Biochar and Drying Technologies as Integrated Tools for Sustainable Pea Production and Functional Ingredient Generation

Abstract

The growing demand for sustainable agriculture requires strategies that simultaneously recover soil quality, improve crop yield, and add value to food products. This study evaluates walnut shell biochar (450 °C) as a circular amendment applied at 0, 10, and 20 t ha⁻¹ to an arid soil cultivated with pea (Pisum sativum L. cv. Onward) in San Juan, Argentina. Biochar enhanced soil porosity, respiration, organic carbon, and cation exchange capacity, resulting in higher plant biomass and a 30.9% increase in pod yield for the 20 t ha⁻¹ treatment. Pea grains were dehydrated by far-infrared drying at 70 °C, producing flour with improved lipid content, water absorption, and swelling capacity, which increased from 0.21 to 0.26 mL g⁻¹ under the 20 t ha⁻¹ treatment. The combined use of biochar and controlled drying highlights a viable pathway to close the soil–plant–food loop through resource valorization. This work contributes practical evidence of biochar’s multifunctional role in sustainable agri-food systems, aligned with circular economy principles.

1. Introduction

Population growth and agricultural intensification have put significant pressure on soils, causing their degradation through loss of organic matter, erosion, acidification, and contamination, as well as contributing to greenhouse gas emissions and climate change [1]. In turn, urbanization and industrial development have reduced the availability and quality of agricultural land, with an estimated 33% worldwide being moderately or severely degraded [2]. This deterioration compromises the fertility, productive capacity, and sustainability of agricultural systems, constituting a direct risk to food security systems. In addition, the degradation of soil functions threatens several dimensions of sustainability, including climate regulation, biodiversity conservation, and long-term food system resilience.

In light of these challenges, the restoration of degraded soils is an urgent necessity, and among the available alternatives, the application of biochar emerges as a promising strategy. Biochar is a material with high carbon content and great stability, capable of remaining in the soil for hundreds to thousands of years [3,4]. It is mainly produced through the pyrolysis process, which consists of the thermal treatment of biomass in the absence of oxygen, generating solid products (biochar), liquids (tar and aqueous fraction), and non-condensable gases [5]. It can be obtained from agricultural and forestry waste and various types of organic waste, which also represents an opportunity to recover large volumes of biowaste [6,7]. Therefore, biochar contributes not only to soil rehabilitation but also to circular economy principles, carbon sequestration, and waste-to-resource strategies. Specifically, the incorporation of biochar into agricultural soils represents a viable carbon sequestration pathway, as its highly aromatic structure resists microbial decomposition, enabling long-term carbon storage while simultaneously displacing the need for synthetic soil amendments that are carbon-intensive to produce. In the present study, walnut shells were selected as the biochar feedstock based on their regional relevance and physicochemical suitability. San Juan Province, where the field trial was conducted, is a major walnut-producing region in Argentina, and the processing of walnuts generates substantial quantities of shell waste that currently lack a dedicated valorization route. Walnut shells are characterized by high lignocellulosic content, low ash content, and favorable mechanical properties, which, upon pyrolysis, yield a biochar with high porosity, surface area, and structural stability attributes critical for soil amendment applications. Additionally, the use of walnut shell biochar enables continuity with previous studies from our research group, supporting comparability and reinforcing the reproducibility of findings. The properties of biochar depend on factors such as the type of raw material, the temperature, and the pyrolysis time, among other variables [5,8]. Its benefits have been widely documented, including improved soil structure, increased water and nutrient retention capacity, and promotion of plant growth [9,10,11]. Biochar also enhances the availability of key nutrients (N, P, and K) by reducing leaching losses, improving aeration, and decreasing soil density, all of which are facilitated by its porous nature. It stimulates microbial activity by providing suitable surfaces for beneficial communities, which promotes both nitrogen fixation and phosphorus bioavailability. These combined effects strengthen the soil–plant interface, while reducing dependence on synthetic fertilizers and improving environmental performance at the field scale. This reduction in synthetic input reliance not only lowers greenhouse gas emissions associated with fertilizer production but also enhances the overall sustainability of the supply chain by decreasing external input costs and promoting closed-loop nutrient cycling within agricultural systems.

