MA, USA). The silicone oil was maintained at 90 °C. The distance between the nozzle and the hot silicone oil surface was 10 cm. The broth was dispersed at a flow rate of 3 mL/min, forming spherical sol droplets due to interfacial tension. The sol droplets were then dropped into a hot silicone oil column, where they solidified into gel microspheres within seconds. The gel microspheres were aged in an oil bath at 90 °C for 1 h, followed by an additional hour at room temperature. They were then successively washed with trichloroethylene (TCE, Shanghai Maclean Biochemical Technology Co., Ltd., Shanghai, China), 0.5 mol/L ammonia water, deionized water, and propylene glycol methyl ether (PGME, Shanghai Maclean Biochemical Technology Co., Ltd., Shanghai, China) to remove organic residues, nitrates, and other impurities [9]. Finally, the washed gel microspheres were dried at 60 °C for 12 h.

2.4. Thermal Treatment and Sintering

The pyrolysis profile of the dried gel microspheres was determined by thermogravimetric-differential scanning calorimetry (TG-DSC, NETZSCH STA 449 F3, NETZSCH Pumpen & Systeme GmbH, Waldkraiburg, Germany) under an air atmosphere. The analysis was conducted over the temperature range of 30 to 900 °C at a heating rate of 5 °C/min. Based on the results, a gradient sintering protocol was implemented. The dried gel microspheres were first heated to 200 °C, 320 °C, 470 °C, 600 °C, and 800 °C at 0.5 °C/min, with holding times of 1~5 h at each temperature to ensure complete organic removal. The dried gel microspheres were then subjected to final sintering at 1350 °C for 2 h in air at a rate of 2 °C/min to obtain crack-free YSZ ceramic microspheres.

2.5. Characterization

The viscosity of the broth was measured using a digital rotary viscometer, with stability time defined as the time to reach 20 mPa·s. The pH evolution was simultaneously monitored with a pH meter. FTIR (Nicolet Nexus 470, Thermo Fisher Scientific, Waltham, MA, USA) was used to analyze the functional groups in the broths and gels. The morphology and sphericity of the sintered microspheres were examined by optical microscopy, and their particle size was statistically analyzed from the corresponding micrographs. The bulk density and single-particle crushing strength of the final ceramic microspheres were determined via the Archimedes method and a particle strength tester, respectively.

3. Results and Discussion

3.1. Effect of ACAC and Glucose on the Stability of Zirconium Broth

To isolate the effect of ACAC content on broth stability, the concentrations of all other components (glucose, ZrO²⁺, HMTA, and urea) and the preparation conditions were kept constant. The zirconium broths were prepared with ACAC additions systematically varied from 0 to 3 g, corresponding to ACAC/ZrO²⁺ molar ratios of 0 to 2. The corresponding formulations (samples 1#~5#) are listed in Table 1, and their gelation phenomena are illustrated in Figure 2. As the ACAC content increased, the gel color changed from white to yellow, then to red.

Figure 3a shows the viscosity evolution of zirconium broths with varying ACAC contents at room temperature. Initially, all zirconium broths exhibited a period of low and stable viscosity, with fluctuations of less than 5%. This stable period was followed by a sharp increase in viscosity, marking the onset of gelation in the broth. The data clearly demonstrated that higher ACAC content delayed the gelation point, progressively extending the broth stability time to over 3 h. The trend confirmed the ACAC’s effective stabilizing role in zirconium broth. However, this effect saturated at an ACAC/glucose molar ratio of 1, beyond which additional ACAC had diminishing effects.

To investigate the role of glucose, its content was systematically varied from 0 to 5.4 g (glucose/ZrO²⁺ molar ratio = 0~2) in the zirconium broth. In comparison, ACAC addition was maintained at 1.5 g (ACAC/ZrO²⁺ molar ratio = 1). The concentrations of all other components (ZrO²⁺, HMTA, urea) and the preparation temperature (25 °C) were all held constant. The corresponding formulations are listed as samples 6#~10# in Table 1.

Figure 3b shows the viscosity curves for the zirconium broth with varying glucose content at a fixed ACAC concentration. A comparable trend to the ACAC variation was observed. The stabilization time of the zirconium broth increased progressively with increasing glucose content, extending beyond 3 h at the highest glucose content. This confirmed that glucose could also act as an effective stabilizer. However, the stabilizing effect plateaued at a glucose/ACAC molar ratio of 1, beyond which the stability time showed no significant increase. The gelation phenomena for samples 6# to 10# are presented in Figure 4. As glucose content increased, the gel color changed from yellow to red. The consistent observation of these saturation points in both ACAC and glucose experiments established the optimal ACAC/glucose molar ratio as 1:1.

