Effects of pre- and post-pyrolysis potassium silicate enrichment on potassium fractionation and release behavior of rice husk and sugarcane press mud biochars

1 Introduction

Potassium (K) is a key macronutrient that regulates stomatal conductance, enzyme activation, osmoregulation, assimilate transport, and stress tolerance (Ortel et al., 2024). In modern agriculture, potassium fertilizers are extensively applied in order to ensure the yield and fertility of the crops (Limmer and Seyfferth, 2026). But, with the growing global demand for potash fertilizers and limited availability of the potash reserves in an uneven geographic distribution, it has raised concerns regarding long-term nutrient security and sustainability of potassium fertilizer supply (Subbiah et al, 2025). Resource recovery and waste valorization are promising strategies for improving nutrient security by converting agro-industrial residues into value-added materials while promoting circular-economy goal.

Rice husk and sugarcane press mud are abundant residues from rice milling industry and sugar processing industry, respectively. Rice husk is silica-rich and structurally stable (Premchand et al., 2024), while press mud is high in organic matter and plant nutrients (Ouedraogo et al., 2026). Converting these residues to the biochar addresses waste management and simultaneously creates a porous, carbon-rich carrier suitable for nutrient enrichment and controlled-release biochar. Biochar physicochemical properties contribute to nutrient holding capacity, minimize leaching loss, and increase fertilizer use efficiency (Limmer and Seyfferth, 2026). Moreover, biochar properties can be modified by choosing the right feedstock and methods to optimize them for specific agricultural uses (Premchand et al., 2024; Ravindiran et al., 2024).

Potassium enrichment stands out as one of the most promising biochar modification methods for creating sustainable potassium fertilizers that offer enhanced nutrient availability and release properties (Liu et al., 2024; Yang et al., 2024). Enrichment can be performed either before pyrolysis (co-pyrolysis or biomass impregnation) or after pyrolysis (post-pyrolysis impregnation/impregnated-drying). Potassium binding mechanisms and their release behavior, mineral phase formation, and nutrient fractionation may vary significantly depending on the stage of introduction of the potassium (Yang et al., 2024; Tan et al., 2026). Potassium silicate shows promise as an enrichment agent, as it not only supplies potassium but also interacts with mineral components in the biochar matrix, potentially impacting nutrient stabilization and release (Abdullah et al., 2025; Park et al., 2026).

Comparative studies have examined mineral amendments and potassium sources for modifying biochar properties and nutrient availability. Bentonite incorporation has been investigated for mineral enrichment of manure biochar (Rupngam et al., 2025), while potassium silicate has been investigated as a potassium-enrichment agent during pyrolysis, including through co-pyrolysis with naturally potassium-rich silicate minerals such as glauconite (Leite et al., 2024). However, comparatively limited information is available on potassium silicate enrichment across contrasting agro-industrial feedstocks, particularly in relation to potassium incorporation, mineral-phase characteristics, potassium fractionation, and subsequent release behavior. In addition, the relationships among potassium enrichment, mineral-phase transformation, potassium fractionation, and subsequent release behavior have received limited attention in the existing literature (Yang et al., 2024; Tan et al., 2026).

Accordingly, the present study aimed to produce potassium-enriched biochars from rice husk and press mud using pre- and post-pyrolysis potassium silicate enrichment strategies and to comparatively evaluate their physicochemical and mineralogical properties, potassium fractionation, and potassium release behavior. Rice husk and press mud were selected as contrasting feedstocks because of their differences in native silicon and potassium contents. This comparison was used to examine how feedstock composition and enrichment strategy influence potassium incorporation, mineral-phase characteristics, potassium fractionation, and release behavior, thereby providing an integrated assessment of potassium enrichment in contrasting agro-industrial biochars.

2 Materials and methods

2.1 Materials

Rice husk was collected from the rice mill and press mud from the sugar processing factory. The source of potassium enrichment was potassium silicate (K₂SiO₃, Sigma-Aldrich). All other chemicals and reagents were of analytical grade and were used without further purification.

2.2 Feedstock preparation

The collected rice husk and press mud were washed with deionized water to free the adhered impurities and foreign materials. All the feedstocks were air-dried for two days under shade conditions and then oven dried to constant weight at 105 °C. The dried materials were ground mechanically and sieved to less than 1 mm in size prior to the enrichment and pyrolysis.

2.3 Biochar production

For the control treatments, the prepared rice husk and press mud feedstocks were pyrolyzed without the addition of potassium silicate. For each treatment, 500 g of homogenized feedstock was divided into four 125 g sub-samples, each processed as an independent pyrolysis replicate. Each sub-sample was placed in a laboratory-scale batch pyrolysis reactor (2 kg capacity) and pyrolyzed under oxygen-limited conditions at 450 °C for 1 h at a heating rate of 10 °C min⁻¹. Laboratory-grade nitrogen gas was continuously purged throughout the pyrolysis process to maintain an inert atmosphere. Following pyrolysis, the reactor was allowed to cool naturally to room temperature before the biochars were collected. The resulting biochars were ground, homogenized, and stored in airtight polyethylene containers for subsequent analyses.

2.4 Pre-pyrolysis potassium silicate enrichment

For pre-pyrolysis enrichment, potassium silicate was incorporated into rice husk and press mud at a rate of 5% (w/w, dry weight basis). The required quantity of potassium silicate was dispersed in deionized water and mixed thoroughly with the feedstocks using a water-to-feedstock ratio of 8:1 (v/w). The mixtures were stirred thoroughly to ensure uniform distribution of potassium silicate throughout the biomass matrix and allowed to equilibrate for 30 min. Subsequently, the mixtures were oven-dried at 65 °C for 6–8 h and then pyrolyzed under the conditions described in Section 2.3. The resulting enriched biochars were designated as RH-Pre and PM-Pre for rice husk and press mud, respectively.

2.5 Post-pyrolysis potassium silicate enrichment

For post-pyrolysis enrichment, rice husk biochar and press mud biochar produced as described in Section 2.3 were used. Potassium silicate was incorporated at a rate of 5% (w/w, dry weight basis). The required quantity of potassium silicate was dispersed in deionized water and mixed with the biochars at a water-to-biochar ratio of 8:1 (v/w). The mixtures were stirred continuously and allowed to equilibrate for 30 min to facilitate uniform coating and adsorption of potassium silicate onto the biochar surface. The enriched biochars were then oven-dried at 65 °C for 6–8 h until constant weight was achieved. The biochars produced were referred to as RH-Post and PM-Post for rice husk and press mud, respectively.

2.6 Experimental treatments

Six treatments were prepared in the present study and are summarized in Table 1.

Table 1. Experimental treatments.

For each treatment, four independent pyrolysis replicates were produced as described in Section 2.3 (n = 4), and these replicates were used for the physicochemical properties and potassium fractionation analyses. For the potassium release kinetics experiment, three replicates were used (n = 3 per treatment at each of the 9 sampling times).

