Designed residue-based amendments tailored to soil-specific requirements for wheat within an integrated nutrient management framework

1 Introduction

Modern agriculture remains highly dependent on inorganic fertilizers whose manufacture relies heavily on fossil fuels and non-renewable raw materials (1–3). This is particularly evident for nitrogen fertilizers, since ammonia synthesis through the Haber–Bosch process requires a large amount of natural gas and energy. Consequently, inorganic fertilizer production is intrinsically linked to energy markets, greenhouse gas emissions, and the volatility of fossil fuel prices. In the case of phosphorus fertilizers, the limitation is different but equally strategic, as phosphate rock is a finite and geographically concentrated resource (4, 5), whose extraction and processing are associated with relevant environmental impacts and long-term supply risks. At the same time, this phosphorus source is highly concentrated at a global scale, with Morocco and Western Sahara accounting for the largest share of world phosphate rock reserves, while only a few countries control most economically exploitable reserves (5). This situation has become especially critical in the European Union (EU), where the fertilizer sector has been strongly exposed to fluctuations in natural gas prices and to external supply chains, as highlighted during the energy and fertilizer crisis following the Russian invasion of Ukraine (6, 7). More recently, disruptions affecting strategic trade routes such as the Strait of Hormuz have further illustrated the vulnerability of global fertilizer markets (8, 9). Europe, and especially Spain, are not isolated from this scenario. The Spanish primary sector represents an important component of the total national gross value added and supports highly relevant agri-food value chains (10, 11). The dependence on third countries for essential fertilizer inputs represents a strategic risk for agricultural productivity, food security, and the economic resilience of Mediterranean farming systems. In this context, the transition toward a circular economy is not only an environmental objective but also a strategic necessity (6). The recovery of nutrients and functional materials from wastes and by-products can reduce the pressure on non-renewable resources, diversify fertilizer supply chains, and create locally available alternatives adapted to specific cropping systems (12, 13). However, their nutrient content, release dynamics, potential contaminants, interactions with soil properties, and effects on crop performance must be systematically characterized under controlled and field-relevant conditions to maintain healthy soils and restore degraded soils, while sustaining economically viable productivity. Therefore, knowledge of new residue-based fertilizers and amendments is essential to determine their real contribution to sustainable nutrient management, complying with current regulatory frameworks and environmental protection criteria (14). The European Fertilizing Products Regulation has opened a pathway for the incorporation of recycled, organic, and waste-derived materials into the fertilizer market. This should not be understood as a direct replacement of conventional inorganic fertilizers in all contexts, but as their strategic integration within integrated nutrient management (INM) (14, 15). In line with this approach, fertilization strategies should combine organic, recycled, and inorganic nutrient sources to satisfy crop demands, improve nutrient use efficiency, and reduce nutrient losses through emissions or leaching (16–18).

Beyond their direct use as fertilizers or soil amendments, waste and by-products can also be valorized as functional materials for the mitigation of gaseous emissions from livestock and manure management systems (19). Technologies capable of capturing gaseous emissions can provide a dual environmental benefit by reducing atmospheric pollution while retaining nutrients that may subsequently be recycled and used in agriculture (20–22). In this context, filtration systems based on low-cost waste-derived materials that can subsequently be reused as soil amendments represent a promising circular strategy linking emission mitigation with nutrient recovery. In particular, multi-waste filters have recently been proposed as a strategy for nitrogen gas capture from slurry and successive reuse as a soil amendment, advancing the technology and knowledge of new materials (20, 21). These findings provide the basis for developing integrated systems in which gaseous nitrogen losses from manure are captured, stabilized, and reintroduced into the soil–plant system with the application of these filters as fertilizers or amendments to supply nutrients and restore soil health.