Based on the established optimal 1:1 molar ratio of ACAC to glucose, the influence of their combined concentration on the stability of the zirconium broth was further investigated. The total content of the stabilizing additives was represented by the molar ratios of ACAC/ZrO²⁺ and C₆H₁₂O₆/ZrO²⁺, which were varied simultaneously while maintaining a fixed 1:1 ACAC:C₆H₁₂O₆ ratio. All other parameters, including the concentrations of ZrO²⁺, HMTA, and urea, as well as the preparation temperature of 25 °C, were kept constant. A series of zirconium broths was prepared with ACAC additions ranging from 0.75 g to 3.0 g and corresponding glucose additions from 1.35 g to 5.4 g. The detailed formulations for these zirconium broths are provided as samples 11#~17# in Table 1.

As illustrated in Figure 3c, the stabilization time of the zirconium broth was found to be intensely dependent on the ACAC/ZrO²⁺ ratio. A clear correlation was observed. The higher ACAC/ZrO²⁺ ratios led to progressively longer stabilization times for the zirconium broth. A distinct saturation behavior was observed after the ratio exceeded 1.5, characterized by a significant reduction in the rate of increase, ultimately leading to a maximum stability time of 5.5 h in the zirconium broth.

3.2. Stabilization Mechanism of the Zirconium Broth with ACAC and Glucose

Figure 5a shows the variation in pH values of 11#~17# zirconium broths with standing time at different ACAC/ZrO²⁺ molar ratios. The rate of pH change, as indicated by the slope of each curve, varied significantly throughout the process. Initially, all systems exhibited a rapid increase in pH, indicating rapid chemical reactions during the early stage. When the protonation degree of HMTA reached the critical threshold of 95%, marked reductions in the rate of pH increase were observed [6]. At this stage, the zirconium broth transitioned to a phase dominated by the decomposition of protonated HMTA, during which the reaction rate slowed considerably, and the zirconium broth gradually approached a stable gelation state. These two distinct stages were identified as the protonation process of HMTA and its subsequent decomposition, respectively.

Figure 5b presents the time and pH values corresponding to the gelation point of the zirconium broth, as determined by the viscosity transition, for different ACAC and glucose addition levels. In the conventional nitric–acid–urea–HMTA system, the gelation of ZrO²⁺ was known to occur at approximately pH 3.7. In contrast, Figure 5b showed that the minimum pH at which gelation in zirconium broths supplemented with ACAC and glucose was 3.92. Furthermore, the threshold pH increased progressively with increasing ACAC/ZrO²⁺ molar ratio. These results demonstrated that coordination of ACAC with ZrO²⁺ ions effectively reduced the concentration of ZrO²⁺ in the broth, thereby increasing the pH required for gelation.

As illustrated in Figure 5b, the stabilization time of the zirconium broth was found to be strongly correlated with the ACAC/ZrO²⁺ molar ratio. A distinct positive relationship was observed, wherein higher ACAC/ZrO²⁺ molar ratios resulted in progressively longer stabilization times. When the ACAC/ZrO²⁺ molar ratio was ≤1.5, the correlation was quantitatively described by Equation (1) with a confidence level of 99%, effectively capturing the dependence of zirconium broth stability on the ACAC and glucose content.

t = 268x − 104

(1)

where t denotes the stabilization time of zirconium broth (min), and x represents the molar ratio of ACAC/ZrO²⁺. Note that Equation (1) was an empirical correlation derived under the fixed experimental conditions of this study (ZrO²⁺, HMTA, and urea concentrations; ACAC/glucose molar ratio of 1:1; 25 °C) and may not be directly applicable to other systems without further verification.

Beyond a molar ratio of 1.5, a clear saturation behavior was observed, characterized by a significant reduction in the rate of increase, leading to a stabilization time approaching 5.5 h under optimal conditions.

To quantitatively deconvolute the individual and synergistic effects of ACAC and glucose on broth stability, a two-factor analysis of variance (ANOVA) was performed on the stability time [13]. The results summarized in

Table 2. Analysis of variance (ANOVA) for the effects of ACAC and glucose on the stability time of zirconium broths.