2.7 Physicochemical characterization of biochars

2.7.1 Biochar yield

Biochar yield was determined gravimetrically and expressed as the percentage of solid recovery relative to the initial dry feedstock mass (Sahoo et al., 2021).

Biochar Yield(%)=(Weight of biochar/Weight of feedstock)×100

2.7.2 pH and electrical conductivity

The pH and electrical conductivity (EC) of feedstocks and biochars were measured using a digital pH and conductivity meter fitted with a glass electrode (Systronics). Suspensions were prepared in deionized water at 1:20 (w/v) and stirred for 1.5 h at room temperature and then equilibrated for 30 min before measurement (Khater et al., 2024).

2.7.3 Total potassium content

To determine the total K, 0.5 g of finely ground biochar and feedstocks was digested with nitric acid–hydrogen peroxide solution (3:1, v/v) under heated conditions until complete dissolution was achieved. The digest was then filtered through Whatman No. 42 filter paper and diluted to a known volume with deionized water. Potassium concentration was determined using a Systronics 128 flame photometer (Biswas et al., 2023).

2.7.4 Total silicon content

Total silicon (Si) content of rice husk (RH) and press mud (PM) feedstocks was determined following the method of Le Blond et al. (2011) with minor modifications. Approximately 100 mg of oven-dried, finely ground sample was digested using concentrated HNO₃, HCl, and HF in a closed-vessel microwave digestion system. After digestion, residual HF was complexed with 4% H₃BO₃, and the digest was diluted to 250 mL with deionized water. Silicon concentration was determined using ICP-MS (PerkinElmer) calibrated with a PerkinElmer TruQ Multi-Element Calibration Standard. Results are expressed as mg g⁻¹ dry weight.

2.7.5 Moisture content

Moisture content was determined by drying 1 g of the biochar sample and feedstocks at 105 °C for 2 h until a constant weight was attained (Sathyabama and Firdous, 2025).

Moisture(%)=[(Mass before drying−Mass after drying)/Mass before drying]×100

2.7.6 Ash content

Ash content was determined by heating 1 g of biochar and feedstock samples in a pre-weighed crucible at 750 °C for 6 h in a muffle furnace. After cooling in a desiccator, the residue was weighed (Sathyabama and Firdous, 2025).

Ash(%)=(Mass of residue after heating/Initial mass of sample)×100

2.7.7 Volatile matter content

Volatile matter content was determined by heating 1 g of oven-dried biochar and feedstock samples at 750 °C for 11 min (Sathyabama and Firdous, 2025).

Volatile Matter(%)=[(Mass before heating−Mass after heating)/Mass before heating]×100

2.7.8 Fixed carbon content

Fixed carbon content was calculated by difference according to the following equation (Sathyabama and Firdous, 2025).

Fixed Carbon(%)=100−[Moisture(%)+Ash(%)+Volatile Matter(%)]

2.8 X-ray diffraction analysis

Mineralogical characterization of the biochars was performed using a Bruker D8 Advance X-ray diffractometer equipped with Cu Kα radiation (λ = 1.5406 Å), operated at 40 kV and 40 mA. Finely ground biochar samples were mounted on sample holders and scanned over a 2θ range of 1–80° at room temperature with a scan speed of 2° min⁻¹ and a step size of 0.02°. The diffraction patterns were analyzed using EVA software, and crystalline phases were identified by comparison with standard reference patterns from the International Centre for Diffraction Data (ICDD) database.

2.9 Fourier transform infrared spectroscopy

Surface functional groups present in the biochars were identified using Fourier Transform Infrared Spectroscopy employing Shimadzu IR Affinity-1 and Thermo Nicolet iS50 instruments operating in KBr mode. Approximately 1 mg of finely ground biochar was thoroughly mixed with 100 mg spectroscopic-grade KBr and compressed into transparent pellets. Spectra were recorded over the range of 4000–500 cm⁻¹ at a resolution of 4 cm⁻¹ with 32 scans per sample.

2.10 Potassium fractionation analysis

Potassium fractionation in the biochars was determined using the chemical fractionation analysis (CFA) procedure originally proposed by Miller and Given (1986) and subsequently modified by Baxter et al. (1998). The method operationally separates potassium into water-soluble, ammonium acetate-extractable, hydrochloric acid-extractable, and residual fractions.

Briefly, 0.2 g of biochar was mixed with 30 mL of deionized water and agitated at 80 rpm for 24 h. The suspension was centrifuged at 4000 rpm for 10 min, and the supernatant was collected for potassium determination. The residual solid was subsequently extracted using 1 M NH₄OAc followed by 1 M HCl. After each extraction step, the residue was washed with 15 mL of deionized water before proceeding to the next extractant. The residual potassium fraction was quantified following triple-acid digestion of the remaining solid using a mixture of HClO₄, HF, and HNO₃ at 180 °C for 10 h. Potassium concentrations in all extracts were determined using a flame photometer.

The recovery rate of potassium in biochars after pyrolysis was calculated according to the following equation:

R(%)=[(KBC×Y)/KFS]×100

where R is the potassium recovery rate (%), KBC is the total potassium content in biochar, KFS is the total potassium content in the corresponding feedstock, and Y is the biochar yield expressed as a decimal fraction.

For the control biochars (RH-BC and PM-BC), K_FS corresponded to the native potassium content of the feedstock. For the pre- and post-pyrolysis potassium silicate-enriched treatments (RH-Pre, PM-Pre, RH-Post, and PM-Post), K_FS was calculated as the sum of the native feedstock potassium and the potassium supplied through the 5% (w/w) potassium silicate amendment, thereby representing the total potassium input used in the recovery calculation.

2.11 Potassium release experiment

A batch leaching method was used to measure the release of potassium from the biochars under controlled conditions in the lab. Deionized water was used as the leaching medium to provide a standardized and reproducible condition for comparing the readily soluble K fraction among the biochar treatments (Chen et al., 2023). One gram of biochar was added to 200 mL of deionized water to create a uniform biochar suspension. The suspensions were shaken at 120 rpm with a Remi horizontal orbital shaker at ambient temperature (25 ± 2 °C).

Aliquots (5 mL) were withdrawn at predetermined intervals of 0.5, 1, 2, 4, 6, 8, 24, 48, 72, and 96 h and immediately filtered through Whatman No. 42 filter paper. A flame photometer was used to measure potassium concentration in the filtrates. The cumulative potassium released was calculated after correcting for the withdrawn volume. Results were expressed as cumulative potassium released per gram of biochar (mg g⁻¹ biochar) (Biswas et al., 2023; An et al., 2020).

Leachate samples were diluted ten-fold prior to flame photometric analysis. The measured concentration was corrected for this dilution factor before calculating cumulative release, as follows: Kt = (C × DF × V)/m, where C is the potassium concentration measured by flame photometry (mg L⁻¹), DF is the dilution factor (10), V is the extraction volume (0.2 L), and m is the mass of biochar (1 g).