This study addresses a knowledge gap in research on waste-derived fertilizers by developing and evaluating amendments tailored to plant- and soil-specific nutrient requirements within an INM framework, while also valorizing materials previously used as gas-capturing filters. A key aspect is that waste-derived fertilizers and amendments can be specifically formulated and combined according to specific soil properties and crop requirements, and then complemented with mineral fertilizers if necessary to optimize nutrient supply and use efficiency. We hypothesize that (1) the formulation of residue-based amendments (including materials previously used as filters) should be adapted to different soil types; (2) residue-based amendments can complement inorganic fertilization by promoting soil functions that cannot be provided by inorganic fertilizers individually, and they may generate synergistic effects in plant performance; and finally, (3) nitrogen (N) captured from gaseous emissions from slurry, through pre-sowing incubation, can enhance nutrient bioavailability as effectively as inorganic fertilizers.

2 Material and methods

2.1 Experimental design

Four filters formulated from wastes and by-products (hereafter residues) were reused as residue-based amendments (RBA) in four soils after their previous use to capture slurry gases such as NO₂ and NH₃, which enriched them in N and other nutrients (20). In each soil, four treatments were established following a completely randomized design: (1) control without fertilization (Control); (2) residue-based amendment (RBA); (3) inorganic fertilizer treatment (IF)—supplying the same NPK amount as the RBA—; and (4) a mix of RBA and IF (RBA+IF; based on the same amount of NPK supplied with each one). Each experimental unit consisted of a pot containing 270 g of soil, and four replicates per treatment and soil combination were used (4 pots × 270 g pot^-1 × 4 treatments = 4,320 g of each of the four soils). In our previous studies, this amount of soil together with the nutrients supplied was sufficient for wheat plants to reach maturity and compare results across treatments (23, 24). The experiment was conducted at the Campus of Rabanales (37°54’49.0”N, 4°43’05.5”W), University of Córdoba, Spain. This was conducted under ambient external environmental conditions between December and May (2024–2025) in a shade house.

In all soils, bread wheat (Triticum aestivum L. var. Átomo) was cultivated as a model crop to evaluate the potential agronomic benefits of valorizing formulated residues.

2.2 Residue characterization

The 14 residues used in this study for designing the RBAs were collected from nine companies located in the south of the Iberian Peninsula (Spain and Portugal) and defined in Supplementary Table S1. The residues were ground before being used to homogenize the size of the particles. In addition, they were characterized after sieving through a sieve <2 mm. This fraction was analyzed to determine the texture through a vibrating sieve tower (ø mm: 2–0.5, 0.5–0.2, 0.2–0.05, and <0.05); specific surface area (SSA) was determined as described by Tuller and Or (25), and their density (ρ), pH (1:2.5, m:V; 26), and electrical conductivity (EC, 1:5 m:V (27) were also measured. Total calcium carbonate (CaCO₃) was analyzed by weight loss following treatment with 6 mol L^-1 HCl (28). Cation exchange capacity (CEC) and both exchangeable acid and non-acid cations were extracted with 1 M NH₄OAc at pH 7 and quantified via Flame Atomic Absorption Spectroscopy (F-AAS, Perkin Elmer). Total C, N, and S (C_T, N_T, and S_T) were analyzed using a LECO CN628 analyzer (LECO Instruments S.L.) following the method of LECO Corporation (1999) (29), while organic C (C_o) was quantified according to Walkley and Black (30). Nitrate (NO₃^-) and ammonium (NH₄⁺) concentrations were quantified spectrophotometrically using a BioTek PowerWave XS microplate reader (Agilent Technologies Spain, S.L.), applying the methodologies of Miranda et al. (31) and Mulvaney (32), respectively. Available P was determined following Olsen et al. (1954) (33). The available fractions of Fe, Cu, Mn, and Zn were extracted using the diethylenetriaminepentaacetic (DTPA) method (34) and also measured using F-AAS (Fe_DTPA, Cu_DTPA, Mn_DTPA, Zn_DTPA). The water-soluble fraction of chemical elements (Ag, Al, As, Ba, Ca, Cd, Co, Cr, Cu, Fe, Hg, K, Mg, Mn, Mo, Na, Ni, P, Pb, S, Sb, Se, Si, Sn, Sr, Ti, Tl, V, and Zn) was extracted by mixing the residue with deionized water (1:20 m:V) for 5 days, following Buján et al. (35), and analyzed using Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES, Agilent Technologies Spain, S.L.). The total concentrations of these elements were determined via acid digestion with concentrated HNO₃ (65%), HCl (37%), and H₂O₂ (30%) (36), followed by ICP-OES analysis (Supplementary Table S2). It should be noted that properties were analyzed according to the nature of the residue; therefore, some parameters were not determined because they were not relevant to the specific residue.