Source of Variation	Sum of Squares (SS)	Degrees of Freedom (df)	Mean Square (MS)	Contribution (%)
ACAC (A)	30,757	4	7689.25	46.00
Glucose (B)	21,973	4	5493.25	32.80
Interaction (A × B)	14,160	16	885.00	21.20
Total	66,890	24	14,067.50	100.00

indicate that both ACAC and glucose factors significantly influenced stability time. ACAC emerged as the dominant factor, accounting for 46.0% of the total variance in stability time, underscoring its primary role as a complexing agent in stabilizing ZrO²⁺ ions. Glucose also exerted a substantial effect, accounting for 32.8% of the variance, likely due to its role in modifying gelation kinetics and the broth environment. Notably, the interaction between ACAC and glucose accounted for 21.2% of the variance, statistically confirming a significant synergistic effect (p < 0.05). This synergy validated the observed optimal ACAC/glucose molar ratio of 1:1, in which the combined stabilizing effect exceeded the sum of their individual contributions.

When ACAC alone was utilized as the broth complexing agent, the formulated broth exhibited a limited room-temperature stability of approximately 30 min. Concurrently, the acidic broth would promote various side reactions involving ACAC, such as aldol condensation [14], nucleophilic addition–cyclization [15], and substitution [16,17] and oxidation [18], which consumed the complexing agent and introduced impurities into the zirconium system. Side reactions would accelerate HMTA decomposition and reduce the stability of the zirconium broth. This reactivity was primarily due to the elevated electron density at the α-carbon of the ACAC molecule under acidic conditions, which made it susceptible to electrophilic attack [19] and initiated deleterious side pathways. To mitigate these issues, glucose was introduced as a stabilizing agent. It was proposed that ACAC reacted with glucose, forming a D-glucose acetylacetone enol adduct with a stabilized α,β-unsaturated ketone structure featuring an extended π-conjugation system [19,20]. This reaction product could subsequently coordinate with ZrO²⁺ ions, forming a stable complex that effectively extended the broth stability time and improved the final product purity.

To provide direct experimental evidence for this proposed synergistic stabilization mechanism, FTIR spectroscopy was used to characterize chemical interactions in broths containing either ACAC alone or both ACAC and glucose, before and after gelation. The corresponding spectra and their interpretation were discussed below.

Figure 6 presents the FTIR spectra of different samples to elucidate the chemical interactions among ACAC, glucose, and zirconyl ions. The #1 broth contained ACAC as the broth stabilizing additive, while the #2 broth contained both ACAC and glucose. The corresponding gel samples are denoted as #1′ and #2′, respectively.

For broth #1, characteristic absorption bands at 3465 and 3450 cm⁻¹ correspond to the asymmetric and symmetric stretching vibrations of N-H bonds, which are typical of urea (CO(NH₂)₂), indicating that urea had not yet participated in the gelation reaction at this stage. The peaks observed at 2941 and 2869 cm⁻¹ were assigned to the asymmetric and symmetric stretching vibrations of C-N bonds in methylene groups, characteristic of hexamethylenetetramine (HMTA), confirming that HMTA remained largely undecomposed. In addition, a Zr-O vibration band appeared at approximately 523 cm⁻¹. After gelation (#1′), a broad absorption band appeared near 2960 cm⁻¹, which could be attributed to N–H stretching vibrations of amide groups. Meanwhile, the characteristic peaks of urea and HMTA disappear, indicating that both species participated in the gelation reaction to form urea–formaldehyde resin.

In broth #2, characteristic peaks of urea, HMTA, and Zr-O bonds were also observed before gelation, suggesting a similar initial chemical environment. Notably, although pure glucose exhibited a characteristic aldehyde C-H stretching vibration near 2720 cm⁻¹, this peak was absent in the spectrum of sample #5. This observation indicated that glucose had already reacted with ACAC. Furthermore, a prominent absorption band at 1612 cm⁻¹ was assigned to the C=O stretching vibration. Compared with broth #1, this band shifted to lower wavenumbers, suggesting the formation of an α,β-unsaturated ketone with a π-π conjugated structure. The conjugation effect weakened the C=O bond strength, leading to a red shift in the characteristic peak.