2.12 Potassium release kinetic modelling

The potassium release data obtained from the leaching experiment were fitted to first-order, second-order, Elovich, and parabolic diffusion kinetic models (Biswas et al., 2023; An et al., 2020). All models were fitted directly to the experimental data in their non-linear form using nonlinear least-squares regression (nls function, R v4.2.1) (Kumar, 2007; Lin and Wang, 2009). The mathematical expressions used for data fitting are presented in Table 2.

Table 2. Kinetic models used for describing potassium release.

Kt, cumulative potassium released at time t (mg g⁻¹); t, time (h).

The suitability of each model was evaluated using the coefficient of determination (R²), root mean square error (RMSE), and Akaike Information Criterion (AIC), rather than R² alone, to provide a more robust comparison of model performance. The model exhibiting the highest R², lowest RMSE, and lowest AIC was considered the most appropriate for describing potassium release behavior for each treatment.

2.13 Statistical analysis

The data were analyzed with R software using RStudio (version 4.2.1; R Core Team, 2022). Normality of residuals and homogeneity of variance were initially tested with Shapiro-Wilk and Levene’s tests, respectively. One-way analysis of variance (ANOVA) was used for analyzing the data, while the means of the treatments were compared with Fisher’s Least Significant Difference (LSD) test at a 5% probability level (P ≤ 0.05). Means followed by the same letter were considered not significantly different. Throughout the manuscript, all ± values refer to the standard deviation (SD) of replicates for each treatment.

3 Results

3.1 Physicochemical properties of feedstocks

The physicochemical properties of the raw rice husk and press mud feedstocks used in the current research study are listed in Table 3 as a baseline point of reference for the potassium recovery, fractionation, and ash results of the respective biochars discussed in subsequent sections.

Table 3. Physicochemical properties of rice husk (RH) and sugarcane press mud (PM) feedstocks.

Values are mean ± SD (n = 3).

Rice husk exhibited slightly higher moisture (8.52%) and substantially higher volatile matter (65.11%) than press mud (7.48% and 57.52%). In contrast, press mud contained greater ash (20.02%) and fixed carbon (14.98%) than rice husk (16.26% and 10.11%). The press mud showed much higher concentration of total potassium (32.10 mg g⁻¹) than that of rice husk (21.42 mg g⁻¹) while silica rich rice husk (105.26 mg g⁻¹) showed higher concentration than the press mud (25.76 mg g⁻¹). This represents the intrinsic physicochemical properties of the unenriched feedstocks and serve as reference for the effects of potassium silicate enrichment on biochar properties, potassium recovery, biochar fractionation and potassium release behavior in the following sections.

3.2 Physicochemical properties of potassium-enriched biochars

The physicochemical properties of rice husk and press mud biochars enriched with potassium silicate through pre- and post-pyrolysis methods are presented in Table 4. Significant differences were observed among the treatments for biochar yield, electrical conductivity, moisture content, ash content, volatile matter content, and fixed carbon content (P ≤ 0.05), whereas pH did not differ significantly among treatments (F(5,18) = 0.81, P = 0.555).

Table 4. Physicochemical properties of rice husk and sugarcane press mud biochars subjected to different potassium silicate enrichment strategies.

Values are means ± standard deviation (SD) of four replicates (n = 4). Means followed by the same letter within a column are not significantly different (P > 0.05, Fisher’s protected LSD test, oneway ANOVA).

3.2.1 Biochar yield

Biochar yield varied significantly among the treatments, ranging from 31.31 to 38.22%. Among the rice husk-derived biochars, RH-Pre recorded the highest yield (38.22%), followed by RH-Post (37.20%) and RH-BC (35.89%). Similarly, within the press mud-derived biochars, PM-Pre exhibited a significantly greater yield (34.82%) than PM-Post (32.95%) and PM-BC (31.31%). Overall, pre-pyrolysis enrichment resulted in higher biochar yields than post-pyrolysis enrichment and the corresponding control treatments. The lowest yield was recorded in PM-BC, whereas the highest yield was observed in RH-Pre.

3.2.2 pH

The pH of the biochars ranged from 9.28 to 10.52, indicating the alkaline nature of all treatments. PM-Pre recorded the numerically highest pH (10.52), followed by RH-Pre (10.44), PM-Post (10.25), and RH-Post (10.14), while RH-BC recorded the numerically lowest pH (9.28), followed by PM-BC (9.99). However, one-way ANOVA indicated no significant differences in pH among treatments (F(5,18) = 0.81, P = 0.555), and all treatment means shared the same LSD grouping (Table 4). Therefore, although potassium silicate enrichment tended to increase pH, this trend was not statistically significant.

3.2.3 Electrical conductivity

Electrical conductivity (EC) differed significantly among the treatments and ranged from 2.18 to 3.65 dS m⁻¹. RH-Pre recorded the highest EC value (3.65 dS m⁻¹), followed by PM-Pre (3.15 dS m⁻¹), whereas the lowest EC was observed in RH-BC (2.18 dS m⁻¹). In both feedstocks, potassium silicate enrichment increased EC relative to the controls, with pre-pyrolysis enrichment generally producing higher values than post-pyrolysis enrichment. The increase likely reflects a greater concentration and retention of soluble ionic species, particularly K-bearing compounds, within the matrix.

3.2.4 Moisture content

Moisture content varied from 2.53 to 4.80% among the treatments. The moisture content was lowest for RH-BC (2.53%) and highest for PM-Post (4.80%). The moisture content of the biochars was generally lower for rice husks compared to press mud. The differences observed suggest that the moisture content of the resulting biochars was affected by both the type of feedstock and the method of enrichment.

3.2.5 Ash content

Ash content varied from 33.82 to 51.40% and was significantly different between the treatments. The highest ash content was observed in PM-Pre (51.40%), followed by PM-Post (47.65%) and PM-BC (42.60%). In contrast, RH-BC recorded the lowest ash content (33.82%). In general, the ash content of the press mud biochars was higher than that of the rice husk biochars. In both feedstocks, the enrichment of potassium silicate resulted in higher ash content than the corresponding controls.

3.2.6 Volatile matter content

Volatile matter content ranged between 19.77 and 25.60%. RH-BC recorded the highest volatile matter content (25.60%), whereas RH-Post exhibited the lowest value (19.77%). In general, the enriched biochars had lower volatile matter contents than the corresponding control treatments. These significant differences indicate variations in the extent of thermal decomposition and carbonization among the biochars.

3.2.7 Fixed carbon content

Fixed carbon content varied significantly among the treatments, ranging from 22.50 to 40.75%. RH-Post recorded the highest fixed carbon content (40.75%), followed by RH-BC (38.06%) and RH-Pre (35.30%). Among the press mud-derived biochars, PM-BC exhibited a higher fixed carbon content (30.98%) than PM-Post (25.51%) and PM-Pre (22.50%). The lowest fixed carbon content was recorded in PM-Pre. Overall, rice husk-derived biochars exhibited greater fixed carbon contents than press mud-derived biochars.