2.3 Soil characterization

The soils used in this study were obtained from the MINFOCINC soil collection of the AGR-165 research group of the University of Córdoba (Table 1). The analytical procedures used to determine their physicochemical properties were the same as those described in the residue-based materials section. In addition, active lime (i.e., carbonate reactive to 0.1 M ammonium oxalate, hereafter referred to as ACCE) was determined following the method described by Loeppert and Suarez (37). Mineralogy of the clay fraction was also characterized by X-ray diffraction (XRD). Finally, soil porosity was determined prior to the experiment to estimate the water-holding capacity of the soils and to define irrigation requirements. A known amount of soil was placed in a pot and weighed. The soil was then saturated with water by capillary rise for 24 h. After removing the pot from the water bath and allowing drainage for 30 min, the weight was recorded. Soil porosity was calculated as:

Table 1. Physicochemical characterization of the four soils used in this study.

EC, Electrical conductivity (1:5 extract).

C_o, Organic carbon.

CCE, calcium carbonate equivalent.

Active calcium carbonate equivalent.

CECe, cation exchange capacity.

Fed, Feca, and Feox, Iron extracted with dithionite, citrate-ascorbate, and oxalate, respectively.

DTPA, Diethylenetriaminepentaacetic.

P_Olsen.

within the clay fraction.

Porosity(%)=Wsat−WdryWdry∗100

assuming a water density of 1 g mL^-1. Based on this estimation, irrigation was adjusted to maintain approximately 70% of the soil pore volume filled with water throughout the experiment.

2.4 Soil-tailored formulation strategy

After evaluating the soil properties of the four soils (Table 1), each RBA was specifically formulated to alleviate the main physicochemical constraints previously identified in its target soil—combined with their functions to capture gases from slurry. The identified constraints as well as the desired soil improvements are shown in Table 2 for each soil. It should be noted that data such as the critical soil levels for nutrients such as K and P are not shown here, although soil variables such as P Olsen and exchangeable K, as indicators of nutrient availability for plants are shown in Table 1. Consequently, each amendment differed in composition and functional properties, reflecting the specific requirements of its target soil rather than a standardized fertilization approach. This function-oriented design constituted the basis for evaluating the agronomic effectiveness of residue-based amendments under contrasting soil conditions. For example, an appropriate ash application rate can increase soil pH, thereby increasing P availability and decreasing Mn availability in Soil 1, while also supplying available K.

Table 2. Target soil functional and properties to address through residue formulation.

2.5 Formulation and characterization of RBAs

The formulation of the different RBAs used in this study was designed to meet two objectives: (1) to capture gases from slurry and (2) to enhance soil properties under the regulatory perspective of EU Regulation 2019/1009. First, RBAs were used as multi-layer filters [see 38, for more information], which were enriched in C, N, and S and then re-used as amendments (studied in this work). The design of the RBAs could potentially fall within the categories of soil improvers, organo-mineral fertilizing products, or even inorganic amendments depending on their declared nutrient function. They are composed of a combination of industrial, agro-industrial, and forestry residues selected for their complementary physicochemical properties, providing sufficient compositional flexibility to fulfill both functions described above. Therefore, the amendment properties were estimated with weighted means from residue properties and are shown in Table 3. Each RBA (1, 2, 3, and 4) was formulated to fulfill the main requirements of each specific soil (PEDZ1, FTPZ2, MRSZ2, OTVZ2, respectively). Their formulation is described in Supplementary Table S1.

Table 3. Physicochemical properties of the residue-based amendments (RBA) formulated to capture slurry gases and fertilize the target soils.

*The sum of particles could not sum 100 due to the presence of waste NOS.

EC, Electrical conductivity (1:5 extract).

CCE, calcium carbonate equivalent.

CECe, cation exchange capacity.

C_T,N_T, S_T, Total carbon, nitrogen, and sulfur by LECO.