The cultivation of peas (Pisum sativum L.) is particularly important. Like other legumes, it belongs to the Fabaceae family and is known for its ability to fix atmospheric nitrogen in the soil. This contributes to the sustainability of agricultural systems [12]. Peas rank fourth among the most widely produced legumes worldwide, with an estimated production of 33.3 million tons in 2024 and an average growth rate of 1.6% per year over the last twenty years [13]. Canada is the leading global producer, followed by Russia, China, and the USA [14]. In Argentina, the provinces of Buenos Aires and Santa Fe account for most of the cultivated area, mainly producing dry beans for domestic consumption and export (88–90%), and to a lesser extent, green beans and fresh string beans [15,16]. In addition to its importance in terms of production, peas are a food with high nutritional value, with an average protein content of 22.6%, surpassed only by lentils, and a low lipid content (1.38%) dominated by linoleic acid [17]. Its mineral content (K, P, Ca, and Fe), vitamins (A, K, folate, and carotenoids), and essential amino acids such as lysine reinforce its value in the human diet [18,19,20]. The increasing interest in plant-based proteins positions pea as a strategic crop for sustainable diets and reduced environmental footprints.

Moreover, pea flour and its derivatives (concentrated proteins, starches, and shell fibers) are used in the food industry to produce breads, beverages, supplements, and meat alternatives, and stand out for their ability to improve both the nutritional profile and functionality of products [18,21]. To obtain these ingredients with consistent quality and stability, efficient drying of the raw material is a critical step in the production chain. Drying is an effective strategy for enhancing and preserving agricultural products, reducing water activity in foods, and extending their shelf life by inhibiting microbial growth [22]. In addition to improving safety, drying reduces the weight and volume of products, facilitating their storage and transport [23]. The choice of method depends on the type of raw material and the desired characteristics of the final product. Among the available techniques, indirect solar drying best preserves color, flavor, and nutrients such as vitamins A and C [24,25,26], although it depends on climatic conditions and requires long periods of time [27]. Alternatively, infrared and microwave drying offer greater efficiency, speed, and better quality of the final product [26,28,29]. However, while the agronomic benefits of biochar are well-documented, there is a critical knowledge gap regarding its cascading effects on the post-harvest processing and techno-functional properties of the resulting food ingredients. Specifically, it is unknown whether the improvements in crop quality from biochar-amended soils translate into flours with superior functional attributes after drying. This study addresses this gap by linking a circular soil amendment directly to the quality of a processed ingredient. By integrating soil restoration, waste valorization, and functional food development, the pea–biochar system offers a multidimensional contribution to sustainable production and consumption frameworks.

Alternatively, infrared and microwave drying offer greater efficiency, speed, and better quality of the final product [26,28,29]. Selecting energy-efficient and quality-preserving drying technologies is essential to align food processing with sustainability goals. In this regard, drying and grinding peas to obtain flour represents a value-added alternative, generating a versatile product that can be used in the formulation of healthy and sustainable foods, with potential impact on the agri-food industry. By integrating soil restoration, waste valorization, and functional food development, the pea–biochar system offers a multidimensional contribution to sustainable production and consumption frameworks. Building on this rationale, the present study was designed to assess this integrated approach from soil to food. The main objectives were (1) to produce biochar from regional agricultural by-products (walnut shells) and characterize its physicochemical properties; (2) to evaluate the effects of different application rates of this biochar (0, 10, and 20 t ha⁻¹) on key soil health indicators and the agronomic performance of pea (P. sativum L. cv. Onward) cultivation; (3) to compare different drying technologies (solar, convective, and far-infrared) for pea grains to identify the most efficient method for preserving quality; and (4) to characterize the nutritional and techno-functional properties of the resulting pea flours and assess how they are influenced by the prior soil biochar treatment.

2. Materials and Methods

2.1. Study Site and Biochar

The study area was located in Rodeo, Iglesia Department, San Juan Province, Argentina (30°12′54.7″ S; 69°09′39.8″ W). The trial began in September with the preparation of the land and continued until the end of December with the harvest. The average climatic conditions during this period were temperature of 22.9 °C, precipitation of 65 mm, humidity of 45%, and solar radiation of 310 W m⁻².