3.3 Fourier transform infrared spectroscopy analysis

The FTIR spectra of rice husk- and press mud-derived biochars produced under different potassium silicate enrichment strategies are presented in Figures 1a, b. All biochars exhibited several characteristic absorption bands associated with silicate minerals and aromatic carbon structures.

In rice husk-derived biochars (RH-BC, RH-Pre, and RH-Post), prominent absorption bands were observed in the range of 1026–1075 cm⁻¹, corresponding to Si–O–Si stretching vibrations, with additional bands between 547 and 565 cm⁻¹ attributed to Si–O bending vibrations. A weak band around 1591–1596 cm⁻¹ was observed in all rice husk biochars and assigned to aromatic C=C stretching vibrations.

Similarly, press mud-derived biochars (PM-BC, PM-Pre, and PM-Post) exhibited absorption bands between 1023 and 1089 cm⁻¹ associated with Si–O–Si stretching vibrations and bands between 441 and 564 cm⁻¹ corresponding to Si–O bending vibrations. Peaks around 788–792 cm⁻¹ were also detected and are commonly associated with quartz- or silica-related vibrations. Additional bands were observed in the enriched biochars, with absorption bands at 1403–1413 cm⁻¹ assigned to carbonate (CO₃²⁻) stretching vibrations. A band around 1630 cm⁻¹ in PM-Post was associated with aromatic C=C stretching and other oxygen-containing surface functionalities. Broad bands centered around 3334–3343 cm⁻¹ were observed in PM-Pre and PM-Post, indicating O–H stretching vibrations related to hydroxyl groups and adsorbed water molecules.

Overall, the FTIR spectra of all treatments were dominated by silicate-associated bands together with aromatic carbon structures. The persistence of the Si–O–Si absorption bands in both control and enriched biochars indicates that silicate-related structures were retained within the biochar matrix, while the enriched treatments exhibited additional carbonate- and hydroxyl-associated functional groups.

3.4 XRD analysis

The X-ray diffraction (XRD) patterns of rice husk- and press mud-derived biochars produced under different potassium silicate enrichment strategies are presented in Figures 2A, B. All treatments exhibited a broad diffraction hump between 20° and 35° (2θ), indicating the predominance of amorphous carbonaceous material formed during pyrolysis.

Diffraction peaks corresponding to quartz (SiO₂; PDF No. 00-033-1161) were observed at approximately 21.98°, 31.42°, and 36.20° (2θ) in all biochars. The rice husk-derived biochars showed more intensity of these reflections, which was consistent with the higher silicon content in the feedstock.

In addition to the weak diffraction features present in the unenriched control biochars, the potassium silicate-enriched biochars showed evidence of additional weak diffraction features at ~26.4–26.8° (2θ) and ~28.1–28.5° (2θ) that were more well defined in the expanded region of the inset in Figure 2. A third, weaker feature was observed near 29.9–30.3° (2θ) in some treatments. Search–match analysis using the ICDD database indicated that the strongest of these reflections (26.4–26.8°) was closest to reference positions for potassium carbonate (K₂CO₃; PDF No. 00-016-0820, 26.188°) and potassium sulfate (K₂SO₄; PDF No. 00-005-0613, 26.314°), whereas the weaker features at 28.1–28.5° and 29.9–30.3° were closer to reference positions for potassium silicate (K₂SiO₃; PDF No. 00-031-1076, 28.13° and 29.86°/30.48°). However, because the reference reflections of these potassium-bearing phases overlap closely in this region, definitive identification of an individual potassium-bearing phase based on XRD alone was not possible. These features were absent or considerably weaker in the unenriched control biochars.

Visual comparison of the diffraction patterns indicated that these potassium-associated reflections were more pronounced in the pre-pyrolysis enriched biochars than in the corresponding post-pyrolysis enriched biochars for both feedstocks. As quantitative phase analysis or peak-area analysis was not performed, this difference is considered a qualitative visual observation rather than a quantitative comparison of crystalline phase abundance.

Overall, XRD patterns showed that potassium silicate enrichment did not significantly alter the mineralogical properties of the biochars, but rather maintained its amorphous carbon structure.

3.5 Operational potassium fractionation in biochars and potassium-enriched biochars

Potassium fractions were significantly different among the control and potassium silicate-amended biochars (Table 5). The total K content was significantly enhanced in both rice husk and press mud biochars after potassium silicate enrichment. Among all treatments, PM-Pre showed the highest total K concentration (85.10 mg g⁻¹), followed by PM-Post (74.24 mg g⁻¹), RH-Pre (65.33 mg g⁻¹), RH-Post (55.81 mg g⁻¹), PM-BC (55.60 mg g⁻¹), and RH-BC (35.48 mg g⁻¹). For both feedstocks, pre-pyrolysis enrichment led to higher accumulation of total potassium than post-pyrolysis enrichment.

Table 5. Chemical fractionation analysis of K in rice husk and sugarcane press mud biochars subjected to different potassium silicate enrichment strategies.

Values are mean ± SD (n = 4). Means followed by the same letter within a column are not significantly different (Fisher’s LSD, p ≤ 0.05). RK was square-root transformed prior to ANOVA to satisfy the normality assumption (Shapiro-Wilk p = 0.082 post-transformation vs. p = 0.046 untransformed); means are presented in original units.

The water-soluble K (H₂O-K) was the highest measured fraction of potassium in all treatments, ranging from 6.56 mg g⁻¹ in RH-BC to 29.00 mg g⁻¹ in RH-Pre. The control biochars were found to have a significantly lower amount of H₂O-K than the enriched biochars. RH-Pre exhibited the highest H₂O-K content (29.00 mg g⁻¹), followed by PM-Pre (26.88 mg g⁻¹), RH-Post (22.60 mg g⁻¹), PM-Post (17.95 mg g⁻¹), PM-BC (7.46 mg g^-1), and RH-BC (6.56 mg g⁻¹).

Expressed as a proportion of total K, water-soluble K accounted for 44.4% and 31.6% of total K in RH-Pre and PM-Pre, respectively, compared with only 18.5% and 13.4% in the corresponding RH-BC and PM-BC controls, indicating that enrichment increased not only the absolute amount but also the relative share of readily available K. A similar trend was observed for NH₄ Ac-extractable K (NH₄ Ac-K). RH-Pre and PM-Pre recorded the highest NH₄. Ac-K concentrations of 25.43 and 24.71 mg g⁻¹, respectively, whereas the control biochars contained comparatively lower values ranging from 4.75 to 5.71 mg g⁻¹.

Similarly, NH₄Ac-extractable K represented a much greater proportion of total K in the enriched biochars (38.9% in RH-Pre and 29.0% in PM-Pre) than in the corresponding controls (13.4% in RH-BC and 10.3% in PM-BC). The HCl-extractable K fraction also increased following enrichment. PM-Pre exhibited the highest HCl-K concentration (4.45 mg g⁻¹), followed by RH-Post (3.98 mg g⁻¹), RH-Pre (3.89 mg g⁻¹), PM-Post (2.35 mg g⁻¹), PM-BC (2.08 mg g⁻¹), and RH-BC (1.87 mg g⁻¹). Residual K represented the smallest potassium pool in all treatments, ranging from 0.60 mg g⁻¹ in RH-BC to 2.10 mg g⁻¹ in PM-Pre.