C_o, Organic carbon; P_Olsen, phosphorus extracted with bicarbonate; DTPA, Diethylenetriaminepentaacetic.

2.6 Soil fertilization

Before sowing, soils and fertilizers (IFs and RBAs) were mixed. Approximately 5 kg of each soil was homogenized and equally divided into four different aluminum trays to be mixed with the corresponding fertilizer (Control; RBA; IF; RBA + IF). The mixtures were thoroughly homogenized to guarantee uniform distribution of the amendments. When large particles were present in the amendments, they were manually crushed or ground to improve homogeneity. After mixing, the soils were placed into cylindrical PVC pots 5 cm in diameter and 15 cm tall with a 6 mm drainage hole at the bottom, which was fitted with a cellulose acetate filter. A small portion of the prepared soil mixture was reserved to cover the seeds during sowing. The soil–fertilizer mixtures were incubated for two weeks before sowing, watering them to field capacity every 3 days. This allowed initial interactions between soil particles, Ifs, and RBAs, as well as promoting the onset of organic matter mineralization and microbial activity. The IF treatment was designed to provide the same NPK amount supplied by RBAs in each soil. Therefore, NPK rates varied depending on the soil–amendment combination (Table 4), which were calculated to apply ~33 t amendment ha^-1 and different doses of N with each RBA (RBA 1–4: 84, 345, 581, and 213 kg ha^-1, respectively). Mineral fertilization was supplied using a combination of urea (46% of N), single superphosphate (Super18; 18% P₂O₅ or ~7.85% P), and potassium chloride (KCl; 82.9% K₂O or ~52%K). The IFs were gently ground prior to mixing to improve homogenization and ensure a uniform distribution within the soil matrix because the required quantities were small.

Table 4. Fertilization doses of RBA and inorganic fertilizers (Urea, Super18 and KCl) applied for each treatment (group of four replicates, 1.08 kg) and their equivalent per hectare (18 cm deep).

2.7 Crop management and performance

Four seeds of bread wheat (T. aestivum L. var. Átomo) were sown in each pot to ensure successful germination, and after emergence, seedlings were thinned, leaving one plant per pot (experimental unit). The crop was watered, maintaining soil moisture at approximately 70% pure water saturation. The irrigation frequency was 2–3 days during winter months and 1–2 days during spring months.

To assess the agronomic performance of the crop, the number of tillers, leaf chlorophyll content (estimated in SPAD values with a Konica Minolta, Inc., Osaka, Japan), total biomass, yield, number of grains, and nutrient concentration (grains and straw) were evaluated. Besides, additional parameters were calculated, including nutrient uptake (nutrient concentration × biomass), nitrogen use efficiency (NUE, overall efficiency with which the crop uses applied N to produce yield), recovery efficiency (RE, the proportion of applied fertilizer N recovered by the crop), and physiological efficiency (PE, the efficiency with which additional N uptake is converted into biomass or yield) (39, 40). To assess wheat grain quality, grain N concentration was evaluated because it is related to protein content. N concentrations >2.28% were considered indicative of higher grain quality (41).

2.8 Limitations for human and animal consumption

The applicability of these new fertilizers must comply with Regulation (EU) 2019/1009 and the applicable legal requirements for soil health. Additionally, the concentration of potential toxic elements in grain—i.e., Pb, Cd, and As—was compared to EU Regulation 2023/915 of pollutants in food and, for Ni, Commission Regulation (EU) 2024/1987, to assure food/feed safety. The concentration of these elements in the straw was also compared against reference values of Directive 2002/32/EC of the European Parliament, which regulates the relative chemical elements in animal food, because straw is sometimes used to feed animals. The elemental concentrations determined on a dry matter basis were multiplied by 0.88 to express them at the regulatory grain moisture content of 12%. This study omits comparison with the European Food Safety Authority (EFSA) for elements such as Al, Cu, Fe, Mn, Mo, Se, Zn, and Cr because their assessment requires consideration of estimated daily intake.