The origin, physico-chemical characteristics, and further specific properties of the biochar used in this study (WSB, biochar from walnut shells) were previously published by Zabaleta et al. [7] and Sánchez et al. [30]. Appendix A characterizes the walnut shell biochar (WSB) used in the study, detailing its production via pyrolysis at 450 °C and its key physicochemical properties, such as pH, surface area, and elemental composition, supported by analytical techniques including SEM and FTIR. In brief, biomass was pyrolyzed at 450 °C for 2 h. For the field experiment, the biochar was ground to pass through a 2 mm sieve and homogenized to ensure uniform distribution within the soil matrix.

2.2. Field Experiment

A representative sandy loam was randomly chosen for the field trial. Before sowing, biochar was incorporated into the soil at application rates of 0, 10, and 20 t ha⁻¹ (designated T0, T1, and T2, respectively). The experiment was arranged in a randomized complete block design (RCBD) with three blocks. Within each block, the three treatments were randomly assigned to individual plots, resulting in three replicates per treatment. Each plot measured 3 m × 5 m. The biochar was evenly distributed on the soil surface on the culture line and incorporated to a depth of approximately 10–15 cm using a rake.

2.3. Soil Sampling and Treatment

In the area where the test was carried out, soil samples of each treatment were collected, consisting of 10 subsamples obtained in a zigzag pattern at a depth of 20 cm. These subsamples were stored in a plastic bag, labelled appropriately, and transferred to the laboratory for further analysis. A portion was further air-dried at room temperature. The soil aggregates were then ground, passed through a 2 mm sieve, and homogenised for analysis. In each plot, all measurements were performed in three replications twice: at 60 and 90 days of biochar incubation.

To determine the pH, a soil:water mixture of 1:2.5 w v⁻¹ was prepared [31]. Electrical conductivity (EC), a saturated soil paste was first prepared, and the EC was measured in the saturation extract [31]. Soil organic matter (OM) was determined using the loss on ignition method [32]. Subsequently, organic carbon (OC) estimation was performed using the Van Bemmelen factor (1.724), which assumes that the OM in the soil contains an average of 58% carbon. The cylinder method was used to determine the bulk density (BD) [33]. Gravimetric moisture (GM) was determined by drying a fine soil sample at 105 °C for 24 h and weighing it to a constant weight. Total nitrogen (TN) was analyzed with the Kjeldahl method. Soil respiration was estimated based on CO₂ measurements from soil incubated in a closed system, in which CO₂ is trapped in a NaOH solution, which is then titrated with HCl [34]. The results were expressed as mg CO₂ g⁻¹ dw. Cation exchange capacity (CEC) and water holding capacity (WHC) were determined according to Barton and Karathanasis [35] and Dugan et al. [36], respectively. The ash content was determined by calcination in a muffle furnace at 580 °C for 4 h [37]. All measurements were made in triplicate.

2.4. Plant Material and Agronomic Measurements

Green pea (Pisum sativum L. cvar Onward) was selected as plant material. These are semi-late cultivars, suitable for early and normal sowing due to their tolerance to frost in the vegetative stage. No chemical fertilisers were applied during the growing cycle. Irrigation was conducted at regular intervals, depending on the requirements of the crop, and weed control was done manually.

Plant height and yield components such as number of pods plant⁻¹, pod weight, pod length, pod weight plant⁻¹, number of seeds pod⁻¹, weight of 10 seeds, and seed diameter were recorded. Pods were harvested at the fresh green stage after the pod-filling period, based on phenological and visual criteria (full pod size, bright green color, and fully developed but immature seeds). After the harvest, pod yields were determined in each plot (in kg per plot), and the results were expressed per hectare.

2.5. Flour Production and Characterization

This section details the drying, milling, and analytical procedures used to obtain and characterize pea flour. Drying kinetics were modelled, flours were produced by controlled milling, and their compositional and functional properties were evaluated using standard analytical methods and statistical validation.

2.5.1. Drying Curve Determination and Modeling

To obtain the drying curves, the moisture ratio (MR) was calculated using the following equation (Equation (1)):

$$\mathrm{MR} = {\displaystyle \frac{({\mathrm{M}}_{\mathrm{t}}-{ \mathrm{M}}_{\mathrm{e}})}{({\mathrm{M}}_0-{\mathrm{M}}_{\mathrm{e}})}}$$

(1)

where MR is the moisture ratio, M_t corresponds to the moisture content recorded at time t, M_e is the equilibrium moisture content, and M_o is the initial moisture content [38].