Potassium recovery also varied significantly among treatments. RH-Pre exhibited the highest K recovery (75.58%), followed by PM-Pre (67.79%), RH-Post (62.86%), RH-BC (59.45%), PM-Post (55.97%), and PM-BC (54.22%). Within each feedstock, pre-pyrolysis enrichment resulted in significantly greater potassium recovery than the corresponding control and post-pyrolysis treatments (F(5,18) = 5.90, P = 0.0021; LSD (0.05) = 9.48). Overall, pre-pyrolysis enrichment enhanced potassium retention more effectively than post-pyrolysis enrichment.

3.6 Potassium release

The potassium release and second-order kinetic model profiles of the control and potassium silicate-enriched biochars are presented in Figures 3, 4A–F, while the corresponding kinetic parameters are shown in Table 6. Potassium release increased continuously with incubation time in all treatments, although the magnitude of release varied considerably among feedstocks and enrichment methods. Potassium silicate enrichment markedly enhanced potassium release compared with the control biochars. The enriched treatments exhibited greater cumulative K release throughout the incubation period, whereas RH-BC and PM-BC showed comparatively lower release. Among the enriched biochars, the pre-pyrolysis treatments generally released more potassium than the post-pyrolysis treatments. Rice husk-derived biochars also exhibited relatively greater potassium release than the corresponding press mud-derived biochars.

Table 6. Kinetic parameters (± SE) and coefficient of determination (R²) of potassium release models, fitted by nonlinear regression (nls).

Models fitted directly in non-linearized form (nls, R v4.2.1). First order: Kt = Q_e(1−e^{^−k₁^t}); Pseudo-second-order: Kt = k²Q_e²t/(1+k2Q_e·t); Elovich: Kt = a + b·ln(t); Parabolic diffusion: Kt = A + B√t. Q_e = equilibrium K release (mg g⁻¹); k1 (h⁻¹), k2 (g mg⁻¹ h⁻¹) = rate constants; a, b (Elovich) = intercept and initial release rate constants; A, B (parabolic) = intercept and intraparticle diffusion rate constant (mg g⁻¹ h⁻½).

The release profiles showed a gradual increase during the early stages of incubation, followed by a more pronounced increase after 24 h, indicating continuous release of potassium from the biochar matrix. The enhanced potassium release observed in the enriched treatments corresponded with their higher abundance of water-soluble and exchangeable potassium fractions.

To further characterize the release behavior, the experimental data were fitted to first-order, second-order, Elovich, and parabolic diffusion models using non-linear least-squares regression, and model performance was compared using R², RMSE, and AIC (Table 7). The parabolic diffusion model provided the best overall statistical description of potassium release for RH-BC, RH-Pre, PM-Pre, RH-Post, and PM-Post, based on the combined R², RMSE, and AIC criteria (Table 7). The better fit of this model is consistent with a substantial contribution of diffusion-controlled processes to potassium release from these biochars during the 96 h incubation period. For PM-BC, the Elovich model provided the best fit, consistent with release from a more heterogeneous distribution of surface binding sites.

Table 7. Goodness-of-fit comparison (R², RMSE, AIC) of kinetic models for potassium release, fitted by nonlinear regression.

R², coefficient of determination; RMSE, root mean square error (mg g⁻¹); AIC, Akaike Information Criterion, calculated as AIC, n·ln(RSS/n) + 2p, where n, number of observations, RSS, residual sum of squares, and p, number of fitted parameters. Lower RMSE and AIC and higher R² indicate better model performance.

The pseudo-second-order model consistently produced imprecise rate-constant estimates, with standard errors comparable to or exceeding the parameter estimates themselves (e.g., k₂ = 0.230 ± 0.167 h⁻¹ for RH-BC, 0.131 ± 0.113 h⁻¹ for RH-Pre, and 0.064 ± 0.040 h⁻¹ for PM-Pre), indicating poor identifiability of this model for the present dataset. Together with its comparatively higher RMSE and AIC values (Table 7), this confirms that the pseudo-second-order model was not the most appropriate description of potassium release under the present experimental conditions.

4 Discussion

4.1 Physicochemical properties of feedstocks

The physicochemical characteristics of the feedstocks reflected their distinct botanical origins and are expected to influence the properties of the resulting biochars. The higher volatile matter content of rice husk compared to press mud suggests its lignocellulosic nature and the high volatile matter reported for rice husk biomass (Bushra and Remya, 2024; Hamidu et al., 2025). The higher ash content of the press mud indicates that it has a higher mineral content, while the higher silicon content in rice husk is due to its inherent mineral composition (Asadi et al., 2021; Sathyabama and Firdous, 2025). The significantly higher total silicon content found in rice husk aligns with the natural silica deposition within rice tissues and its accumulation in the husk during plant development (Nawaz et al., 2023; Hamidu et al., 2025). Rice husk, by contrast, had potassium content that was lower than the press mud, suggesting that it may have had a higher potassium reserve for subsequent mineral transformation during pyrolysis. Previous studies have demonstrated that potassium-bearing minerals facilitate the formation of potassium silicate and other stable K-containing mineral phases, thereby improving potassium retention and influencing nutrient release behavior in enriched biochars (Wu et al., 2021; Chen et al., 2023; Buss et al., 2022). Therefore, the contrasting silicon-rich nature of rice husk and potassium-rich nature of press mud provide an important basis for understanding the subsequent differences in biochar characteristics, potassium recovery, potassium fractionation, and potassium release behavior following potassium silicate enrichment and pyrolysis.

4.2 Physicochemical properties of potassium-enriched biochars

The physicochemical properties of the biochars were significantly influenced by both feedstock type and potassium silicate enrichment strategy (Table 4). The higher biochar yields observed in the enriched treatments, particularly under pre-pyrolysis enrichment, suggest that potassium silicate addition may have enhanced the retention of inorganic constituents during thermal conversion (Buss et al., 2022). Similar increases in biochar yield following mineral enrichment have been reported by Nan et al., 2022 who attributed the effect to increased retention of mineral constituents within the char matrix. The comparatively greater yield observed in rice husk-derived biochars may also be associated with the naturally high silica content of rice husk, which has been reported to contribute to greater residual mass and ash retention following pyrolysis (Hamidu et al., 2025). These observations indicate that feedstock composition plays an important role in determining the response of biomass to nutrient enrichment during thermal conversion (Sornhiran et al., 2022).

