The experiment was conducted in a greenhouse at the Guangxi Academy of Agricultural Sciences Research and Experiment Base from March to August 2023. Potting soil was collected from Yandong Township, Debao County, Baise City, Guangxi Province. Its basic physicochemical properties were: organic matter 30.3 g/kg, total nitrogen 1.8 g/kg, total phosphorus 1.25 g/kg, total potassium 29.22 g/kg, available nitrogen 92 mg/kg, available phosphorus 77 mg/kg, available potassium 183 mg/kg, pH 7.0, and total selenium 0.82 mg/kg (Table 1).

The soil conditioners used were as follows: organic fertilizer containing 461 g/kg organic matter, N 29 g/kg, P_2O5 11 g/kg, K₂O 19 g/kg, pH 5.5; potassium humate with 373 g/kg organic matter, N 24.1 g/kg, P₂₀₅ 6 g/kg, K₂O 61.2 g/kg; and biochar with 423 g/kg organic C, total nitrogen 11.1 g/kg, P₂₀₅ 7 g/kg, K₂O 32.6 g/kg, pH 9.3. The maize variety used was Guidan 162.

The experiment was performed in pots with five treatments: organic fertilizer (OF, 3.13 g/kg soil), quicklime (QL, 1.25 g/kg soil), biochar (BC, 10 g/kg soil), potassium humate (FA, 5 g/kg soil), and a conventional fertilization control (CK). Chemical fertilizers were applied at the same rate across all treatments. Each treatment was replicated four times. All soil conditioners were thoroughly mixed with the soil and then placed in the pots before planting the first batch of corn. No soil conditioners were added for the second and third batches of corn planted thereafter.

Biochar was thoroughly mixed with the soil before planting the first maize batch; no biochar was added for subsequent batches. Each batch received only basal fertilizer (compound fertilizer 15-15-15, 3.00 g/pot), mixed with the soil prior to planting. Maize seeds were sown at five per pot, thinned to two plants per pot at the three-leaf stage. Irrigation was applied daily to maintain 70% of field capacity. Three consecutive maize batches were planted, each grown for 30 days.

The data were organized using Microsoft Excel 2010 (Microsoft Corp., Raymond, Washington, USA) and SPSS version 26. ANOVA was employed to analyze the effects of differences between treatments were assessed by least significant difference (LSD) tests (P < 0.05). Figures, correlation analysis and plotting were conducted using Origin 2021 (Origin Lab, Corp., Hampton, Massachusetts, USA).

Figure 1 illustrates the dynamics of these parameters under different soil conditioner treatments. Soil pH under lime, biochar, and conventional treatments initially increased and then gradually declined. In contrast, potassium humate treatment exhibited a decrease followed by an increase and a subsequent decline, while organic fertilizer initially reduced pH, which then gradually increased. Among all treatments, lime consistently maintained the highest pH. Except for the first maize batch under organic fertilizer, which showed significantly lower pH than biochar and conventional treatments, there were no significant pH differences among biochar, conventional, potassium humate, and organic fertilizer treatments after the second and third maize batches. Overall, pH remained higher under biochar and conventional treatments compared to potassium humate and organic fertilizer.

Available phosphorus levels were significantly higher only under the conventional treatment during the first two maize batches, while no significant differences were observed among lime, biochar, potassium humate, and organic fertilizer treatments. After the third batch, available phosphorus levels showed no significant differences among any treatments.

Available sulfur content was significantly higher in soils treated with organic fertilizer after the first and second maize batches, whereas other treatments showed no significant differences. In the third batch, available sulfur increased under potassium humate treatment, resulting in both organic fertilizer and potassium humate treatments having significantly higher sulfur content than the other three treatments. No significant differences were observed between organic fertilizer and potassium humate, or among lime, biochar, and conventional treatments.