For mathematical modelling, different thin-layer drying models were tested (see

Table A3. Mathematical models used to describe the thin-layer drying kinetics of pea.

N°	Model Name	Model Expression
1	Newton	$\mathrm{MR}=\exp (-\mathrm{kt})$
2	Page	$\mathrm{MR}=\exp (-\mathrm{k}{\mathrm{t}}^{\mathrm{n}})$
3	Modified Page	$\mathrm{MR}=\exp \left(\right(-{\mathrm{k}\mathrm{t})}^{\mathrm{n}})$
4	Henderson and Pabis	$\mathrm{MR}=\mathrm{a}· \exp (-\mathrm{kt})$
5	Modified Henderson and Pabis	$\mathrm{MR}=\mathrm{a}·\exp \left(-\mathrm{kt}\right)+\mathrm{b}·\exp (-\mathrm{gt})+\mathrm{c}·\exp (-\mathrm{ht})$
6	Wang and Singh	$\mathrm{MR}=1+\mathrm{at}+{\mathrm{b}\mathrm{t}}^2$
7	Logarithmic	$\mathrm{MR}=\mathrm{a}·\exp (-\mathrm{kt})+\mathrm{c}$
8	Midilli	$\mathrm{MR}=\mathrm{a}·\exp (-\mathrm{k}{\mathrm{t}}^{\mathrm{n}})+\mathrm{bt}$
9	Two-term Exponential	$\mathrm{MR}=\mathrm{a}·\exp (-\mathrm{kt}) +(1-\mathrm{a})·\exp (-\mathrm{akt})$
10	Two-term	$\mathrm{MR}=\mathrm{a}·\exp (-\mathrm{kt})+\mathrm{b}·\exp (-\mathrm{nt})$
11	Noomhorm and Verma	$\mathrm{MR}=\mathrm{a}·\exp (-\mathrm{kt})+\mathrm{b}·\exp (-\mathrm{nt})+\mathrm{c}$

MR (moisture ratio), t (time), k, n, a, b, g, and c (constants).

from Appendix B.1). The adjustment indices corresponded to the adjusted coefficient of determination (Equation (3)).

2.5.2. Flour Analyses

Each pea sample (from the different treatments T0, T1, and T2) was ground separately in a stainless-steel blade mill (TDMC, TecnoDalvo, Santa Fe, Argentina) until a particle size of 0.10 to 0.20 mm was achieved [39]. The flours were stored in plastic bags at room temperature. The AOAC methods [40] were followed to determine ash (AOAC 923.03), protein (AOAC 960.52), lipid (AOAC 920.39), and crude fiber (AOAC 962.09) contents. Moisture content was determined at 105 °C using a moisture analyser (PMR 50 NH, Radwag, Radom, Poland). Water activity (a_w) was determined according to Román et al. [23], using an electronic water activity meter with dew point (Aqualab 3TE, Decagon, Washington, USA). Appendix B.2 describes the methodology used to estimate the levels of specific amino acids (lysine, methionine, cysteine, tryptophan, and threonine) in the pea flours using predictive regression equations based on the protein content [41]. Water and oil retention capacity (WHCf and OHC, respectively) were determined as described by Silva et al. [42]. Swelling capacity (SC) was measured according to the procedure established by Onipe et al. [43]. The total carbohydrate content was determined using the difference method [44]:

$$\% \mathrm{Carbohydrates} = 100 - (\%\mathrm{Moisture} + \%\mathrm{Ash} + \%\mathrm{Protein} + \%\mathrm{Lipids})$$

(2)

2.6. Statistical Analyses

The data set shows the values as averages ± standard deviation (SD). The data were found to meet the assumptions of normality and homogeneity of variance. Significance was determined using a one-way or two-way ANOVA according to factors, and tested with Tukey’s post hoc multiple comparison test. p values < 0.05 were considered statistically significant.