These results demonstrated that ACAC alone could effectively coordinate with ZrO²⁺ ions and extend the broth’s room-temperature stability. More importantly, the reaction between ACAC and glucose generated α,β-unsaturated ketone species, which further coordinated with ZrO²⁺ to form more stable chelated complexes. This synergistic coordination mechanism explained the significantly enhanced stability observed in the zirconium broth with ACAC and glucose.

Specifically, the proposed reaction between ACAC and glucose proceeded via a nucleophilic addition–elimination pathway, as shown in Figure 7. The hydrogen at the α-carbon of ACAC was activated by the strong electron-withdrawing effect of the two adjacent carbonyl groups, resulting in the formation of a carbanion (C⁻). This carbanion acted as a nucleophile, attacking the electrophilic carbonyl carbon of the open-chain form of glucose to form a new C–C bond. Concurrently, the C=O double bond of the aldehyde group opened, with electrons transferring to the oxygen atom, generating a negatively charged alkoxide intermediate. In the acidic medium, the oxyanion was protonated, yielding a neutral intermediate for aldol addition. Upon heating, this intermediate underwent dehydration. The newly formed hydroxyl group was protonated, converting it into a good leaving group (H₂O). At the same time, a proton was abstracted from the adjacent carbon atom. Simultaneous elimination of water led to the formation of a C=C double bond, yielding a D-glucose acetylacetone enol adduct with the structure of conjugated α,β-unsaturated ketone [21] as the final product. The conjugated α,β-unsaturated ketone could complex with ZrO²⁺ ions, thereby significantly contributing to the stability of the broth.

3.3. Comparative Study of YSZ Microsphere Properties

Thermogravimetric-differential scanning calorimetry (TG-DSC) was performed on the washed and dried gel microspheres, and the results are shown in Figure 8a. The dried gel microspheres consisted primarily of ZrO₂·2H₂O, acetylacetonate complexes, glucose, urea, HMTA, and related organic species. The thermal analysis revealed three characteristic exothermic peaks at 75 °C, 250 °C, and 480 °C, along with endothermic events at 185 °C and 360 °C. Specifically, the exothermic peak at 75 °C was attributed to the oxidation of glucose. The endothermic event at around 185 °C corresponded to the release of physically bound and crystalline water from hydrated zirconia, as described by Equation (2).

$$Zr{O}_2·2H_{2}O\to Zr{O}_2+2H_{2}O$$

(2)

A pronounced exothermic peak at approximately 250 °C was associated with the thermal decomposition and oxidation of urea-formaldehyde resin formed in situ during the internal gelation process. This resin decomposed into gaseous products, including CO₂, CO, NH₃, and H₂O, accompanied by significant heat release [9]. The endothermic feature near 360 °C was attributed to the breakdown of residual organic complexes and the rearrangement of acetylacetonate-coordinated species. Subsequently, the broad exothermic peak centered at approximately 480 °C corresponded to the oxidative combustion of acetylacetonate-derived organic residues and carbonaceous intermediates, leading to the formation of pure zirconia. All major mass loss and thermal events were completed below 500 °C, after which the TG curve reached a stable plateau, with a final residual weight of about 47% at 900 °C. This indicated the effective removal of organic components and the completion of oxide formation.

To mitigate the risk of microspheres cracking, a slow heating rate and staged holding steps were employed, as shown in Figure 8b. In particular, the heating rate in the 200~300 °C range was limited to 0.5 °C/min, allowing gradual decomposition of urea–formaldehyde resin and controlled gas release. This approach minimized internal pressure buildup and thermal gradients within the microspheres.

Figure 9 shows optical micrographs of the YSZ microspheres prepared from samples 11#~16#. It was observed that YSZ microspheres produced with an ACAC/ZrO²⁺ molar ratio below 1.5 exhibit smooth surfaces and maintain excellent sphericity, as shown in Figure 9a–e. When the ACAC/ZrO²⁺ ratio was 2, the sol droplets gelated too slowly in the hot silicone oil bath, settling to the bottom before solidification was complete, thereby preventing the formation of intact gel microspheres. These results indicated that while the complexing agent ACAC and the additive glucose could effectively enhance the room-temperature stability of the zirconium broth, excessive addition led to over-stabilization of the broth at the gelation temperature (90 °C). The overly stabilized state inhibited the uniform contraction and formation of gel microspheres during the sol–gel transition, thereby degrading the morphological quality of the final YSZ microspheres, as shown in Figure 9f. The effect of equal proportions of ACAC and glucose on the morphology of gel and YSZ microspheres is summarized in

Table 3. Effect of equal proportions of ACAC and glucose on the morphology of microspheres.