2.9 Statistical analysis

Statistical analysis of the measured variables was performed using R software and STATISTIX v.10.0 (42, 43). A two-way factorial analysis of variance (ANOVA) was done considering Soil (four levels) and Treatment (Control, RBA, IF, and RBA + IF) as fixed factors, including the Soil × Treatment interaction. When a significant Soil × Treatment interaction was found, treatment effects were subsequently evaluated separately within each soil. In the absence of significant interaction, the main effects were evaluated across soils. SPAD measurements, recorded repeatedly on the same plants throughout crop development, were analyzed using a repeated-measures ANOVA, with Time as the within-subject factor and Soil and Treatment as between-subject factors, including their interactions. Differences among means were assessed using the post-hoc Least Significant Differences (LSD) test at a significance level of p < 0.05. In addition, differences among soils under unfertilized conditions were evaluated by comparing the Control treatments of the four soils through one-way ANOVA. When data did not fulfill residual normality and/or homogeneity of variances, the non-parametric test Kruskal–Wallis was performed.

3 Results

3.1 Soil fertility on wheat performance

Total biomass was significantly higher in Soils 1 and 4, which also showed greater straw biomass and tiller production (Figure 1). No differences were observed for either Harvest Index or Weight of Grains (Supplementary Figure S1).

Plant uptake of N, K, and Ca was higher in Soil 4, followed by Soil 1, than in the rest of the soils. The lowest uptake of P was found for plants grown in Soil 1 (Supplementary Figure S2), while these plants also had the highest uptake of Na and Fe, with Mn uptake being ~100 times higher than in wheat grown in the other soils (Supplementary Figure S2). The uptake of potential toxic elements such as Al, As, B, Pb, and Se followed biomass production. Element uptake of Ba, Cd, Co, Cr, Mo, and Ni were seemingly conditioned by soil pH and its geochemical behavior, as these elements were higher in Soil 1 (with acid pH), except Mo, which was higher in Soil 2 with a very alkaline pH.

3.2 Effect of treatments on wheat yield performance

Treatment effects on crop performance differed among soils, as indicated by significant Soil × Treatment interactions for all variables except HI and grain weight. Treatments containing RBA showed total biomass similar to the IF treatment, but no statistical differences were observed except in Soil 2, where IF was significantly higher than the other treatments (Figure 2). HI under RBA + IF treatment was significantly higher than under IF.

The number of tillers was significantly higher with IF treatment in Soils 2 and 4 (Figure 3). In contrast, grain weight and HI (parameters that did not show a significant Soil × Treatment interaction) were significantly higher with RBA + IF than with IF alone.

3.3 Effect of treatments on wheat nutrition

Wheat nutrition performance showed a significant Soil × Treatment interaction for all variables except for K and S uptake. These elements showed a significant increase with RBA, IF, and RBA + IF in comparison with the Control but not between them. N, Ca, Fe, Cu, Mn, and Zn uptake followed a similar pattern to that described for wheat biomass in the different soils, with micronutrients uptake being significantly higher when wheat was grown in Soil 2 supplied with IF (data not shown). Plants grown in Soil 1 showed higher Mn uptake, whereas RBA significantly reduced it (Figure 4). Mg, Na, and Fe uptake was also higher under RBA in Soil 1 (Supplementary Figure S3). Figure 4 shows the concentration of Zn in grain (absence of a Soil × Treatment interaction), where a significant decrease under IF treatment occurred for all soils together.

3.4 Concentrations of potential toxic elements in wheat

No significant treatment differences in grain element concentrations were observed except for Ba, which was significantly higher under IF2 than RBA2 (Supplementary Figure S4). On the other hand, chemical element concentrations in straw showed more differences across treatments. IF1 produced a significantly higher Mn concentration than IF2 (Figure 4), as well as a significantly higher Ni concentration (data not shown). The latter, however, was significantly higher with RBA1 than IF1. The concentration of Cr was significantly higher with IF2 than RBA2, and Al concentration with IF3. Finally, all RBAs showed a significantly higher Na concentration (except in Soil 3) than IFs.