In addition, the following statistical coefficients were used for the evaluation of models: chi-square (χ²), sum of squared errors (SSE), and root mean square error (RMSE). The model that achieved the best fit was the one that exhibited the highest R² value and the lowest χ², SSE, and RMSE values. The statistical parameters evaluated are described below (Equations (3)–(6)):

$${\mathrm{R}}^2 = 1 - \left[{\displaystyle \frac{{\sum }_{\mathrm{i}=1}^{\mathrm{N}}{({\mathrm{MR}}_{\mathrm{pre} \mathrm{i}} -{ \mathrm{MR}}_{\exp \mathrm{i}})}^2}{{\sum }_{\mathrm{i}=1}^{\mathrm{N}}{({\overline{\mathrm{MR}}}_{\exp }- {\mathrm{MR}}_{\exp \mathrm{i}})}^2}}\right]$$

(3)

$${\mathsf{{\rm X}}}^2={\displaystyle \frac{{\sum }_{\mathrm{i}=1}^{\mathrm{N}}{({\mathrm{MR}}_{\mathrm{pre} \mathrm{i}} - {\mathrm{MR}}_{\exp \mathrm{i}})}^2}{\mathrm{N} - \mathrm{z}}}$$

(4)

$$\mathrm{SSE}={\displaystyle \frac{1}{\mathrm{N}}}\sum _{\mathrm{i}=1}^{\mathrm{N}}{({\mathrm{MR}}_{\mathrm{pre} \mathrm{i}} - {\mathrm{MR}}_{\exp \mathrm{i}})}^2$$

(5)

$$\mathrm{RMSE}=\sqrt{{\displaystyle \frac{1}{\mathrm{N}}}\sum _{\mathrm{i}=1}^{\mathrm{N}}{({\mathrm{MR}}_{\mathrm{pre} \mathrm{i}} - {\mathrm{MR}}_{\exp \mathrm{i}})}^2}$$

(6)

MR_{pre i} corresponds to the moisture ratio predicted by mathematical models, and MR_{exp i} corresponds to the moisture ratio obtained experimentally. $\overline{\mathrm{MR}}$_{pre i} corresponds to the average of the predicted moisture ratio, N corresponds to the number of data points, and z corresponds to the number of model constants.

3. Results and Discussion

3.1. Soil Physicochemical Properties

Table 1. Properties measured in the soil for both samples. Values are shown as average ± SD.

Properties	Days	Treatments
		T0	T1	T2
pH	60	5.48 ± 0.02 b	6.40 ± 0.20 a	6.44 ± 0.15 a
	90	6.26 ± 0.30 a	6.09 ± 0.04 a	6.05 ± 0.02 a
EC (µS cm−1)	60	305 ± 11 a	285 ± 15 a	294 ± 19 a
	90	233 ± 20 b	357 ± 22 a	393 ± 10 a
GM (%)	60	1.40 ± 0.08 a	1.39 ± 0.01 a	1.45 ± 0.07 a
	90	1.47 ± 0.02 a	1.59 ± 0.01 a	1.74 ± 0.05 a
WHC (%)	60	21.50 ± 1.02 a	22.18 ± 1.94 a	21.80 ± 1.98 a
	90	16.51 ± 4.71 a	20.78 ± 0.81 a	21.30 ± 0.81 a
BD (g cm−3)	60	1.45 ± 0.16 a	1.45 ± 0.16 a	1.47 ± 0.16 a
	90	1.47 ± 0.12 a	1.36 ± 0.12 b	1.35 ± 0.12 b
Ash (%)	60	6.82 ± 0.07 a	6.75 ± 0.25 a	6.81 ± 0.18 a
	90	7.12 ± 0.27 b	7.93 ± 0.03 b	12.01 ± 0.49 a
Soil respiration (mg CO2 g dw−1)	60	21.99 ± 1.10 a	23.18 ± 2.1 a	23.54 ± 1.25 a
	90	5.12 ± 0.39 c	12.76 ± 0.56 b	22.78 ± 1.33 a
OM (%)	60	1.02 ± 0.01 a	1.05 ± 0.02 a	0.98 ± 0.03 a
	90	1.08 ± 0.02 b	1.21 ± 0.09 ab	1.59 ± 0.34 a
OC (%)	60	0.59 ± 0.01 a	0.60 ± 0.04 a	0.57 ± 0.02 a
	90	0.62 ± 0.01 b	0.70 ± 0.05 ab	0.92 ± 0.19 a
CEC (meq 100 g−1)	60	202 ± 6 a	200 ± 15 a	194 ± 12 a
	90	186 ± 4 b	193 ± 2 b	203 ± 2 a
TN (%)	60	0.34 ± 0.02 b	0.33 ± 0.02 a	0.33 ± 0.01 a
	90	0.31 ± 0.02 a	0.38 ± 0.04 a	0.39 ± 0.02 a

