Highly robust Tremella polysaccharides stabilized selenium nanoparticles via avoiding polysaccharides self-aggregation Authors: Jianwei Dong, Jingyuan Liu, Huan Wang, Jiahao Zhang, Zhihai Wei, Da Song, Hao Han, Jingzhang Geng Published: 2026 ABSTRACT AbstractThis study employed the growth curve of selenium nanoparticles (SeNPs) to reveal the mechanism by which Tremella polysaccharides (TP) concentration affects SeNPs. At TP concentrations of 0.25–3.0 mg/mL, TP-stabilized SeNPs (TP-SeNPs) prepared with 1.5 mg/mL TP showed the smallest size (56.8 nm) and optimal stability, as this concentration sufficiently limited the diffusion of Se0 species and nuclei. The increased particle size (71.8 nm) at 3.0 mg/mL TP was due to TP self-aggregation, evidenced by marked rises in viscosity and hydrodynamic diameter, along with atomic force microscopy-detected aggregates. This aggregation was probably correlated with uronic acid content (17.1%) and the rigid conformation of TP. Thermogravimetry, zeta potential, and storage modulus further confirm that self-aggregation diminishes coating effectiveness. This work provided a theoretical basis for determining stabilizer concentrations in SeNP preparation. Moreover, TP-SeNPs maintained a stable, monodisperse spherical structure under simulated gastrointestinal digestion, showing great potential as a low-toxicity, high-bioavailability Se supplement.Keywords: Selenium nanoparticles, Tremella polysaccharides, Highly stability, Polysaccharides self-aggregation, Digestion Introduction The essential trace element selenium (Se) exhibits a narrow margin between its required and toxic concentrations, so the lower toxicity SeNPs have become the most promising source of Se (Gao et al., 2023; Wei et al., 2026). However, the bioavailability of bare SeNPs is severely limited by their tendency to aggregate in aqueous suspensions and physiological intake environments (Yang et al., 2025). Stabilizers were required to maintain their colloidal state, and polysaccharides (PS) were often used as stabilizers for SeNPs (Qi et al., 2025; Subhash et al., 2025), which improve SeNPs' bioactivity and stability. PS stabilize SeNPs through non-covalent bonding by their high surface area and abundance of hydroxyl groups. This lowers the surface free energy of SeNPs and prevents them from aggregating. Molecular weight (Song et al., 2020a), conformation (He et al., 2024), and functional groups (Xie et al., 2024) have been the primary topics of research on PS's stabilizing effect on SeNPs. However, due to the structural complexity of PS, it is difficult to reach a unified understanding. In our previous study, by investigating the growth pattern of SeNPs, we found that limiting the diffusion of Se0 species and Se nuclei may be the underlying mechanism by which PS stabilize SeNPs, and this requires sufficiently strong interactions with SeNPs (Dong et al., 2026). Of course, the concentration of PS will also affect the formation of SeNPs. Many studies have shown that the size of SeNPs gradually decreases due to rising stabilizer concentration or ratio, and after reaching a minimum particle size, further increasing the stabilizer concentration causes the size of SeNPs to grow (Chen et al., 2023; Cui et al., 2019). A few studies have also shown that SeNPs' size continuously decreases as stabilizer concentration rises (Zhang et al., 2025a). There is widespread agreement that a low stabilizer concentration is insufficient to disperse and prevent the collision and aggregation of SeNPs, resulting in an increase in particle size (Chen et al., 2023). However, several explanations have been proposed for the increase in SeNPs size at high stabilizer concentrations. One attributes it to an excessively thick stabilizer layer coating the SeNP surface (Chen et al., 2023), thereby increasing particle size. Another suggests that molecular bridging leads to size enlargement (Zhang et al., 2015). A third explanation is that excess stabilizer polymerizes on the SeNPs (Ye et al., 2020a). Regrettably, no direct evidence for these explanations has been found, and there are still many studies that do not explain the reasons. Therefore, the effect of PS concentration on the formation of SeNPs needs to be systematically studied. With increasing polysaccharide concentration, intermolecular self-aggregation occurs via hydrogen bonds, hydrophobic interactions, ionic interactions, and van der Waals forces, leading to increased viscosity and eventually the formation of a strong gel network (Xia et al., 2026; Ye et al., 2020c). While no studies have so far been identified that address the influence of concentration-induced polysaccharide self-aggregation on the morphology of nanomaterials, investigations that employ polysaccharide triple-helical self-assemblies or aggregates as stabilizers for nanomaterials may serve as a useful reference. For instance, the triple-helical conformations of β-glucan (Jia et al., 2015) and Curdlan (Xu et al., 2023) were found to occlude the binding sites that enable strong interactions with SeNPs, thereby compromising their stabilizing capability and yielding SeNPs with poor stability. In contrast, when the dissociated single chains of these polysaccharides were used as stabilizers, highly stable SeNPs were obtained in both cases. Furthermore, although stable nanoparticles could be successfully prepared using the self-assembled aggregates of macromolecules in tea infusion (Ye et al., 2020a) and the self-assemblies of diphenylalanine induced by alginate dialdehyde (Shen et al., 2019) as template molecules, transmission electron microscopy revealed that these nanoparticles were adhered together by the coating layers of the template molecules. Therefore, it can be inferred that self-aggregation of polysaccharides may shield the active sites responsible for interaction with SeNPs and reduce their spatial accessibility (Chang et al., 2026), thereby weakening their ability to induce the nucleation and growth of SeNPs, ultimately leading to increased particle size and decreased stability. On the other hand, self-aggregation of polysaccharides may also produce a bridging effect, causing agglomeration of the nanoparticles. Hence, avoiding concentration-induced stabilizer self-aggregation may be beneficial for the preparation of highly stable SeNPs. In this study, TP was used as a stabilizer to investigate the effects of its concentration on the growth of SeNPs, and to characterize how TP concentration affected its own hydrodynamic properties, aiming to better understand the mechanism by which polysaccharide concentration regulates SeNP formation. Elucidating this mechanism will provide theoretical support for the determination of stabilizer