3.5 Plant physiology and nitrogen efficiency

SPAD values, used as a proxy for leaf chlorophyll content, showed no significant Soil × Treatment interaction, and the lowest values were found in Control plants while no differences were observed between fertilizer treatments over the monitoring time. In Soil 1, recovery efficiency was significantly higher with RBA than with the RBA + IF treatment, while in Soil 2, IF showed significant differences with respect to RBA and RBA + IF treatments (Figure 5). Although no significant differences were observed in physiological efficiency variables, it should be mentioned that the RBA + IF treatment produced increases of 43% and 88% for IF and RBA, respectively, in comparison with the Control plants (Figure 5).

4 Discussion

4.1 Soil-tailored formulation enables function-oriented residue valorization

The significant Soil × Treatment interaction observed for most analyzed variables is the central result of this study. This demonstrates that the effectiveness of RBAs cannot be generalized across soils but depends on understanding the limitations of a soil for a specific crop. Additionally, the effect of an RBA on soil is also dependent on the interaction between the applied residues themselves. This finding supports a function-oriented approach to residue formulation for soil fertility and health, where amendments are designed to modify specific soil properties rather than simply to supply nutrients (basic approach). This approach is consistent with the principles of INM, which emphasize improving nutrient use efficiency through the interaction between soil, fertilizer, and crop demand rather than maximizing nutrient inputs (44, 45).

Among the four amendments evaluated, RBA1 represents the clearest example of successful soil-tailored formulation. Its application to the soil produced the highest grain yield, plant P uptake comparable to IF, and a lower Mn uptake despite the naturally high Mn availability of Soil 1. These responses indicate that simultaneously correcting soil constraints is more effective than addressing individual nutrient deficiencies independently (46). These benefits are the result of choosing the right materials that supply the right functions (e.g., nutrient supply, pH modification, and contribution to finer texture), even when they are combined with other widely different residues such as ashes with cork. The latter is also a key point because the selection of compatible materials also requires experimental validation, as testing all possible residue combinations is impractical. In this case (RBA1 applied to Soil 1), the observed increase in soil pH, the enhanced plant P uptake, and the lower Mn and Fe uptake—potentially reducing the risk of micronutrient antagonism or toxicity—were a result of an adequate formulation of the RBA for this specific soil.

Additionally, early P availability is particularly important for tillering and root development in cereals (47), whereas K promotes photosynthesis, assimilate translocation, and water regulation, thereby contributing to biomass accumulation and grain production (48). Therefore, the improved nutrient availability provided during the early stages of crop growth—through the inclusion of pure slag and white ash, among other components—likely contributed substantially to the superior agronomic performance of RBA1. However, the potential limitation of P—initially under the soil critical level—could have caused less tillering when RBA1 and RBA1 + IF1 were applied to the soil in comparison with IF1 alone. RBA1 + IF1 contained highly soluble KCl, which was probably the reason our results pointed to a P limitation rather than a K limitation. Similar improvements were obtained with formulated amendments in blueberry cultivation and vineyards, where tailoring the physicochemical characteristics of the amendment to the target soil and crop consistently enhanced crop productivity and soil quality (38, 49). Regarding toxicological effects, RBA1 significantly increased Ni concentration in straw relative to IF1, in contrast to what was observed in Soil 2. This was an unexpected result because even though the total concentration was 39 mg kg^-1, the availability of this element was low and the amendment increased the pH, which favors its precipitation.