Values followed by different superscript letters in the same row are statistically significant. p < 0.05. T0 (soil); T1 (soil + 10 tn ha⁻¹ WSB); T2 (soil + 20 tn ha⁻¹ WSB); EC (electrical conductivity); GM (gravimetric moisture); WHC (water holding capacity); BD (bulk density); OM (organic matter); OC (organic carbon); CEC (cation exchange capacity); TN (total nitrogen).

shows the results obtained from soil samples analysed 60 days (sampling 1) and 90 days (sampling 2) after the application of the amendment. Previous studies [45,46] suggest that with biochar amendment, incubation times greater than 60 days, improvements in soil properties can be observed, which encouraged a detailed analysis of the changes that occurred in the different parameters evaluated.

Analysis of soil BD showed a reduction in soils amended with WSB in the second sampling. Specifically, BD decreased from 1.47 g cm⁻³ in T0 to 1.35–1.36 g cm⁻³ in the biochar-treated plots, representing a reduction of approximately 8%. This effect is consistent with previous studies that have indicated that the incorporation of biochar, due to its low density and porous structure, improves aeration [47]. The reduction in BD is a key aspect in soils with compaction problems, as it facilitates root penetration, promotes water infiltration, and encourages microbial activity [48].

The results reveal that pH values increased significantly due to the addition of biochar to the soil in the first sampling, while the EC value remained unchanged. However, one month later, the pH showed a tendency to decrease with the addition of biochar, whereas the EC increased significantly with the amendment compared to the control soil (p < 0.01). Soil pH is a crucial characteristic that influences nutrient availability and plant development [49]. The initial increase in pH could be attributed to the release of carbonates and alkaline elements (K⁺ and Ca²⁺) present in WSB, which is consistent with what has been reported in the literature [50]. However, the subsequent decrease in pH could be due to the interaction of biochar with soil microbiota, OM mineralisation, and the possible leaching of basic ions. In parallel, the increase in EC suggests that, as soluble minerals from biochar are released into the soil solution, the ionic concentration increases, which could influence nutrient availability [51,52].

GM was higher in soils with biochar than in the control in both samplings; this value is closely related to WHC, which increased gradually. However, they were not statistically significant. The absence of a statistically significant increase in WHC during the first sampling may be attributed to the relatively short incubation period (60 days), which is often insufficient for biochar particles to become fully integrated into the soil matrix and for surface oxidation processes to enhance hydrophilicity, both of which are necessary to observe measurable improvements in water retention. This effect has been widely reported [53,54]. Reasons that could explain this phenomenon include the increase in CEC and the porous structure of biochar, which impacts water retention and moisture availability for plants [55]. The ash content constitutes the mineral fraction of the samples after calcination, and this value was higher in soils with organic amendments than in the control in the second sampling. For the first sampling, the values were similar for the control and the treated soils. A higher ash content could imply a higher level of macro and micronutrients that were concentrated in the biochar during the pyrolysis process [56].

Soil respiration was observed to increase significantly in treated soils compared to control soils, and this trend was observed in both samplings. By the second sampling, soil respiration in the T2 treatment was 22.78 mg CO₂ g⁻¹, which was 4.4 times higher than in the control soil (5.12 mg CO₂ g⁻¹). In the second sampling, OM and OC values were higher in soils with biochar than in the control (p < 0.05). The T2 treatment showed a 47% increase in OM (from 1.08% to 1.59%) and a 48% increase in OC (from 0.62% to 0.92%) compared to the control. These results are relevant because the addition of biochar as a soil amendment provides degradable OC to soil organisms, creating more favourable habitats and thus facilitating soil biological activities [57,58]. The observed increases in soil organic carbon, respiration, and cation exchange capacity under biochar amendment underscore the material’s dual role in enhancing soil health while contributing to carbon sequestration. These improvements are foundational to long-term agricultural sustainability, as they enhance soil resilience against degradation, reduce vulnerability to climate variability, and support productive capacity without reliance on non-renewable inputs. However, no significant variations were observed in the first sampling, which could suggest that biochar requires a period of time to be incorporated into the soil biogeochemical cycles. The delayed increase in soil respiration and organic matter content reflects the time needed for microbial colonization of biochar surfaces and the gradual mineralization of labile biochar fractions, which together stimulate microbial activity and contribute to the accumulation of stabilized organic carbon in the soil.