Table 2 presents the proportions of soil selenium forms following application of different soil conditioners. After the first maize batch, soil conditioners primarily influenced exchangeable and residual selenium, whereas soluble selenium, iron-manganese-bound selenium, and organically bound selenium showed no significant differences among treatments. Exchangeable selenium was highest under organic fertilizer, significantly exceeding all other treatments, and lowest under the control (CK). Residual selenium was highest in the biochar treatment, surpassing lime, potassium humate, organic fertilizer, and CK by 3.91, 5.03, 5.78, and 2.97 percentage points, respectively. Available selenium proportions in organic fertilizer, potassium humate, lime, and biochar treatments were 3.79%, 2.38%, 2.56%, and 2.23%, respectively, all higher than CK.

After the second maize harvest, soil conditioners significantly affected soluble, exchangeable, and organically bound selenium, while iron-manganese-bound and residual selenium remained unaffected (Table 2). Biochar and lime significantly increased soluble selenium, whereas potassium humate and organic fertilizer treatments were comparable to CK. Exchangeable selenium increased significantly in the biochar treatment, but decreased under lime, potassium humate, and organic fertilizer compared to CK. Organically bound selenium did not differ significantly from CK, though organic fertilizer treatment exceeded biochar. Available selenium decreased in organic fertilizer, potassium humate, and lime treatments but increased in the biochar treatment relative to CK.

After the third maize batch, soil conditioners significantly influenced soluble, exchangeable, and residual selenium, but not iron-manganese-bound or organically bound selenium (Table 2). Biochar, lime, and potassium humate increased soluble selenium; lime also enhanced exchangeable selenium, while biochar and potassium humate increased it by 2.14 and 1.61 percentage points, respectively. Exchangeable and soluble selenium under organic fertilizer did not differ from CK. Residual selenium was significantly higher in potassium humate than in biochar, while CK values were comparable across treatments. Available selenium decreased by 0.57 percentage points under organic fertilizer, but increased by 1.25, 2.81, and 0.74 percentage points under potassium humate, lime, and biochar, respectively.

Overall, soil selenium continuously transforms among its forms. Iron-manganese-bound and residual selenium vary significantly between batches, reflecting dynamic soil-environment interactions. Despite these transformations, soluble and exchangeable selenium, which are directly available to crops, remain consistently low.

Total biomass of the first maize batch followed the order: biochar > organic fertilizer > CK > potassium humate > lime (Figure 2). Compared to CK, biochar increased total biomass by 7.89%, whereas lime reduced it by 41.17%. Biochar notably enhanced root biomass by 30.20%, while lime reduced root biomass by 18.54%. Stem and leaf biomass did not differ significantly among biochar, potassium humate, organic fertilizer, and CK; however, lime significantly decreased stem and leaf biomass by 46.39%.

After the second maize batch, total biomass ranked as potassium humate > organic fertilizer > CK > biochar > lime (Figure 2). Root biomass did not differ significantly among most treatments and the control, although potassium humate roots were significantly heavier than those under biochar and lime. Stem and leaf biomass was highest under potassium humate, increasing 25.85% compared to CK, whereas lime reduced stem and leaf biomass by 18.29%. Organic fertilizer and biochar had no significant effect on stem and leaf biomass relative to CK.

Following the third maize batch, total biomass followed the order: potassium humate > organic fertilizer > biochar > CK > lime (Figure 2). Total biomass increased by 9.04%, 4.42%, and 2.54% under potassium humate, organic fertilizer, and biochar, respectively, relative to CK, whereas lime reduced total biomass by 19.42%. Potassium humate significantly enhanced root biomass by 29.53%, while other conditioners showed no significant difference compared to CK. Lime treatment caused a 21.42% reduction in stem and leaf biomass, whereas the other treatments were slightly higher than CK but not significantly different. Across the three maize batches, lime consistently inhibited growth. Organic fertilizers, potassium humate, and biochar exerted modest growth-promoting effects that became more apparent across consecutive plantings, with potassium humate showing the strongest positive impact on maize biomass.

Correlation analysis between maize selenium nutritional indicators and key soil physicochemical properties and selenium speciation is presented in Table 4. Among selenium fractions, soluble selenium showed significant or highly significant positive correlations with selenium content in stems and leaves, selenium accumulation in stems and leaves, and the selenium transfer coefficient. Iron–manganese–bound selenium exhibited highly significant positive correlations with root selenium content, stem and leaf selenium content, selenium accumulation in both roots and shoots, total plant selenium accumulation, and the selenium transfer coefficient. In contrast, residual selenium was highly negatively correlated with all of these indicators. Exchangeable and organically bound selenium showed no significant correlations with maize selenium nutritional parameters.