concentration. We hypothesize that the stabilizer concentration may affect the nucleation and growth of SeNPs by altering its hydrodynamic characteristics, leading to the formation of SeNPs with different morphologies, and these SeNPs may also exhibit differences in stability. In addition, as a prebiotic (Sun et al., 2025), TP will play a critical role in protecting SeNPs during gastrointestinal digestion, which a key factor in improving their bioavailability (He et al., 2025), and this aspect has been specifically evaluated herein. Materials and methods Materials Tremella polysaccharide (TP), sodium selenite (Na2SeO3), and ascorbic acid (AA) were obtained from Aladdin (Shanghai, China). Artificial saliva (ISO/TR10271) was purchased from Shanghai Yuanye Biotechnology Co., Ltd. (Shanghai, China). Trypsin (from porcine pancreas, 1:250) and pepsin (from porcine gastric mucosa, 1:3000) and were purchased from Beyotime Biotechnology Co., Ltd. (Shanghai, China). The water used in the experiment was deionized water, and all of the chemicals were analytical grade. Synthesis of TP-stabilized SeNPs (TP-SeNPs) After dissolving a certain quantity of TP in 20 mL of deionized water (the final concentration of the reaction solution was in the range of 0.25–3.0 mg/mL), 8.64 mg of Na2SeO3 was added and rapidly mix at room temperature for 0.5 h. Add 5 mL of 40 mM AA solution to the solution and stir continuously at 25 °C for 0.5 h. To get rid of extra reactants, the TP-SeNPs solution was dialyzed for two days in a dialysis bag with a molecular weight cutoff of 3500 Da (Aladdin, Shanghai, China). It was then kept at 4 °C for further use. With the exception of not using TP, the preparation method of bare SeNPs was the same as that described above. The physicochemical characteristics of TP Chemical composition analysis The total sugar, protein, and uronic acid contents of TP were determined by the phenol‑sulfuric acid method, the coomassie brilliant blue method assay, and the meta-hydroxydiphenyl assay, respectively (Chen et al., 2025). Conformational characterization The multiple parameters of TP, including weight average molecular weights (Mw), intrinsic viscosity ([η]), radius of gyration (Rg) and hydrodynamic radius (Rh) were directly derived from a GPC-MALLS-VS-RI system (Xu et al., 2020), which consists of a Waters 1515 gel permeation chromatography (GPC, Waters Corp., USA) attached to a DAWN Heleos multi-angle laser light scattering equipped with 658 nm laser (MALLS, Wyatt Technology Co., USA), a Viscostar-II viscometer (VS, Wyatt Technology Co., USA) and an Optilab T-rEX refractive index detector (RI, Wyatt Technology Co., USA). 2.5 mg/mL TP solution was filtered through a 0.45 μm syringe membrane, and the injection volume was 50 μL. The eluting procedure was performed at 0.6 mL/min with in OHpak SB-806 M and SB-804 columns (Shodex, Japan) in series, which were maintained at 40 °C, and the elution buffer was 0.1 M NaNO3. The refractive index increment (dn/dc) was set to be 0.138 mL/g, and data acquisition and processing were performed using ASTRA 6.1.4 software (Wyatt Technology Co., USA). Morphology observation Molecular chain morphologies of TP dissolved in water were observed on a SPM9700 atom force microscopy (AFM, Shimadzu Corp., Japan) with a commercial NCHR tips. 5 μL of TP aqueous solutions with varying concentrations (0.25, 1.5 and 3.0 mg/mL) was dropped onto the freshly cleaved mica substrate, respectively. After air drying at room temperature, AFM measurements were performed using the tapping mode under ambient conditions. Characterization of TP-SeNPs Following suitable dilution and ultrasonic dispersion, a drop of the treated TP-SeNPs was suspended on a 200-mesh copper net, naturally dried, and viewed using a JEM-F200 transmission electron microscopy (TEM, JEOL, Japan) with an accelerating voltage of 200 kV. UV–vis absorption spectra (190–780 nm) of the 5-fold diluted TP-SeNPs were acquired with a UV–Visible spectrophotometer (Thermo Evolution 201, USA). A DelsaMax PRO dynamic light scattering (DLS, Beckman, USA) was utilized to assess the particle size and zeta potential of TP-SeNPs in the undiluted TP-SeNPs solution. The instrument was operated at a laser wavelength of 532 nm, a detection angle of 163.5°, and a refractive index of 1.333. Every measurement was carried out in triplicate at 25 °C. Lyophilized SeNPs were ground and tableted with dry KBr for recording the Fourier transform infrared spectroscopy (FT-IR) on a Nicolet iS10 instrument (Thermo Scientific, USA) in the spectrum range of 400–4000 cm−1 with a resolution of 0.5 cm−1. The X-ray Photoelectron Spectroscopy (XPS) of lyophilized TP-SeNPs were recorded with a ESCALab 250Xi instrument (Thermo Scientific, USA) equipped with a monochromatic Al Kα X-ray source (1486.6 eV). The thermogravimetric analysis (TGA) and differential thermal analysis (DTA) of the lyophilized SeNPs were carried out on a TGA/SDTA851 thermogravimeter (Mettler Toledo, Switzerland) with a heating rate of 10 °C/min in a flowing high-purity nitrogen gas in the range of 30 to 800 °C. Storage modulus (G') of undiluted TP-SeNPs was measured at 25 °C on a HAAKE Mars 40 rheometer (Thermo Scientific, USA) in oscillation mode (1 Hz, 0.01–100% strain). The Se content in the simulated digestive fluid was determined using an AES-230E hydride generation-atomic fluorescence spectrometer (HG-AFS, Beijing Haiguang, China). Briefly, 1 mL of the fluid was mixed with 5 mL of HNO₃, microwave-digested, evaporated to near dryness, and then treated with 5 mL of HCl (6 mol/L) and 2.5 mL of K₃Fe(CN)₆ (100 g/L). The mixture was diluted to 10 mL with water and used for analysis. Stability of TP-SeNPs in various environments The storage stability of SeNPs under different conditions of temperature, pH, alcohol concentration, and ionic strength was evaluated. Section 1 of the Supplementary Material contained specific instructions. Growth curve of TP-SeNPs During TP-SeNP preparation, 1 mL aliquots were taken every 5 min, immediately measured for particle size by DLS (Anderson et al., 2014; Zvereva et al., 2024), and the size-vs-time data were plotted as the growth curve. The average nanoparticle radius and reaction time are related by Lifshitz-Slyozov-Wagner (LSW) growth theory, as demonstrated by Eq. (1). (Lee et al., 2024; Lifshitz et al., 1961; Weiner et al., 2016) where, t is the reaction time; Rt is the average radius of SeNPs at time t; K is the proportionality coefficient between Rt and t; n is the growth index of SeNPs. Determination of Se release rate of TP-SeNPs during in vitro digestion To investigate the potential bioactivities of TP-SeNPs, their digestion characteristics under simulated gastrointestinal conditions