The effects observed when RBA2 was applied to the soil highlight another critical aspect: the relevance of synchronizing nutrient release with crop demand. In addition to amendment incubation, a more labile OM fraction was included (40% orange peel) in order to acidify the soil through its nature, produce rapid oxidation, and increase nutrient availability during the first phenological stages. However, plant performance under RBA2 remained below that of IF2, which could be due to a limited nutrient release from the organic fraction that did not match the period of maximum crop demand (50, 51). This interpretation is further supported by the K recovery relative to the total K available (i.e., K applied plus soil-available K); IF2 promoted a K recovery of 29%, whereas plants under RBA2 only recovered 18%, suggesting that a longer incubation period may be required to improve nutrient mineralization and availability, particularly when the concentration of this element is below the critical level. Consequently, the lower number of tillers was reflected in a reduced biomass production and ultimately in a lower yield. On the other hand, organic materials such as cork and uncooked biomass from the cellulose extraction process (UBCP) had high C/N ratios, which may have slowed their decomposition and caused temporary microbial N immobilization (52, 53). Moreover, because their decomposition continued during crop growth, the associated microbial activity may have competed with plants for nutrients and increased soil O₂ consumption, potentially generating transient conditions that constrained root activity and crop development (54, 55). These results indicate that formulation must consider both total element concentrations (particularly for regulatory compliance) and element availability, mainly to synchronize nutrient release with crop physiological demands—attending to the residue nature, the soil properties, and environmental conditions. This aspect becomes particularly relevant when integrating RBAs with IF in sustainable strategies aligned with the circular economy. Thus, the authors suggest that residue mineralization dynamics should be incorporated as a key formulation parameter, allowing recycled nutrients to become available when crop demand is greatest and thereby maximizing nutrient use efficiency and soil quality.

The responses observed with RBA3 and RBA4 further indicate that formulation flexibility is also constrained by the diversity of functional materials available. This is based on the fact that the effectiveness of these formulations was lower than that observed under RBA1, suggesting that simplified formulations (the use of fewer materials) offer fewer opportunities to simultaneously enhance multiple soil properties and functionalities. Differences in total nutrient uptake in plants under different fertilization treatments were small and largely driven by differences in biomass production as in both previous cases. Grain N concentration was significantly higher under RBA3 than under IF3, although RBA3 captured the least N during its previous use as a filter. However, this increase in grain quality did not translate into a significant improvement in grain yield (physiological efficiency). A particularly relevant finding was that grain Zn concentration was consistently lower under inorganic fertilization than in the unfertilized control across all soils. This result is especially important for wheat, as Zn deficiency remains a major nutritional concern in many developing countries, where cereals constitute the main dietary Zn source and economically affordable biofortification strategies are often unavailable (56–58). Then, the consistent reduction in tiller number observed under RBA4—despite the absence of yield decrease—identifies Al sludge as a candidate factor because it was unique to RBA4; however, its individual contribution cannot be isolated from the present formulation design. Its proportion should therefore be evaluated in future formulations.

Generally, the soil-tailored RBA formulations consistently increased HI when combined with IF compared with IF alone across all evaluated soils. This response may reflect the complementary nutrient supply from both fertilizer sources: IF provided readily available nutrients during the early growth stages, whereas the RBA may have contributed to a more sustained nutrient supply during crop development. Sustained nutrient availability, particularly during the reproductive and grain-filling stages, may support post-anthesis growth and dry-matter allocation to the grain, thereby increasing grain production relative to straw biomass (59). This interpretation is further supported by the consistently higher—although non-significant—physiological efficiency observed under RBA + IF, suggesting that absorbed N was utilized more efficiently for grain production. However, our initial hypothesis that RBA + IF would outperform inorganic fertilization in all soils was not fully supported. Finally, one common pattern repeated in all soils was the significant increase in straw Na concentration under RBAs relative to IFs (except in Soil 3). The available K/available Na ratio in RBA1, 2, 3, and 4 was 6.3, 5.3, 1.6, and 6.8, respectively; we consider this unbalanced ratio relevant, especially in Soil 3, where crop K uptake can be limited by an excess of Na, an element associated with tillering abortion (60).

4.2 Limitations for human and animal consumption

No regulatory exceedances were observed except in the Control treatment of Soil 1, where Pb was 0.48 mg kg^-1 and Ni, 1.26 mg kg^-1, and the maximum levels allowed are 0.2 and 0.8 mg kg^-1, respectively. This may be due to the lower biomass production and the higher concentration of elements in this soil. Regarding the concentration of these elements in straw, none of the element concentrations were higher than the allowed maximum levels. It should be noted that Cd quantification limits (0.1 mg kg^-1) are close to the regulatory limits and therefore warrant special attention in future studies. This supports the importance of analyzing the soil prior to formulating the amendments.