Potassium silicate enrichment altered the chemical properties of the biochars by significantly increasing EC, with a numerical but non-significant increase in pH (Nan et al., 2022). The numerically higher pH values observed in enriched treatments are broadly consistent with previous reports indicating that alkaline mineral amendments promote the accumulation of basic ash-forming constituents, including silicates, oxides, and carbonates, during pyrolysis (Tomczyk et al., 2020; Nan et al., 2022), although this trend did not reach statistical significance in the present dataset (F(5,18) = 0.81, P = 0.555). Similarly, the increase in EC indicates a greater concentration of soluble ionic species in the biochar matrix. Comparable increases in pH and EC were observed by Rodrigues et al. (2025) for potassium-rich biochars and were associated with improved nutrient availability. Overall, the significantly higher EC values in the pre-pyrolysis treatments suggest more effective incorporation of K-bearing mineral phases during pyrolysis than in post-pyrolysis treatments (Nan et al., 2022; Buss et al., 2022). From an agronomic perspective, these properties are desirable because such biochars may function both as soil amendments and as nutrient carriers, although the potential effects of elevated EC on salt-sensitive crops should also be considered (Joseph et al., 2021).

There were also significant differences among the proximate composition of the biochars. Press mud-derived biochars, particularly PM-Pre, exhibited greater ash contents than rice husk-derived biochars, reflecting the inherently greater mineral content of press mud and the additional contribution of potassium silicate during enrichment (Nan et al., 2022). Similar increases in ash accumulation following mineral enrichment have been reported in engineered nutrient-rich biochars and are often associated with enhanced nutrient retention within the biochar matrix (Buss et al., 2022). In contrast, rice husk-derived biochars generally exhibited lower volatile matter and higher fixed carbon contents. Reductions in volatile matter accompanied by increases in fixed carbon are commonly interpreted as indicators of a higher degree of carbonization and improved structural stability (Tomczyk et al., 2020; Ippolito et al., 2020). The relatively high fixed carbon content of RH-Post may therefore indicate a more stable carbonaceous structure relative to the press mud-derived treatments (Tomczyk et al., 2020). Overall, the inverse relationship between volatile matter and fixed carbon highlights the combined influence of feedstock type and enrichment strategy on biochar stability (Ippolito et al., 2020; Nan et al., 2022).

Overall, the results demonstrate that potassium silicate enrichment modified both the mineral-associated and carbon-associated characteristics of the biochars (Nan et al., 2022; Buss et al., 2022). Pre-pyrolysis enrichment generally resulted in greater yield, electrical conductivity, and ash accumulation, together with a numerically higher (though not statistically significant) pH, whereas post-pyrolysis enrichment of rice husk biochar was associated with greater fixed carbon content (Nan et al., 2022; Tomczyk et al., 2020). These findings suggest that the timing of potassium silicate incorporation can influence the physicochemical properties of biochars and may consequently affect potassium retention, Fractionation, and release behavior (Buss et al., 2022). Such effects are explored further in the subsequent sections.

4.3 FTIR

The FTIR spectra indicated that potassium silicate enrichment modified the surface chemistry of both rice husk- and press mud-derived biochars, while the fundamental carbonaceous structure formed during pyrolysis was largely retained (Nan et al., 2022; Buss et al., 2022). Biochars prepared using the enriched material were generally similar in their overall spectral profiles, but showed variations in the intensity and distribution of the mineral-associated bands, which indicated the interaction of potassium silicate with the biochar matrix. Similar responses have been reported for mineral-modified biochars, where amendment primarily affects surface-associated functionalities without substantially disrupting the aromatic carbon framework (Gęca et al., 2023; Nan et al., 2022).

The Si–O–Si and Si–O-related bands identified in the enriched treatments suggest that there are silicate-associated structures in the enriched treatments after potassium silicate enrichment (Nan et al., 2022). The aromatic C=C bands found in all the treatments show that the overall aromatic carbon structure remained intact after the application of potassium silicate enrichment (Tomczyk et al., 2020; Nan et al., 2022). Hence, the enrichment of potassium silicate did not seem to significantly affect the carbonized structure of the biochars (Nan et al., 2022; Buss et al., 2022). Furthermore, hydroxyl- and carbonate-associated bands were observed in the enriched treatments, which suggested a change in mineral-associated surface functionalities after enrichment (Buss et al., 2022; Joseph et al., 2021). Such functional groups may contribute to surface reactivity and nutrient interactions in modified biochars (Zhang et al., 2020; Joseph et al., 2021; Ippolito et al., 2020).

The mineral-associated bands were more prominent for the pre-pyrolysis treatments than for the post-pyrolysis enrichment, especially in the Si–O–Si and Si–O regions. This suggests that the presence of potassium silicate during thermal conversion promoted closer association of silicate-derived species with the developing biochar matrix (Buss et al., 2022; Nan et al., 2022). Comparable enhancement of mineral-associated FTIR features following pre-pyrolysis incorporation of mineral amendments has also been reported (Rupngam et al., 2025). Such interactions may contribute to potassium retention through association with mineral-containing surface sites and the developing carbon matrix, thereby influencing potassium distribution and subsequent release. In contrast, post-pyrolysis enrichment occurs after formation of the carbon matrix and is therefore more likely to result in surface-associated potassium and silicate species. Thus, the stronger mineral-associated FTIR features observed following pre-pyrolysis enrichment suggest that the timing of potassium silicate addition influenced its interaction with the developing biochar matrix and consequently the chemical environment associated with potassium retention and release.

Overall, the FTIR findings indicate that potassium silicate enrichment primarily influenced the mineral-associated surface chemistry of the biochars while largely preserving their aromatic carbon framework. The stronger expression of silicate-associated features following pre-pyrolysis enrichment suggests that introducing potassium silicate before thermal conversion promoted closer interaction between the amendment and the developing biochar matrix. These changes in surface chemistry may contribute to the altered potassium association and mobility observed in the enriched biochars, particularly when considered together with the potassium fractionation and release results.

4.4 XRD

The persistence of the broad amorphous diffraction feature across the treatments indicates that potassium silicate enrichment did not substantially disrupt the carbonaceous structure developed during pyrolysis. This suggests that the enrichment primarily influenced the mineral-associated environment of the biochars rather than causing a major alteration of the underlying carbon framework, which is consistent with the structural characteristics commonly reported for lignocellulosic biochars (Zhou et al., 2025).

The stronger quartz reflections observed for rice husk-derived biochars are consistent with the greater native silicon content of rice husk and the persistence of biogenic silica during pyrolysis. Rice husk contains silica that is relatively resistant to thermal transformation and can remain within the resulting biochar as a crystalline mineral component (Ouedraogo et al., 2026; Hamidu et al., 2025). Therefore, the mineralogical differences between the two feedstocks constitute a background for the following association of potassium introduced through enrichment.

The emergence of additional potassium-associated crystalline features following potassium silicate enrichment indicates that the applied potassium did not remain exclusively in its original chemical environment during pyrolysis. Instead, thermal conversion appears to have promoted interactions between potassium and the inorganic constituents of the biomass matrix, resulting in the development of potassium-associated mineral phases. Potassium introduced into biomass can undergo redistribution and transformation during pyrolysis through reactions with inorganic constituents and other mineral-forming species present in the feedstock (Zhang et al., 2020; Tan et al., 2026; Xia et al., 2024). The close overlap among the reference reflections of potential potassium-bearing phases in the present XRD analysis means that these features are best interpreted as evidence of potassium mineral formation or association rather than definitive identification of a single crystalline phase.