were examined. Section 2 of the Supplementary Material contained specific instructions. Statistical analyses Every experiment was conducted at least three times, and the mean ± SD (standard deviation) was used to express the results. Student's t-test and analysis of variance (ANOVA) were performed using OriginPro Software Version 8.5 (OriginLab Corp., MA, USA) for statistical analysis, and P < 0.05 was considered statistically significant. Results and discussion Effect of stabilizers concentration on the morphology of SeNPs As shown in Fig. 1A, TP-SeNPs prepared with 0.25 mg/mL TP exhibited poor monodispersity and relatively large particle sizes. As the TP concentration gradually increased, the dispersibility of the obtained TP-SeNPs improved progressively, and the particle size decreased accordingly. The smallest particle size and optimal monodispersity were achieved with 1.5 mg/mL TP (Fig. 1B, C, D). However, with a further increase in TP concentration to 3.0 mg/mL, the prepared TP-SeNPs began to aggregate again, accompanied by an increase in particle size (Fig. 1E, F, G). The particle sizes of all TP-SeNPs in the TEM images were measured using Nano Measurer 1.2 software, and the particle size distributions of TP-SeNPs were obtained through Gaussian fitting, as shown in Fig. S1 in the Supplementary materials and Fig. 1H. The greatest peak particle size (126 nm) and the widest particle size distribution were shown by TP-SeNPs0.25. As the TP concentration increased to 1.5 mg/mL, the peak particle size of the prepared TP-SeNPs gradually decreased to 56.8 nm, and the particle size distribution reached its narrowest range. With further increases in TP concentration, the particle size of the obtained TP-SeNPs began to increase again, and the particle size distribution gradually widened. These results were consistent with UV–Vis spectroscopy (Fig. S2). As TP concentration increased from 0.25 to 1.5 mg/mL, the maximum absorption peak of SeNPs blue-shifted from 268 nm to 255 nm, indicating a gradual decrease in particle size (Yang et al., 2023). However, when TP concentration further increased from 1.5 to 3.0 mg/mL, the particle size gradually increased as the peak red-shifted from 255 to 260 nm. Morphology of TP-stabilized SeNPs (TP-SeNPs) prepared with different concentrations of TP as determined by Transmission electron microscopy (TEM). A, B, C, D, E, F, and G correspond to TP concentrations of 0.25, 0.5, 1.0, 1.5, 2.0, 2.5, and 3.0 mg/mL, respectively. H shows the particle size distribution of the seven types of TP-SeNPs described above. Although an optimal stabilizer concentration exists for minimizing both the particle size and size distribution of SeNPs (Cui et al., 2019), the underlying mechanisms remain underexplored. It is widely accepted that stabilizers reduce the surface energy of SeNPs, thereby preventing aggregation and leading to smaller particle sizes and narrower distributions (Dong et al., 2026). However, no consensus has been reached on whether an excess of stabilizer causes an increase in particle size. One viewpoint holds that excess stabilizer thickens the adsorbed layer on SeNPs, thereby increasing their hydrodynamic size (Chen et al., 2023; Ye et al., 2020a; Zhang et al., 2025a). Another viewpoint attributes the aggregation of SeNPs to the bridging effect of excess stabilizer (Qi et al., 2025; Zhang et al., 2015), but conclusive evidence remains lacking. Liu et al. (2026) found that increasing the TP concentration from 2 to 8 mg/mL gradually reduced the SeNPs-TP's size from 245 nm to 184 nm. However, because TP at 8 mg/mL was already markedly viscous at room temperature, further increasing its concentration to obtain smaller SeNPs was impractical. We speculate that high-concentration-induced changes in the hydrodynamic properties of TP may affect SeNP growth, ultimately influencing their particle size and monodisperse state. Indeed, as shown in Fig. S3, further increasing the TP concentration to 4, 5, and 6 mg/mL led to markedly greater aggregation of the obtained SeNPs. Impact of stabilizers concentration on SeNP stability Temperature High temperatures can hasten the aggregation and loss of biological activity of nanomaterials with high free energy and surface energy (He et al., 2024; Zhai et al., 2024). As shown in Fig. 2A, the freshly prepared TP-SeNPs0.25 (0.25 mg/mL TP) solution was turbid, whereas TP-SeNPs1.5 (1.5 mg/mL TP) was clear and transparent. Since colloidal clarity is negatively correlated with particle size (Xie et al., 2024), this indicated that TP-SeNPs0.25 had the largest particle size and TP-SeNPs1.5 the smallest. At 4 °C, TP-SeNPs0.25 cleared after 25 days due to sedimentation of large SeNPs, while others remained unchanged. At 46 days, all TP-SeNPs showed slight clarity loss from mild aggregation. At 25 °C, after 25 days, TP-SeNPs0.25 almost fully precipitated; TP-SeNPs0.5 and TP-SeNPs3.0 also noticeably precipitated, while others slightly decreased in clarity. After 46 days, TP-SeNPs0.25 fully precipitated, TP-SeNPs0.5 and TP-SeNPs3.0 showed significant precipitation, and the remaining TP-SeNPs further decreased in clarity. Compared to the 4 °C environment, the 25 °C accelerated the aggregation and precipitation of TP-SeNPs (Zhai et al., 2024). However, TP-SeNPs1.5 maintained relatively good stability under both temperature conditions, benefiting from the stabilizing effect of an appropriate concentration of TP. The variations in particle size, polydispersity index (PDI), and zeta potential of several TP-SeNPs during storage at different temperatures were shown in Fig. 2B, C, and D. Notably, TP-SeNPs0.25 showed the greatest variations across the three indicators. The particle size increased by 107% at 4 °C (from 78.8 ± 3.1 to 163.9 ± 12.6 nm) and by 668% at 25 °C (from 84.5 ± 3.6 to 649.2 ± 38.5 nm) over 36 days. Meanwhile, the PDI increased from 0.2 ± 0.06 to 1.18 ± 0.05 at 4 °C and from 0.22 ± 0.05 to 3.07 ± 0.30 at 25 °C, while the zeta potential increased from −13.0 ± 0.44 mV to −4.8 ± 0.64 mV and from −12.8 ± 0.38 mV to −2.3 ± 0.40 mV at the respective temperatures, suggesting the occurrence of TP-SeNP aggregation. The insufficient amount of stabilizer was the primary cause of the significant aggregation under both temperatures. Additionally, the aggregation of SeNPs was accelerated by the higher temperatures, which was in line with the experimental observations reported by Chu et al. (2024). The increases in particle size for TP-SeNPs 1.5 and TP-SeNPs 3.0 under 4 °C and 25 °C storage were 36% and 58%, 58% and 127%, respectively. Furthermore, the PDI and zeta potential of the two TP-SeNPs showed only a slow increase over the storage period. In particular, after 36 days of storage, the PDI of TP-SeNPs increased to 0.23 ± 0.07 (4 °C) and 0.40 ± 0.06 (25 °C), and the zeta potential increased to −15.8 ± 0.49 mV (4 °C) and − 13.8 ± 0.43 mV (25 °C). Among the three TP-SeNPs, TP-SeNPs₁.