However, a deeper evaluation showed that Ba concentrations in grain were consistently higher under IF than RBAs. This may be related to competition with Ca applied with RBA and the immobilization through sorption by organic ligands. Nevertheless, the most probable cause is the natural occurrence of Ba in the phosphate rock used to manufacture inorganic fertilizers (61–64)—superphosphate in this study (65). The same explanation may apply to the higher concentration of Cr found under IF2 because, despite the greater biomass production, Cr concentration in straw was significantly higher than that measured under RBA2. This indicates that the difference was not explained by a simple biomass-concentration effect. Cr and other trace elements occur in phosphate rock, but their concentrations vary considerably among deposits and fertilizer products (62, 64). Finally, straw Al concentration under IF3 showed a significant increase, which is undesirable in animal feed. These results highlight the importance of characterizing trace-element impurities in both conventional and recycled fertilizers, because although inorganic fertilizers are essential in agriculture, their potential secondary effects are not always sufficiently considered.

4.3 Feasibility of coupling gaseous mitigation with crop fertilization

The present study extends the concept of residue-based amendments beyond conventional waste valorization by demonstrating that the same combination of materials can mitigate gaseous emissions and subsequently function as a soil amendment. Unlike conventional filtration systems, these multifunctional RBAs are able to recycle the nutrients captured during gas filtration into crop production. This enrichment (N, C, and S) was not accounted for in the RBA formulation, since the objective was to determine its agronomic benefit. The gaseous N recovered during filtration represented an additional N input of approximately 34%, 27%, 13%, and 40% for RBA1, RBA2, RBA3, and RBA4, respectively. The RBAs produced crop performance comparable or superior to that with IF in three of the four soils. This indicates that total nutrient concentration alone does not determine crop response, reinforcing the importance of nutrient availability and release kinetics during the early stages of crop establishment, as was discussed previously. These findings suggest that the N captured during the previous use of RBAs as slurry gas filters could further reduce the performance gap between residue-based amendments and inorganic fertilization, highlighting the potential of this dual-purpose (“2-in-1”) formulation strategy as a viable approach for simultaneously mitigating gaseous emissions and improving nutrient recycling. The agronomic results demonstrate that this dual functionality can be compatible with crop performance, supporting a circular strategy that simultaneously mitigates ammonia emissions, recovers nutrients, and improves nitrogen use efficiency.

5 Conclusions

The formulation of residue-based amendments to improve the properties and functions of specific soils was appropriate and promising, with some formulations maintaining or improving crop performance relative to inorganic fertilizer (such as grain N concentration) and the combined RBA + IF treatment increasing the harvest index relative to IF. However, contrary to our initial hypothesis, the fertilization strategy based on INM did not consistently produce the best results. This finding indicates that, beyond adopting an INM approach, nutrient-release timing must be more accurately synchronized with crop demand. This study also demonstrated that the combined and controlled use of residues and by-products can maintain or improve soil and crop health, as illustrated by the performance of plants grown in Soil 1 under residue-based amendment 1. We can therefore state that the agronomic value of residue-based amendments depends less on their origin than on their capacity to alleviate the main physicochemical constraints of the target soil. Although our results are promising, it should be noted that they were obtained from a pot experiment using small experimental units, without leachate analysis, and under controlled incubation conditions. Therefore, this work provides a conceptual framework for advancing circular fertilization strategies beyond conventional residue valorization; however, the results require field validation, particularly to improve our understanding of nutrient dynamics, salinity, and material mineralization. The future development of this management strategy within a circular-economy context requires a better understanding of potentially recoverable materials, the nutritional and functional requirements of crops, and the rates and timing of nutrient release. Technological advances will enable the more efficient integration of data on soils, crops, residues, legislation, and environmental conditions, thereby facilitating the optimization of RBA formulations.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. Adrian thanks ChatGPT4o for helping him to improve English language of the first draft of the manuscript, and also for correcting the code needed in the software R to perform either statistical analysis or figures.

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.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fsoil.2026.1955073/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: José María De La Rosa, Spanish National Research Council (CSIC), Spain

Reviewed by: Marco E. Mng’ong’o, Mbeya University of Science and Technology, Tanzania

Nighet Musa, Azad Jammu & Kashmir, Pakistan