Regarding soil properties, soil pH displayed significant or highly significant positive correlations with root selenium content, stem and leaf selenium content, selenium accumulation in roots and shoots, total selenium accumulation, and the selenium transfer coefficient. Among available nutrients, available sulfur showed a significant negative correlation only with the selenium transfer coefficient, whereas available phosphorus displayed no significant associations.

Overall, these findings indicate that soil pH is a key environmental factor regulating maize selenium nutrition under soil conditioner treatments. Soluble selenium, iron–manganese–bound selenium, and residual selenium serve as critical indicators of selenium uptake and accumulation, while available sulfur significantly influences selenium translocation within the plant.

Figure 5 shows principal component analysis (PCA) of the three batches of maize data with added soil conditioner. The results show that the first two principal components (PC1 and PC2) cumulatively explained 60.5%, 60.1%, and 59.9% of the total variance, respectively, indicating that these two dimensions can reflect the original structure of the data well. In the first batch of maize, the lime application treatment showed a significant difference from the control (CK) treatment. Soil pH, available phosphorus in the soil, and selenium content in different parts of the maize plant were closely related to lime application. In the second and third batches of maize planting, there was significant overlap among the treatments, indicating a weakening of the treatment effect. In the second batch of maize, soluble selenium content, transfer coefficient, selenium content in maize stems and leaves, and soil pH were closely related; selenium content in maize roots was closely related to available phosphorus in the soil. In the third batch of maize, the selenium content in both maize roots and stems and leaves was closely related to exchangeable selenium, while the selenium content in maize stems and leaves, soil pH, and selenium transfer coefficient were closely related.

This study continuously monitored shifts in soil selenium speciation across three sequential maize seedling harvests and found substantial variation in several selenium fractions among batches under the same treatment, particularly in iron–manganese-bound and residual selenium (Table 1). Soil pH, available phosphorus, and available sulfur are key factors influencing soil selenium speciation and crop selenium uptake. These fluctuations were likely driven not solely by the applied soil conditioners but also by interacting climatic, hydrologic, and crop-management factors. Research on arsenic speciation indicates that arsenic is initially adsorbed onto iron–aluminum clay mineral surfaces and, as soil pH fluctuates, these minerals transform and incorporate the adsorbed arsenic into their lattices, resulting in more stable fixation (Ran, 2023). Given the structural and chemical similarities between selenium and arsenic, iron–manganese-bound and residual selenium likely undergo comparable inter-conversion during clay-mineral transformations under variable environmental conditions.

Soil conditioners significantly influenced exchangeable selenium after all three maize harvests; soluble selenium after the second and third harvests; residual selenium during the first and third harvests; and organically bound selenium after the second harvest (Table 1). Soluble and exchangeable selenium represent the bioavailable fractions in soil, and their availability is strongly regulated by soil pH and organic matter content (Liu et al., 2015). The addition of soil conditioners markedly altered soil pH (Figure 1), with treatments containing organic fertilizer or potassium humate producing the lowest pH values and correspondingly the lowest available selenium.

The four soil conditioners used in this study differed substantially in organic-matter characteristics, which influenced selenium sorption and stabilization. Because selenium fixation by organic matter requires time to develop, significant differences in organically bound selenium among treatments did not emerge until the second harvest. Organic fertilizer and potassium humate, both rich in active organic components, produced higher levels of organically bound selenium at that stage. Biochar, characterized by a more stable carbon matrix, exerted a delayed effect, displaying elevated organically bound selenium only after the third maize harvest.

The observed effects of soil conditioners on residual selenium were consistent with previous findings (Tan et al., 2024; Lu et al., 2023). Residual selenium, immobilized within mineral lattices, is largely inaccessible under stable natural conditions. During the first planting cycle, the combined initial activation from irrigation, maize cultivation, and soil-conditioner application likely disrupted lattice structures, leading to treatment-specific alterations in residual selenium. The pronounced differences among conditioners consequently produced varying degrees of influence on this most recalcitrant selenium fraction.