₅ displayed the smallest variations in all three measured parameters and the highest resistance to high temperature, which was consistent with the visual observations in Fig. 2A. In summary, the storage stability of SeNPs was influenced by the TP concentration. Under the Na2SeO3 concentration of 2 mM, 1.5 mg/mL TP exerted the optimal stabilizing effect. It is possible that 1.5 mg/mL TP effectively encapsulated the formed NPs, minimizing the surface energy of the SeNPs and preventing their aggregation. In contrast, 0.25 mg/mL TP was insufficient, while 3.0 mg/mL TP was excessive and produced a counterproductive effect. Photographs(A), particle size and polydispersity index (PDI) variations at 4 °C (B), 25 °C (C), and zeta potential (D) evolution of TP-SeNPs during storage at different temperatures. 1, 2, 3, 4, 5, 6, 7 in (A) represent TP concentrations of 0.25, 0.5, 1, 1.5, 2, 2.5, and 3 mg/mL, respectively. TP-SeNPs0.25, TP-SeNPs1.5 and TP-SeNPs3.0 in (B), (C) and (D) represent TP-SeNPs prepared with 0.25, 1.5, and 3.0 mg/mL of TP, respectively. pH, alcohol content and ionic strength To preserve their bioactivity and bioavailability, SeNPs must remain monodisperse in application environments (Lim et al., 2024; Tian et al., 2026). Their suitable liquid food carriers (e.g., tea, juice, milk, low-alcohol beverages) possess specific pH, alcohol content, and ionic strength. Moreover, during gastrointestinal transit, SeNPs encounter varying ionic strengths and pH values (Xie et al., 2025). So, this work examined the stability of SeNPs (30 mg/L) under different pH (2–7), ethanol (5–21% vol), and NaCl (50–400 mM) conditions. Samples were stored in the dark at 25 ± 1 °C and periodically photographed to track colloidal changes. After 1 day across pH conditions, all TP-SeNPs were translucent light brown except TP-SeNPs0.25, which showed a slight reddish hue (Fig. 3A), confirming that all but TP-SeNPs0.25 remained stable. After 40 days, TP-SeNPs0.25 turned markedly red, and all except TP-SeNPs1.5 became reddish and faded. These color changes were indicative of the aggregation of SeNPs (Chopada et al., 2026), and TP-SeNPs1.5, which exhibited the weakest change, demonstrated superior tolerance to pH. After 1 day in various alcohol concentrations, all TP-SeNPs appeared translucent light-brown color (Fig. 3B), indicating that they are insensitive to alcohol over short periods of time. After 40 days, TP-SeNPs0.25 turned reddish, while TP-SeNPs2.5 and TP-SeNPs3.0 became significantly lighter, with higher alcohol concentration leading to lighter color. The color changes and fading reflected TP-SeNPs aggregation, suggesting that TP-SeNPs1.0, TP-SeNPs1.5, and TP-SeNPs2.0 possess good alcohol tolerance. After one day under various NaCl concentrations, all TP-SeNPs were semi-transparent light brown, comparable to the NaCl-free control (Fig. 3C), indicating short-term ionic strength insensitivity. After 40 days, only TP-SeNPs0.25 turned slightly reddish, with no notable changes in the others, suggesting that ionic strength does not affect TP-SeNP storage stability. Stability of TP-SeNPs under acidic (A), alcoholic (B) and ionic strength conditions (C). 0.25, 0.5, 1, 1.5, 2, 2.5, and 3.0 represent the concentration of TP (mg/mL) used in the prepared TP-SeNPs, respectively. It is generally accepted that strong acidic conditions can disrupt the hydrogen bonding and electrostatic interactions between PS and SeNPs, causing collisions and aggregation of SeNPs (Song et al., 2020b). The strongest contact force with SeNPs may have been shown by 1.5 mg/mL TP, thus demonstrating the best tolerance to pH. The two main mechanisms conferring colloidal stability are electrostatic repulsion (Liu et al., 2024) and steric hindrance repulsion (Tao et al., 2025). Strong acidity and high ionic strength can shield the surface charges of NPs and reduce electrostatic repulsion, thereby promoting aggregation. As a result, electrostatic stabilization's efficacy in complex matrices is limited by its sensitivity to pH and ionic strength. Steric stabilization, on the other hand, is a preferred approach in complicated settings because it is insensitive to electrolytes (Lim et al., 2024). The type of surface coating determines the colloidal stability of SeNPs in various environmentally relevant aqueous media. The steric hindrance that TP provides may be connected to TP-SeNPs' resistance to pH and ionic strength (Dong et al., 2026). 1.5 mg/mL TP was likely to provide the strongest steric hindrance for TP-SeNPs, thereby endowing them with optimal tolerance to pH and ionic strength. Nevertheless, it may be worth noting that direct evidence of steric hindrance effects could potentially provide stronger support for this conjecture. By breaking hydrogen bonds with water and strengthening hydrophobic interactions between polysaccharide chains, ethanol encourages polysaccharide aggregation in aqueous solutions. However, for polysaccharides to be severely impacted, an ethanol concentration of more than 30% is needed (Hu et al., 2018). Therefore, low-alcohol beverages have minimal impact on polysaccharides. However, as the concentration of polysaccharides in the solution increases, the effect of ethanol-induced polysaccharide aggregation may be enhanced. Dialysis with a 3500 Da membrane did not remove TP from TP-SeNPs2.5 and TP-SeNPs3.0 solutions, leading to elevated polysaccharide levels. This promoted aggregation and bridging even under low-alcohol conditions, decreasing their stability. Overall, TP-SeNPs1.5 exhibited good stability in environments with low pH, low alcohol concentration, and high ionic strength, which supports their prospective use as a dietary supplement or nanomedicine to treat Se deficiency (He et al., 2025). Influence of stabilizer concentration on the growth model of SeNPs A classical model for nanocrystal growth kinetics, the LSW hypothesis is equally applicable to the growth of organic-stabilized inorganic nanoparticles (Canepa et al., 2018; Zvereva et al., 2024). When the relation of nanocrystal radius (R) and time (t) is R ∝ t1/2, the growth of nanocrystals is controlled by the internal reaction kinetics, also known as reaction-limited growth. On the other hand, growth is dominated by precursor diffusion when the equation is R ∝ t1/3, referred to as diffusion-limited growth (Song et al., 2018; Woehl et al., 2012). According to the growth curve in Fig. 5, the radius growth of TP-SeNPs0.25, TP-SeNPs1.5 and TP-SeNPs3.0 followed R ∝ t0.55, t0.24, and t0.26, respectively. The growth curve of TP-SeNPs0.25 was closely approximate t1/2, corresponding to reaction-limited growth, while those of TP-SeNPs1.5 and TP-SeNPs3.0 were approximate t1/3, indicating a diffusion-limited growth (Lee et al., 2024; Weiner et al., 2016). In the absence of stabilizers to limit diffusion, Se0 species could easily diffuse to the crystal surface, resulting in the growth rate of SeNPs depending on their own reaction kinetics. This resulted in a reaction-limited growth mode, leading to relatively larger particle sizes. Additionally, the absence of stabilizers that restrict the collision of Se nuclei could also lead to agglomeration. A diffusion-limited growth mode was produced when stabilizers were present because they prevent Se0 species from diffusing to the crystal surface. In addition to slowing the formation of SeNPs, this growth mode stopped Se nuclei from colliding and clumping together, which decreased the size of NPs (Dong et al., 2026). TP-SeNPs₀.