Samples	n(ACAC/ZrO2+)	Gel Microspheres	Sintered Microspheres
11#	0.50	good	good
12#	0.75	good	good
13#	1.00	good	good
14#	1.25	good	good
15#	1.50	good	good
16#	1.75	Deformed microspheres	Deformed microspheres
17#	2.00	No microspheres	No microspheres

.

Based on a systematic analysis of factors influencing room-temperature stability and formulation optimization, formulation 15# was selected for preparing the zirconium broth used to prepare the YSZ microspheres, with an ACAC/ZrO²⁺ molar ratio of 1.5 and an ACAC/glucose molar ratio of 1. The zirconium broth remained stable for more than 5 h at room temperature. This performance fully satisfied the industrial processing requirements of the internal gelation process. Furthermore, the resulting ceramic microspheres exhibited a smooth surface morphology, as shown in Figure 9e.

The YSZ microspheres synthesized from the 15# broth are characterized and compared with literature data, as summarized in

Table 4. The property comparison of YSZ microspheres.

Samples	Temperature of Preparation	Density (g/cm3)	Sphericity	Crushing Strength (kg)	Size (μm)
This work	25 °C	5.84	1.04 ± 0.01	8.0	508 ± 15
Ref. [6]	25 °C	5.87	1.02 ± 0.01	3.0	608 ± 6
Ref. [22]	5 °C	5.85	1.04 ± 0.04	8.1	345 ± 15

. The YSZ microspheres exhibited a sphericity of 1.04 ± 0.01, a density of 5.84 g/cm³, and a particle crushing strength of 8.0 kg per sphere, indicating excellent overall properties. Notably, the crushing strength of the YSZ microspheres was comparable to that of those prepared from low-temperature broths [6,22]. Furthermore, compared to existing room-temperature broth systems, the YSZ microspheres in this work exhibited superior mechanical strength. These results highlighted the unique advantage of the developed formulation. It combined the high strength typically associated with low-temperature processes with the prolonged stability and practical handling benefits of a room-temperature broth. Moreover, the fact that both preparation and storage of the zirconium broth could be conducted at room temperature represented a notable advance, simplifying operational requirements and reducing energy consumption.

3.4. Stabilization Mechanism Extended to the Uranium System

Having validated the efficacy and mechanism of the ACAC and glucose stabilizing system in the zirconia surrogate, the investigation was extended to the actual nuclear fuel precursor solution (a uranyl nitrate (ADUN) solution) to assess the practical relevance of this approach. A uranium broth was prepared by first introducing a specified amount of ACAC into a 2.8 mol/L ADUN solution, followed by the slow addition of an HMUR solution (containing 3 mol/L HMTA and urea). As shown in Figure 10, the uranium broth prepared in the absence of ACAC exhibited minimal stability, with a room-temperature lifetime of only 4 min. In contrast, a clear trend was observed whereby increasing the ACAC content progressively prolonged the stability time of the uranium broth.

To identify the key functional groups responsible for effective complexation, a series of complexing agents was added to a 2.8 mol/L ADUN solution. The HMUR solution was then added to each mixture to form the uranium broth. The room temperature stability time of each broth and its ability to form gel microspheres in hot silicone oil were evaluated. The results are summarized in

Table 5. Effects of complexing agents on the stability of uranium broth and on gel microsphere formation.

Complexing Agents	Stability Time (min)	Formation of Gel Microspheres
Urea	4	Good
Glucose	Precipitation	No formation
ACAC	60	Good
ACAC and glucose	360	Good
Citric acid	>360	No formation

.

When urea was used as the complexing agent, the prepared uranium broth rapidly gelled, indicating that the coordination between urea and uranyl ions was too weak to allow long-term storage at room temperature. In the case of glucose, it was found to react with nitric acid to form gluconic acid [23], which increased the broth pH and caused immediate precipitation [24,25]. When a combination of ACAC and glucose was used, the reaction between glucose and the hydrogen atom at the α-carbon of ACAC occurred, mitigating side reactions of ACAC with formaldehyde and ammonia decomposed by protonated HMTA. As a result, the stability time of the uranium broth at room temperature was extended from 1 to 6 h after the addition of an appropriate amount of glucose. Conversely, when citric acid was used as the complexing agent, the uranium broth failed to form gel microspheres in 70 °C silicone oil, demonstrating that citric acid complexation with uranyl ions was too strong to allow gelation.