The enrichment strategy also appears to influence the extent of potassium incorporation into the mineral fraction. The more pronounced potassium-associated features observed for pre-pyrolysis enrichment suggest that introducing potassium silicate before thermal conversion provides greater opportunity for potassium to interact with the inorganic constituents released during biomass decomposition and to become incorporated into the developing biochar–mineral matrix. In contrast, post-pyrolysis enrichment occurs after the carbon and mineral framework has already developed, which may limit potassium interactions primarily to existing surface or near-surface sites. Similar effects of enrichment timing on the incorporation and distribution of mineral constituents have been reported for mineral-enriched biochars (Buss et al., 2022; Rupngam et al., 2025). The greater prominence of the potassium-associated features in the pre-pyrolysis treatments should, however, be regarded as a qualitative indication because quantitative phase or peak-area analysis was not performed.

These mineralogical interactions have implications for the subsequent retention and mobility of potassium. Potassium associated with newly formed mineral phases can differ in its accessibility from potassium present in readily soluble or exchangeable forms. Thus, the alteration of the mineral environment during potassium silicate enrichment may contribute to the differences in potassium fractionation observed among the treatments. In particular, the greater incorporation of potassium during pre-pyrolysis may provide additional sites or mineral associations capable of retaining potassium within the biochar matrix while still allowing a proportion of the nutrient to remain available for subsequent mobilization.

The XRD observations therefore provide a mechanistic basis for understanding the differences in potassium release observed among the enrichment strategies. Potassium associated with more readily accessible pools can be mobilized during contact with water, whereas potassium incorporated more strongly into mineral-associated environments may require dissolution or diffusion before becoming available. This is consistent with the diffusion-controlled release behavior observed for most treatments, suggesting that the mineral environment created during enrichment may contribute to controlling the subsequent movement of potassium from the biochar matrix into solution.

Overall, the XRD findings indicate that potassium silicate enrichment altered the mineral-associated environment of potassium without substantially disrupting the carbonaceous structure of the biochars. The stronger mineral association apparent following pre-pyrolysis enrichment suggests that the timing of potassium addition is an important factor determining potassium incorporation during thermal conversion. These mineralogical changes, together with the potassium fractionation and release results, indicate that enrichment strategy can influence both the retention and subsequent mobility of potassium in the resulting biochars.

4.5 Operational potassium fractionation in biochars and potassium-enriched biochars

Potassium silicate enrichment substantially altered the distribution of potassium among the different operational fractions of the biochars. Water-soluble K (H₂O-K) represented the dominant potassium pool in all treatments, followed by NH₄Ac-K, whereas HCl-K and residual K formed comparatively smaller fractions. The predominance of H₂O-K and NH₄Ac-K indicates that a considerable proportion of potassium was retained in readily available forms after pyrolysis (Chen et al., 2023). Similar patterns have been reported for potassium-enriched biochars, in which soluble and exchangeable fraction constitute the major K pools governing nutrient availability after soil application (Chen et al., 2023; Bilias et al., 2023).

The marked increase in H₂O-K and NH₄Ac-K following potassium silicate enrichment suggests that a large proportion of the added potassium remained in bioavailable forms after pyrolysis (Leite et al., 2024; Xiu et al., 2023). In particular, RH-Pre and PM-Pre exhibited the highest concentrations of both fractions, indicating that pre-pyrolysis enrichment favored potassium retention during thermal conversion (Chen et al., 2023; Sirijuncheun et al., 2023). This effect may be attributed to stronger interactions between potassium silicate and biomass components during pyrolysis, which promoted the incorporation of potassium into soluble and exchangeable pools (Xia et al., 2021; Chen et al., 2021). Similar increases in bioavailable potassium fractions have been reported following co-pyrolysis of biomass with potassium-containing additives (Rajput et al., 2024; Wu et al., 2023).

Variability in potassium distribution was also evident between feedstocks. Rice husk biochars, especially the RH-Pre, contained higher H₂O-K and NH₄Ac-K fractions, while the press mud biochars showed relatively higher HCl-K and residual K fraction contents (Leite et al., 2024). This indicates that rice husk favored retention of more easily available potassium forms, while press mud promoted the formation of relatively stable potassium pools (Nguyen et al., 2020). The higher ash and mineral content of press mud may have enhanced the association of potassium with inorganic constituents during pyrolysis, resulting in greater stabilization of potassium within the biochar matrix (Xia et al., 2021).

Although HCl-K and residual K represented smaller proportions of total potassium, both fractions increased significantly after enrichment, particularly in PM-Pre (Leite et al., 2024). These fractions are generally associated with mineral-bound or structurally retained potassium forms and are less susceptible to immediate dissolution than H₂O-K and NH₄Ac-K (Xiu et al., 2023). The increase in these fractions, therefore, suggests the formation of more stable potassium pools that may contribute to sustained potassium release following soil application (Chen et al., 2023). Similar increases in stable potassium fractions have been reported in mineral-enriched biochars where potassium became associated with inorganic phases formed during thermal conversion (Rajput et al., 2024).

The higher potassium recovery observed in RH-Pre and PM-Pre further indicates that pre-pyrolysis enrichment improved potassium retention efficiency compared with post-pyrolysis enrichment (Leite et al., 2024). This difference is most plausibly explained by the timing of potassium introduction relative to the development of the carbon and mineral matrix during pyrolysis. When potassium silicate is introduced before pyrolysis, K and silicate species are present throughout the biomass during thermal conversion, providing greater opportunity for their interaction with the developing carbonaceous and mineral phases. Such interactions may promote greater incorporation of K into the developing matrix and its association with relatively stable mineral-bound and exchangeable pools. Similar deeper mineral integration during pre-pyrolysis treatment and greater surface-associated interactions following post-pyrolysis treatment have been reported for mineral-enriched biochars (Rupngam et al., 2025). Studies of silica-rich biomass have also shown that pyrolysis can redistribute K into different forms, including silica-associated pools, indicating that thermal conversion can influence K retention and subsequent release behavior (Nguyen et al., 2020). Mineral-assisted pyrolysis has further been reported to promote transformation of readily soluble K into exchangeable and K-silicate-associated forms, thereby modifying K retention and release (Chen et al., 2023; Wu et al., 2021). In contrast, post-pyrolysis enrichment occurs after formation of the biochar matrix, limiting the opportunity for incorporation during thermal conversion and potentially leaving a greater proportion of the added K associated with accessible surface or near-surface sites (Buss et al., 2022; Rupngam et al., 2025). However, because the post-pyrolysis treatments contained a lower total K pool, their cumulative K release remained lower than that of the pre-pyrolysis treatments despite the potentially greater accessibility of surface-associated K. Thus, the higher K retention and recovery observed in the pre-pyrolysis treatments may reflect greater incorporation and stabilization of K during thermal conversion, while the subsequent release represents mobilization of the larger retained K pool.