₂₅ showed reaction-limited growth and weak diffusion restriction, explaining their large particles and interconnected structures. This aroused from too few TP molecules, which could not prevent Se0 species diffusion to Se nuclei, block Se nuclei collision, or fully shield SeNP surface free energy, leading to aggregation. Conversely, TP-SeNPs₁.₅ and TP-SeNPs₃.₀ exhibited a diffusion-limited growth pattern, indicating that when the concentration of TP reaches 1.5 mg/mL, it is sufficient to completely hinder the diffusion and collision of Se0 species and Se nuclei. This result was consistent with the findings of Canepa et al. (2018), who reported that at a low cetyltrimethylammonium bromide concentration (7.5 mM), the growth of gold nanocrystals followed a reaction-limited mechanism, whereas at a higher concentration (50 mM), a diffusion-limited regime was observed. As previously mentioned, compared to other TP-SeNPs, TP-SeNPs₁.₅ exhibits the smallest particle size and the best monodispersity (Fig. 1), indicating that 1.5 mg/mL of TP is precisely the concentration required to achieve control over the particle size of SeNPs without being excessive, making it the optimal concentration for TP to stabilize SeNPs. When TP was in excess, the surplus polysaccharides may undergo self-aggregation and bridging (Qi et al., 2025; Zhang et al., 2015), leading to the agglomeration of the initial Se nuclei. The relatively large initial particle size of TP-SeNPs₃.₀ in the growth curve confirms this hypothesis. This phenomenon may also be attributed to the self-aggregation of TP itself, which aggregates Se ions during the mixing process of TP and Na2SeO3. TG (A) and DTG (B) curves, FT-IR spectra (C) and XPS spectra (D) of TP-SeNPs. Bare SeNPs (B-SeNPs), TP-SeNPs0.25, TP-SeNPs1.5 and TP-SeNPs3.0 represent TP-SeNPs prepared with 0, 0.25, 1.5, and 3.0 mg/mL of TP, respectively. Effect of stabilizers concentration on its interaction with Se The diffusion limitation of polysaccharides on Se0 species and Se nuclei requires the support of interfacial interaction forces between them. The interaction between TP and Se was examined using thermogravimetric analysis (TGA), fourier transform infrared spectroscopy (FT-IR), and X-ray photoelectron spectroscopy (XPS). The TG curve of B-SeNPs displayed one main thermal decomposition stage, starting from approximately 300 °C and ending at around 530 °C, with a mass loss of about 97% (Fig. 5A). This should be attributed to the decomposition of Se, the main component of B-SeNPs. The DTG curve of B-SeNPs exhibited a maximum degradation rate temperature (Tmax) of 514 °C (Fig. 5B), which was near the previously mentioned Tmax of SeNPs at approximately 409 °C (Chen et al., 2015), confirming that the weight loss in this stage was mainly caused by the decomposition of Se. The thermal decomposition curves of TP-SeNPs were significantly different from that of B-SeNPs. The significant decrease in Tmax indicated the polysaccharides in TP-SeNPs have substantially altered the decomposition properties of SeNPs (Fig. 5B), given that PS' Tmax was roughly 310 °C (Chen et al., 2019), which was lower than that of SeNPs. Two stages mainly constituted the thermal decomposition of the three types of TP-SeNPs (Fig. 5A). The first stage ranged from 30 °C to 230 °C, and the second stage ranged from 230 °C to 500 °C. The weight losses of TP-SeNPs₀.₂₅, TP-SeNPs₁.₅, and TP-SeNPs₃.₀ in these two stages were 13.7% and 42.9%, 20.6% and 38.8%, 17.6% and 39.6%, respectively. The total weight losses for the three in the two stages were 56.6%, 59.4%, and 57.2%. Therefore, the greatest weight loss exhibited in TP-SeNPs₁.₅, indicating that TP-SeNPs₁.₅ contained the highest amount of polysaccharides. As shown in Fig. 5C, the absorption peaks of TP at 3363 cm−1, 2926 cm−1 and 1031 cm−1 correspond to the stretching vibrations of O—H, C—H and C-O-C groups, respectively (Wang et al., 2025). All TP-SeNPs exhibited characteristic peaks at similar wavenumbers to those of TP, indicating that TP was successfully coated onto the SeNPs (Chen et al., 2023). Compared to TP, the infrared spectrum of TP-SeNPs showed no new absorption peak, suggesting that there was no new covalent link between TP and SeNPs (Zhai et al., 2024). Additionally, the TP-SeNPs had clear peak shifts at wavenumbers that correspond to the O—H, C—H, and C-O-C groups of TP, suggesting that TP interacts with SeNPs through these three functional groups (Qi et al., 2025). Compared to TP, the three TP-SeNPs (0.25, 1.5, 3.0) displayed blue shifts (69, 61, 63 cm−1) in O—H stretching, red shifts (6, 5, 5 cm−1) in C—H stretching, and red shifts (9, 10, 9 cm−1) in C-O-C stretching, respectively. Based on the magnitude of absorption peak shift, it could be inferred that TP primarily interacts with SeNPs through hydrogen bonding via O—H groups (O—H⋯Se), followed by hydrogen bonding of C—O⋯Se (Song et al., 2021). Although TP-SeNPs0.25 exhibited the largest shift in the peak of O—H stretching vibration, it showed the smallest shift in the peak of O—H bending vibration (1613 cm−1). Therefore, no significant differences were observed in the infrared absorption peak shifts among the three types of TP-SeNPs. However, based on the depth of the absorption peaks, it can be inferred that TP-SeNPs1.5 had the highest amount of TP coated, followed by TP-SeNPs3.0, and TP-SeNPs0.25 had the lowest (Gu et al., 2021), which was consistent with the TG result. As shown in Fig.5D, the XPS spectra of all SeNPs showed a Se 3d peak at ∼55.5 eV for elemental Se (Tang et al., 2021) and no peak at 59.5 eV for Se (IV) (Chen et al., 2023; Ye et al., 2020a), indicating complete reduction of Se (IV) to Se0. While the Se 3d₅/₂ and 3d₃/₂ binding energies of B-SeNPs were