The application of soil conditioners increased selenium concentrations in maize roots, though the magnitude of this increase varied markedly among treatments (Figure 3). This pattern is consistent with previous findings. For example, Huang et al. (2019) reported that selenium uptake by rice roots differed substantially depending on how sodium selenite was combined with various organic materials, while Ma et al. (2023) found that biochar application in dry-land systems enhanced plant selenium absorption. These differences arise because soil conditioners modify soil physicochemical properties and selenium speciation, influencing plant uptake of distinct selenium chemical forms (Dinh et al., 2019).

Lime application significantly elevated root selenium content, largely due to its strong alkalizing effect on soil pH (Figure 1). Soil pH showed a highly significant positive correlation with selenium content in both roots and leaves (Table 4), consistent with earlier studies (Lu et al., 2023). Although all conditioners increased root selenium, the selenium content in the aboveground tissues of maize treated with organic fertilizer, potassium humate, or biochar was lower than that of the control (Figure 3). This suggests that selenium absorbed in these treatments was not effectively translocated from the roots to shoots. Excessive selenium (Se) acts as a pro-oxidant in plants, causing significant toxicity (selenosis) characterized by stunted growth, root inhibition, chlorosis, and necrosis (Gu et al., 2025). It causes damage by replacing sulfur in amino acids (cysteine and methionine), resulting in dysfunctional proteins, alongside inducing severe oxidative stress (Cao et al., 2025).

Crops can absorb all available selenium species in soil, including Se(IV), Se(VI), and dissolved organic selenium. However, selenium must undergo intracellular transformation before being transported to aboveground tissues. Se(IV) taken up by roots is quickly converted into organoselenium compounds primarily selenomethionine and methylselenocysteine most of which remain in the roots, with only limited translocation to stems (Winkel et al., 2015). In contrast, selenate (Se(VI)) is readily transported through the xylem to aboveground tissues, where it is reduced to Se(IV); following reduction, selenium is incorporated into organoselenium forms and redistributed to various tissues through both xylem and phloem pathways [32]. Transport efficiencies differ widely across chemical forms, with the general order Se(VI) > SeMet > Se(IV)/SeCys (Kikkert and Berkelaar, 2013).

Different soil selenium fractions also contain distinct chemical species (Qu et al., 1997). Soluble selenium is dominated by SeO₄²^-, with smaller proportions of SeO₃²^-and soluble organic selenium, whereas exchangeable selenium exists mainly as SeO₃²^-(Gissel et al., 1984; White, 2016). In this study, soil conditioners significantly affected exchangeable selenium (Table 1), and selenium content and transfer coefficients in stems and leaves were significantly positively correlated with soluble selenium (Table 4). This indicates that the application of soil conditioners on dryland may mainly affect the content of Se(IV) in available selenium in the soil, thereby affecting the absorption and accumulation of selenium by crops.

Furthermore, plant absorption of selenium involves complex synergistic and competitive relationships with other nutrients. Generally, Se(IV) in the soil is absorbed by phosphorus transport proteins (Zhang et al., 2014), while Se(VI) mainly enters the plant root system through sulfur transport proteins (Gupta and Gupta, 2017). Therefore, the availability of phosphorus and sulfur in the soil plays an important regulatory role in the efficiency of selenium translocation, absorption, and accumulation in plants. Experiments have shown that the combined application of phosphorus and Se(IV) significantly reduces selenium translocation and accumulation in wheat, but the combined application of phosphorus and Se(VI) increases selenium absorption and accumulation in wheat (Zhang et al., 2017). The effect of sulfur on plant selenium absorption and accumulation also varies depending on the crop’s sulfur status (Stroud et al., 2010). The addition of soil conditioners also significantly affects the transformation characteristics of available phosphorus, available sulfur, and selenium forms in the soil, thereby affecting the absorption and translocation of selenium by maize plants. However, as mentioned above, we regret that we did not detect changes in soil Se(IV) and Se(VI). Therefore, we believe that future related studies must measure changes in soil Se(IV) and Se(VI) content to help understand the mechanism of selenium transfer and transformation in soil and plants.