The dual requirement of room-temperature stability and thermal gelation necessitated a complexing agent with precisely calibrated coordination strength. Excessively weak complexation, exemplified by the N-donor bonds from urea in a quadrangular bipyramidal configuration, provided insufficient stability for storage at room temperature. Excessively strong complexation, as demonstrated by citric acid, rendered the uranyl ions inert to thermally induced gelation. The efficacy of ACAC stemmed from its optimal O-donor coordination [26]. The binding strength of its carbonyl groups was sufficient to significantly reduce the free uranyl ion concentration, thereby achieving extended room-temperature stability of the uranium broth. Moreover, this coordination was also labile enough to be disrupted upon heating, allowing the controlled hydrolysis and polycondensation reactions required for gel microsphere formation to proceed. The uranium broth containing ACAC and glucose not only maintained excellent stability for 6 h at room temperature, the longest reported for uranium broth [12,27], but also showed excellent stability for 6 h at 37 °C. It also formed crack-free, high-strength gel microspheres, thereby establishing a robust foundation for the scalable production of high-quality UO₂ microspheres. Detailed work on the preparation of UO₂ microspheres using this strategy will be published in a separate publication.

Therefore, the results from the uranium system directly corroborated those from the zirconia surrogate. It was demonstrated that the chelation between dicarbonyl oxygen atoms of α,β-unsaturated ketone in D-glucose acetylacetone enol adduct and metal ions (ZrO²⁺ or UO₂²⁺) constituted the fundamental mechanism for achieving room-temperature stability while preserving the sol–gel transition capability upon heating. The parallel behavior observed across both zirconium and uranium systems underscored the applicability of the ACAC and glucose stabilization strategy in modulating the kinetics of the internal gelation process [28]. Specifically, ACAC served as the primary complexing agent [29]. At the same time, glucose acted as a synergistic stabilizer, mitigating the side reactions of ACAC and enhancing coordination efficiency. This work proposed a stabilization strategy compatible with the internal gelation process used for HTGRs fuel kernel production. It was not yet the commercial production formula, but it provided a scalable optimization direction.

4. Conclusions

In this work, a five-component internal gelation system comprising ZrO(NO₃)₂, HMTA, urea, acetylacetone (ACAC), and glucose was developed to address the long-standing challenge of limited room-temperature stability in the broths of the internal gelation process. Unlike conventional stabilization approaches that relied on single complexing agents or empirical parameter tuning, this study introduced a mechanism-based synergistic stabilization strategy. The key novelty lay in the cooperative role of ACAC and glucose. ACAC provided effective dicarbonyl coordination with metal ions. At the same time, glucose reacted with ACAC to form an α,β-unsaturated ketone with extended π-conjugation, thereby suppressing deleterious side reactions and enabling more stable metal-ligand complexation.

Through systematic optimization, optimal ACAC/glucose and ACAC/ZrO²⁺ molar ratios of 1:1 and 1.5, respectively, were identified, achieving room-temperature broth stability exceeding 5 h without compromising gelation kinetics. YSZ microspheres prepared from the optimized formulation exhibited excellent sphericity (1.04), high density (5.84 g/cm³), and strong mechanical integrity (8.0 kg per sphere), demonstrating that enhanced stability can be achieved without sacrificing final material performance.

Importantly, this stabilization mechanism was successfully extended to a uranium-based system, increasing the room-temperature stability of the uranium broth from only a few minutes to 6 h. The consistent behavior observed for both ZrO²⁺ and UO₂²⁺ systems highlighted the general applicability of the dicarbonyl-based coordination and conjugation strategy for modulating internal gelation kinetics. By effectively decoupling the stability–strength trade-off, this approach eliminated the need for energy-intensive low-temperature processing. It provided a practical pathway to scalable, energy-efficient production of nuclear fuel kernels.

It should be noted that the present study focused on elucidating the mechanism of internal gelation stabilization in room-temperature broths. Detailed cross-sectional microstructural characterization of UO₂ fuel kernels and full fuel performance qualification were not addressed due to radiological constraints and are beyond the scope of this work. These aspects will be investigated in future studies under appropriate facilities and conditions.