centered at 54.61 and 55.38 eV, those of TP-SeNPs0.25, TP-SeNPs1.5, and TP-SeNPs3.0 were located at 55.19/56.05, 55.17/56.04, and 55.16/56.09 eV, respectively. The observed shift of the Se 3d peak in TP-coated SeNPs implied the participation of Se 3d in PS bonding (Chen et al., 2022). No significant changes were observed in the Se 3d peak shifts of three different TP-SeNPs, indicating that the TP concentration does not affect the binding affinity between TP and SeNPs. In conclusion, the TP concentration did not affect the affinity between TP and SeNPs, but it influenced the amount of TP coated onto the SeNPs. A TP concentration of 1.5 mg/mL resulted in the highest coating amount of TP on the SeNPs, which could also account for its optimal stability. Mechanism of the influence of stabilizer concentration on SeNPs Previous research had indicated that the mechanism by which polysaccharides control the particle size of SeNPs involved limiting the process of Se0 species diffusing toward Se nuclei, along with the diffusion and collision of Se nuclei with each other (Dong et al., 2026). In previous growth curve study, it was also observed that a sufficient amount of TP could effectively limit the diffusion of Se nuclei as well as Se0 species (Fig. 4). However, it was less intuitive that when the TP concentration above 1.5 mg/mL (e.g., 2.0, 2.5, and 3.0 mg/mL), the particle size of TP-SeNPs gradually increased again. We speculate that this may be related to the hydrodynamic properties of TP at different concentrations. Growth curve of TP-SeNPs0.25 (A), TP-SeNPs1.5 (B), and TP-SeNPs3.0 (C). TP-SeNPs0.25, TP-SeNPs1.5 and TP-SeNPs3.0 represent TP-SeNPs prepared with 0.25 mg/mL, 1.5 mg/mL, and 3.0 mg/mL of TP, respectively. Fig. 6A shown that as TP concentration increased from 0.25 to 1.5 mg/mL, viscosity rose from 3.7 to 43.7 mPa·s, with both the overall viscosity and its rate of increase staying relatively low over this range. As the TP concentration further increased from 1.5 to 3.0 mg/mL, the corresponding viscosity increased rapidly from 43.7 to 266.3 mPa·s. The viscosity of the 1.5 mg/mL TP solution was essentially at the inflection point of the rate of increase; therefore, viscosity might be a factor influencing the role of TP in stabilizing SeNPs. Leangtanom et al. (2025) found that as the solvent viscosity increased from 4 to 90 mPa∙s, the size of silver phenylselenide nanocrystal decreased; however, at excessively high viscosity (1.0–1.5 Pa∙s), large particles formed instead. They attributed the reduced nanocrystal size to slower reactant mass transfer and reaction kinetics caused by increased solvent viscosity (Altorbaq et al., 2021), while excessively high viscosity hindered stirring, leading to nanocrystal agglomeration (Moral et al., 2023). Viscosity (A), hydrodynamic diameter (B) and atom force microscopy (AFM) images (C) of TP polysaccharide solutions at different concentrations; Zeta potential (D) and storage modulus (E) of B-SeNPs and TP-SeNPs prepared with various concentrations of TP. Studies have shown that an increase in PS concentration may lead to molecular entanglement and aggregation (Ye et al., 2020b), thereby causing an increase in solution viscosity (Guo et al., 2026). This was consistent with the finding in this study that the viscosity of the TP solution continuously increases with increasing concentration. Therefore, we confirmed the aggregation of TP by measuring the hydrodynamic diameter. As shown in Fig. 6B, TP (804 kDa, Fig. S4) exhibited a hydrodynamic diameter of 6.6 nm at 0.25 mg/mL, which increased only slightly to 9.7 nm at 1.5 mg/mL. This suggested that below 1.5 mg/mL, TP molecules were predominantly freely dispersed with negligible entanglement or aggregation, resulting in low solution viscosity. However, the hydrodynamic diameter of TP sharply increased to 47.3 nm at 3.0 mg/mL, with a few large particles (913.7 nm), indicating significant entanglement and aggregation, which aligned with the sharp rise in viscosity. TP aggregation may bridge Se nuclei, leading to an increase in SeNPs' size, which was already confirmed in Fig. 1. Additionally, self-aggregated TP would reduce its interaction with Se nuclei and SeNPs, resulting in insufficient coating of TP on SeNPs. The TG and FT-IR results confirmed this speculation, ultimately leading to an increase in SeNPs' size. The irregular spherical morphology of TP observed by AFM (Xu et al., 2020) may be attributable to the negative charges on both mica flakes and the polysaccharides (Yuan et al., 2022). Except for the higher particle count, the morphology of TP at 1.5 mg/L showed no significant difference from that at 0.25 mg/mL, with a height of 38.17 nm, which was close to the 41.42 nm observed at 0.25 mg/L (Fig. 6C). This indicates that neither concentration exhibited notable self-aggregation. Although the concentration of 3.0 mg/L TP was only twice that of 1.5 mg/mL TP, significant aggregation into large particles was observed, and the height sharply increased to 161.44 nm, indicating pronounced aggregation at 3.0 mg/mL. These results were in good agreement with the sharp increases in viscosity (Fig. 6A) and hydrodynamic diameter (Fig. 6B) at the same concentration. The aggregation behavior of TP may be closely related to its structural characteristics. The total sugar content of 89.9 ± 7.6% (w/w), the protein content of only 0.06 ± 0.01% (w/w) (Table 1), and the weak absorption peaks at 260–280 nm in the UV spectrum (corresponding to proteins and nucleic acids) (Fig. S5A) collectively indicated that TP possesses high purity with minimal interfering impurities. The broad molecular weight distribution (PDI > 2) was consistent with the typical features of native polysaccharides (Yang et al., 2018). Furthermore, the high content of uronic acid (17.1 ± 0.6%, w/w) (Table 1) imparted a strong aggregation tendency of TP via carboxyl-mediated hydrogen bonding (Pieczywek et al., 2021). Upon reaction with I₂–KI solution, the UV absorption spectrum of TP exhibited a pronounced peak at 350 nm (Fig. S5B), suggesting complex chain structures characterized by long side chains and numerous branches (Zhang et al., 2025b). The shape factor ρ (=Rg/Rh) and the conformational parameter v derived from the conformational power-law equation Rg = KgMv were important parameters for conformational analysis (Guo et al., 2020). The ρ value of 2.6 (Table 1), together with the conformational parameter ν = 0.90 ± 0.02 obtained from the double logarithmic plot of M versus Rg (Fig. S6), indicated that TP adopted an extended rigid conformation (Jia et al., 2025). TP exhibited relatively high intrinsic viscosity ([η] = 517.0 ± 1.5 mL/g) (Table 1), which may be attributed to its