In general, the fractionation results show that the potassium silicate enrichment increased both readily available (H₂O-K and NH₄Ac-K) and relatively stable (HCl-K and residual K) potassium fractions in biochars (Leite et al., 2024). The coexistence of these pools suggests that the enriched biochars may provide an initial supply of readily available potassium while maintaining a reserve of more slowly released potassium, thereby supporting both immediate and sustained nutrient availability after soil application.

4.6 Potassium release

The potassium silicate enrichment had a significant influence on the release behavior of K from the biochars. The enriched biochars showed significantly greater potassium release than the control biochars during the incubation period, indicating that the addition of potassium silicate led to greater releasable potassium in the biochar matrix. Increases in nutrient release after mineral enrichment have been reported for potassium-modified biochars and biochar-based fertilizers (Rajput et al., 2024).

In terms of the enrichment strategies, the potassium release from pre-pyrolysis enrichment was higher than that obtained from post-pyrolysis enrichment for both feedstocks (Leite et al., 2024). The trend indicated that pre-pyrolysis application of potassium silicate facilitated the retention and preservation of potassium in readily releasable forms during thermal conversion (Xia et al., 2021). There have been similar reports on nutrient-enriched biochars produced through co-pyrolysis with mineral additives, with enrichment during pyrolysis leading to enhanced nutrient retention and subsequent release (Nguyen et al., 2020).

There were also significant feedstock differences in K release, with rice husk biochars showing higher release compared to the press mud biochars (Leite et al., 2024). These feedstock-dependent differences are consistent with the contrasting proximate and mineral composition of the two feedstocks (Table 3). The higher volatile matter content of rice husk (65.11%) relative to press mud (57.52%) suggests more extensive devolatilization during pyrolysis, likely generating a more open, accessible pore structure that facilitates diffusion of readily soluble potassium to the biochar surface. In addition, the markedly higher native silicon content of rice husk (105.26 mg g⁻¹) relative to press mud (25.76 mg g⁻¹) may favor the formation of a more porous, silica-associated matrix; comparably silica-rich biochars have been reported to exhibit relatively unrestricted potassium release kinetics (Nguyen et al., 2020). In contrast, the higher ash content of press mud (20.02%; Table 3) is consistent with a denser inorganic matrix in which potassium may become more strongly associated with mineral surfaces and structurally bound within silicate- and carbonate-associated phases, thereby restricting its mobility. This is consistent with the higher HCl-K and residual K fractions observed for press mud biochars relative to rice husk biochars.

The kinetic modelling results indicated that potassium release from most biochars (RH-BC, RH-Pre, PM-Pre, RH-Post, and PM-Post) was best described by the parabolic diffusion model, indicating that release was governed primarily by diffusion of potassium from within the biochar matrix rather than by simple, rapid dissolution of surface-associated potassium alone. Similar diffusion-controlled release behavior has been reported for nutrient-enriched biochars and slow-release fertilizer materials (Xia et al., 2021; Rajput et al., 2024).

The dominance of the parabolic diffusion model across most treatments suggests that, after an initial release of readily available surface-associated potassium, the movement of potassium from internal pores and mineral-associated sites within the biochar structure increasingly controlled the release pattern over the 96 h incubation period (Xia et al., 2021). Similar diffusion-controlled release mechanisms have been reported in porous biochar-based nutrient carriers and controlled-release fertilizer systems (Leite et al., 2024). For PM-BC, potassium release was instead best described by the Elovich model, which is typically associated with release from energetically heterogeneous surfaces, suggesting a comparatively more heterogeneous distribution of potassium-binding sites in this unenriched press mud biochar.

The relatively high fitting coefficients obtained for the parabolic diffusion model are consistent with, though not direct proof of, a diffusion-influenced release process (Xiu et al., 2023). A plausible interpretation is that, after the initial release of readily available potassium fractions, diffusion of potassium from less readily available pools and mineral-associated sites within the biochar contributed to the subsequent release pattern (Xia et al., 2021), as has been proposed for other porous biochar-based nutrient carriers and controlled-release fertilizer systems (Leite et al., 2024). Therefore, these mechanisms should be regarded as plausible rather than definitive, as pore structure and morphology were not directly characterized in the present study. Confirmation of these mechanisms would require complementary characterization techniques, such as BET surface area analysis and SEM imaging.

The release behavior observed in this study agrees with the potassium fractionation results, which showed that readily exchangeable potassium fractions were found to be associated with relatively stable pools of potassium (Nguyen et al., 2020). The highly soluble and exchangeable potassium fractions were responsible for the fast release, while the more stable potassium compounds were responsible for the release in later incubations (Chen et al., 2023). Such a combination of potassium pools is advantageous because it can provide both immediate nutrient availability and sustained potassium supply following soil application (Xiu et al., 2023).

Overall, potassium silicate enrichment enhanced potassium release from both rice husk- and press mud-derived biochars, with pre-pyrolysis enrichment showing the strongest effect. The kinetic modelling results indicate that potassium release was influenced by both dissolution and diffusion processes, highlighting the potential of these enriched biochars as potassium sources capable of providing both rapid and sustained nutrient release in agricultural soils.

5 Conclusion

The present study shows that the abundant agro-industrial wastes, such as rice husk and sugarcane press mud, can be effectively converted into potassium-enriched biochars using potassium silicate enrichment. The enrichment strategy significantly influenced biochar properties, potassium fraction distribution, and the potassium release behavior, with pre-pyrolysis treatments generally showing greater potassium retention and more favorable release characteristics than post-pyrolysis treatments. The enriched biochars contained both readily available and relatively stable potassium fractions, indicating their potential to provide immediate as well as sustained nutrient supply.

The findings of potassium incorporation within the biochar matrix and the progressive potassium release during incubation demonstrate their suitability as alternative potassium sources in agriculture. It should be noted that these findings are based on a single pyrolysis temperature (450 °C) and a single potassium silicate enrichment rate (5% w/w); the pre- versus post-pyrolysis comparisons reported here should therefore be interpreted as indicative of trends under these specific conditions, and further work across a wider range of pyrolysis temperatures and enrichment rates is needed before these conclusions can be generalized. Similarly, potassium release was evaluated under a standardized deionized-water system, and release under field soil conditions may differ given the influence of soil pH, ionic strength, competing cations, and mineral surfaces on potassium mobility. By converting agricultural wastes into value-added nutrient carriers, this approach contributes to resource recovery, waste valorization, and improved nutrient-use efficiency. These results indicate that K-enriched biochar produced from rice husk and sugarcane press mud might be an alternative source of potassium for sustainable nutrient management in agriculture.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Fotis Bilias, Aristotle University of Thessaloniki, Greece

Reviewed by: Anjana J. Atapattu, Coconut Research Institute, Sri Lanka

Moh Moh Thant Zin, University of Missouri, United States