rigid molecular conformation tending to possess larger hydrodynamic dimensions, thereby resulting in higher viscosity (Jia et al., 2025). Table 1Basic physicochemical and conformational characterizations of TP.ItemsValueCompositionTotal sugar (%, w/w)89.9 ± 7.6Total protein content (%, w/w)0.06 ± 0.01Uronic acid (%, w/w)17.1 ± 0.6Molecular parametersMw (kDa)804.8Mn (kDa)304.7PDI (= Mw/Mn)2.64Rh (nm)34.9 ± 0.07Rg (nm)91.8 ± 0.1[η] (mL/g)517.0 ± 1.5Conformational parametersShape factor ρ (= Rg/Rh)2.6v of conformational power-law equation: Rg = Kg Mv0.90 ± 0.02Mw: weight-average molecular weight; Mn: number-average molecular weight; PDI: polydispersity index;Rh: hydrodynamic radius; Rg: radius of gyration; [η]: intrinsic viscosity; Kg: proportionality factor; M: Molar mass. Basic physicochemical and conformational characterizations of TP. Mw: weight-average molecular weight; Mn: number-average molecular weight; PDI: polydispersity index; Rh: hydrodynamic radius; Rg: radius of gyration; [η]: intrinsic viscosity; Kg: proportionality factor; M: Molar mass. The coating of TP on the surface of TP-SeNPs could also be characterized by changes in Zeta potential and storage modulus (G'). As depicted in Fig. 6D, The Zeta potential of TP (−56.4 ± 0.7 mV) and that of B-SeNPs (−6.13 ± 2.4 mV) differ significantly. The Zeta potential absolute values of TP-SeNPs0.25 (−13.5 ± 1.2), TP-SeNPs1.5 (−19.3 ± 0.3), and TP-SeNPs3.0 (−17.3 ± 0.1) were all significantly higher than that of B-SeNPs (P < 0.05), confirming successful coating of TP on the surface of TP-SeNPs (Dong et al., 2026; Yang et al., 2023). Among the three types of TP-SeNPs, the absolute Zeta potential of TP-SeNPs1.5 was significantly higher than those of the other two (P < 0.01), indicating that TP-SeNPs1.5 had the highest amount of TP coating on its surface. Meanwhile, the amount or density of TP coated on the surface of SeNPs was characterized by the storage modulus (G') measured via dynamic rheology. G' represents the energy stored in a viscoelastic system and is positively correlated with the grafting density and amount of the stabilizer on the nanoparticle surface (Feng et al., 2025). The G'of the four types of SeNPs measured at a constant oscillation frequency of 1 Hz within a shear strain range of 1–10% at 25 °C were shown in Fig. 6E. B-SeNPs exhibited the lowest storage modulus, indicating that they were coated with a trace amount of elastic coating and were prone to aggregation. TP-SeNPs1.5 showed the highest storage modulus, corresponding to the maximum amount of TP coating and the weakest tendency to aggregate. The low storage modulus of TP-SeNPs0.25 was attributed to an insufficient amount of TP, whereas the thin coating of TP-SeNPs3.0 was likely due to excessive self-aggregation of TP. The maximum amount coating obtained for TP-SeNPs1.5 not only gives them the largest absolute zeta potential value (Fig. 6D), but also provides the greatest steric stabilization effect (Zhan et al., 2025). Electrostatic repulsion from zeta potential and steric hindrance are the two main mechanisms that prevent nanoparticle aggregation and maintain them in a monodisperse stable state (Liu et al., 2024; Tao et al., 2025). However, steric stabilization offers robust stability that is insensitive to electrolytes and suitable for concentrated and mixed nanoparticle systems (Yu et al., 2017). This renders it the recommended approach for complex environments (Dong et al., 2026; Lim et al., 2024). Therefore, TP-SeNPs1.5 could maintain good stability under different conditions. Fig. 7 summarized the proposed mechanism underlying the effect of TP concentration on the formation of TP-SeNPs. The state of TP in aqueous solution varied with concentration. At low concentrations, they existed as individual molecules with low viscosity and small hydrodynamic diameter, and their self-association leads to increased viscosity and hydrodynamic diameter at high concentrations (Fig. 6A, B). Consequently, this may lead to different scenarios when TP was mixed with selenite at various concentrations. At an insufficient TP concentration (0.25 mg/mL), all TP were surrounded by SeO₃2− ions, leaving substantial free SeO₃2− in the solution. At the optimal concentration (1.5 mg/mL), nearly all SeO₃2− ions were adsorbed by TP via hydrogen bond (Gao et al., 2023; He et al., 2024). At excessive TP concentrations, all SeO₃2− ions were also adsorbed, but they coexist with TP that undergoes self-aggregation and does not adsorb SeO₃2−. As illustrated in Fig. 7, the reduction reaction was initiated by the addition of ascorbic acid (AA), SeO₃2− was initially reduced into Se nuclei, which subsequently aggregate to form SeNPs. Under conditions of insufficient TP (0.25 mg/mL), free SeO₃2− and the resulting Se0 species could diffuse toward the surfaces of Se nuclei, exhibiting a reaction-limited growth mode, which leads to the formation of larger SeNPs (126 nm). Furthermore, the surface of TP-SeNPs lacks a sufficient TP coating, making them prone to mutual aggregation. An appropriate TP concentration (1.5 mg/mL) effectively hinders the diffusion between Se nuclei, resulting in a diffusion-limited growth mode, which yields the smallest TP-SeNPs (56.8 nm). These NPs were coated with a thick layer of TP molecules, providing effective electrostatic repulsion and steric hindrance, thereby preventing aggregation and demonstrating optimal stability. An excessive TP concentration (3.0 mg/mL) could also effectively hinder the diffusion between Se nuclei. However, the increased viscosity due to the self-aggregation of excess TP reduced mixing efficiency, and coupled with large hydrodynamic volume and the bridging effect of TP, they caused premature aggregation of Se0 species and Se nuclei, resulting in an increased size of TP-SeNPs (71.8 nm). Due to the self-aggregation of TP, its adsorption efficiency onto SeNPs is reduced, and a substantial amount of free TP aggregates exists. Consequently, the PS coating amount on the surface of TP-SeNPs3.0 is not the maximum. These points have been substantiated through TG, FT-IR, Zeta potential, and storage modulus. Based on the mechanism by which TP concentration influences the formation of TP-SeNPs, it could be inferred that to stabilize SeNPs, the PS must be present in sufficient quantity, while self-aggregation that leads to increased viscosity and bridging effects should be avoided. Mechanism of the effect of TP concentration on the formation of TP-SeNPs. In vitro digestion release rate of SeNPs Fig. 8A shown the release profiles of B-SeNPs and TP-SeNPs during digestion. Very little Se was released from B-SeNPs (2.09 ± 0.63%) and TP-SeNPs (1.01 ± 0.36%) in the simulated oral digestion stage. In simulated gastric fluid (SGF), Se release from B-SeNPs increased sharply to 12.28 ± 0.63% at 120 min, whereas TP-SeNPs released only 3.83 ± 0.15%. In simulated intestinal fluid (SIF), the Se release rate from SeNPs reached a release rate of 24.12 ± 1.33% at 250 min, while TP-SeNPs showed a slight initial rise then stabilization, with overall release of 10.14 ± 0.44%. Owing to the protection of TP, the Se release rate of TP-SeNPs after simulated gastrointestinal digestion was much lower than that of B-SeNPs. Consistent with the release profiles, similar results were obtained for the morphological changes of B-SeNPs and TP-SeNPs during digestion (Fig. 8B). Before digestion, B-SeNPs were large, irregular, and partially aggregated, while TP-SeNPs were small, regular, and monodisperse. After SGF digestion, B-SeNPs showed blurred boundaries and worsened aggregation, with further TEM contrast loss after SIF exposure. In contrast, TP-SeNPs exhibited only moderate aggregation after SGF, while their TEM contrast remained largely unchanged, indicating resistance to gastric digestion. After SIF digestion, a small number of low-contrast particles with blurred boundaries appeared, indicating that a small fraction of TP-SeNPs were digested. Overall, TP-SeNPs retained a monodisperse regular spherical morphology after gastrointestinal digestion, enabling intact TP-SeNPs to successfully cross the intestinal barrier, which has positive implications for enhancing the bioavailability and biological effects of SeNPs. The high bioavailability of SeNPs is attributed to their nanoscale size-mediated mucus penetration and endocytic uptake across the intestinal epithelium (Abdollahi et al., 2025; Redolfi-Bristol et al., 2025). Their low toxicity, on the other hand, stems from the low chemical reactivity of elemental selenium, which reduces oxidative degradation and conversion, enables controlled release (K et al., 2022; Márquez et al., 2020), and supports safer administration at higher doses (Ye et al., 2020b). Furthermore, Se in SeNPs predominantly exists in a cyclic (Se5) conformation, which facilitates its intracellular conversion into selenoproteins and bioactive secondary metabolites, thereby exerting stronger biological effects (He et al., 2025; Xiong et al., 2025). However, these benefits depend on whether SeNPs can maintain their nanoform without rapid degradation during gastrointestinal digestion (Tian et al., 2026). Therefore, TP-SeNPs capable of resisting simulated gastrointestinal digestion in vitro hold the potential for high bioavailability and low toxicity, its genuine bioactivity remains to be confirmed by further in vivo studies, which was a limitation of the present study. In vitro release profile of Se from B-SeNPs and TP-SeNPs1.5 in simulated gastric fluid (SGF) and simulated intestinal fluid (SIF) (A) and the TEM images of SeNPs and TP-SeNPs1.5 during the in vitro gastrointestinal digestion (B). Conclusion In this study, TP was used as a stabilizer to systematically investigate the influence of stabilizer concentration on the morphology formation of SeNPs. The results indicated that there existed an optimal range of stabilizer concentration. Insufficiently low concentrations failed to effectively inhibit the diffusion and collision of Se0 species and Se nuclei, leading to an increase in nanoparticle size. However, an extremely high concentration accelerated polysaccharide self-aggregation and significantly increased viscosity. While this restricted the diffusion of Se0 species and Se nuclei, the high viscosity reduced stirring efficiency and generated bridging effects, encouraging the aggregation of primary Se nuclei and producing bigger nanoparticles. TG analysis, Zeta potential, and storage modulus further confirmed that an excessively high stabilizer concentration did not provide the maximum coating amount on TP-SeNPs, demonstrating that the self-aggregation of polysaccharides reduced the amount available for nanoparticle formation. In summary, the concentration of polysaccharide stabilizers should be chosen before severe self-aggregation occurs, and maintained at an appropriate ratio with selenite anions. Under such conditions, TP-SeNPs with the smallest particle size and optimal stability could be obtained, achieving synergistic stabilization through electrostatic repulsion and steric hindrance. This study provided a theoretical basis for optimizing stabilizer concentration in the preparation of SeNPs. Furthermore, the as-prepared TP-SeNPs could withstand simulated gastrointestinal digestion and maintain a stable, monodisperse spherical structure, demonstrating great potential as a low-toxicity, high-bioavailability, and high-bioactivity Se supplement. Corroboration of the in vivo efficacy of bioactives through animal experiments deserves attention in future research. CRediT authorship contribution statement Jianwei Dong: Writing – review & editing, Supervision, Investigation, Funding acquisition, Conceptualization. Jingyuan Liu: Writing – original draft, Methodology, Formal analysis. Huan Wang: Writing – original draft, Project administration, Methodology. Jiahao Zhang: Writing – original draft, Visualization, Resources. Zhihai Wei: Writing – original draft, Visualization, Validation, Data curation. Da Song: Validation, Investigation, Funding acquisition, Data curation. Hao Han: Validation, Investigation, Funding acquisition. Jingzhang Geng: Writing – review & editing, Supervision, Funding acquisition. Ethics declaration The author(s) declare(s) that the study does not involve humans nor animal subjects. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments This work was supported by the Key Research and Development Program of Shaanxi Province (Nos. 2025NC-YBXM-133 & 2025QY-GJHX-44), the Local Special Service Program of Shaanxi Provincial Education Department (No. 24JC025), the Scientific Research Program of Shaanxi University of Technology (Nos. SLGRC202404 & SLGRC202409), and the Key Scientific Research Program of Shaanxi Provincial Department of Education (No. 23JY018). Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2026.104404. Contributor Information Jianwei Dong, Email: jwdong@snut.edu.cn. Jingzhang Geng, Email: gengjingzhang@163.com. Supplementary data Supplementary material Data availability The original contributions presented in the study are included in the article/Supplementary Material; further inquiries can be directed to the corresponding author. References Associated Data Supplementary Materials Supplementary material Data Availability Statement The original contributions presented in the study are included in the article/Supplementary Material; further inquiries can be directed to the corresponding author.