CEC increased significantly (p < 0.01) only in the second sampling, with T2 having the highest value. CEC in T2 was 203 meq 100 g⁻¹, which is 9% higher than in the control soil (186 meq 100 g⁻¹). This delayed response is consistent with the mechanism by which freshly applied biochar undergoes gradual surface oxidation upon exposure to soil oxygen and moisture, leading to the formation of carboxyl and phenolic functional groups that increase net negative charge and, consequently, cation exchange capacity over time. This has been reported by several authors [58,59,60]. When freshly applied biochar comes into contact with oxygen and water present in the soil, spontaneous surface oxidation reactions occur, generating an increase in its net negative charge and, therefore, an increase in CEC. Over time, biochar particles tend to have a high negative charge, thus favouring cation retention in the soil [61]. The results revealed that the addition of biochar did not statistically influence the TN of the soils studied. It is important to note that biochar obtained from biomass residues is often deficient in plant nutrients, especially nitrogen, and therefore cannot serve as a primary source of nutrients to replace chemical fertilisers in modern agriculture [57]. It can incorporate other nutrients as well as mitigate leaching losses and improve nutrient retention in the soil profile. The value found for biochar is consistent with that of biochar obtained from crop residues (TN < 2.7%) [62].

3.2. Pea Growth and Productivity

The pea cultivation cycle in this study began with the emergence of seedlings on 12 September 2024. Following this period, the vegetative growth stage lasted until early November. Flowering began on 3 November 2024. After flowering, fruiting, and subsequent pod filling occurred. This period lasted until the end of December 2024, when fresh green pods were harvested. The growing cycle lasted approximately 90–100 days (see Figure 1).

Table 2. Agronomic variables in pea plants. Values are shown as mean ± SD.

Variable	Treatments
	T0	T1	T2
Plant height (cm)	45.26 ± 3.25 a	46.25 ± 4.18 a	46.75 ± 5.01 a
Internode length (cm)	3.96 ± 0.21 b	4.41 ± 0.34 a	4.40 ± 0.30 a
Relative chlorophyll content (SPAD)	28.85 ± 1.56 b	35.47 ± 2.65 a	31.58 ± 3.25 ab
Fresh weight of pods (g)	4.45 ± 0.35 b	5.53 ± 0.45 a	5.84 ± 0.51 a
Pod length (cm)	6.86 ± 0.59 a	7.27 ± 0.66 a	7.42 ± 0.63 a
Number of seeds per pod−1	5.05 ± 0.25 b	5.74 ± 0.48 a	5.89 ± 0.48 a
Weight of 10 grains (g)	4.22 ± 0.40 b	4.31 ± 0.33 b	4.95 ± 0.37 a
Grain diameter (cm)	1.10 ± 0.01 a	1.11 ± 0. 01 a	1.11 ± 0. 01 a
Pod yield (ton ha−1)	5.76 ± 0.93 b	7.23 ± 0.62 a	7.54 ± 0.54 a

Values followed by different superscript letters in the same row are statistically significant.

shows the effects of the treatments on various morphophysiological variables of pea plants under different biochar amendment treatments: T0, T1, and T2. In terms of plant height, a slight increase was observed with the application of biochar, although there were no significant differences between T1 and T2. This result coincided with that obtained by Bhattarai et al. [63] when they applied biochar amendments from different feedstock to pea crops. The length of the internodes showed a significant increase (p < 0.05) in the treatments with biochar (T1 and T2) compared to the control (T0). The values for plant height and internode length were similar to those obtained in the study by Khichi et al. [64], in which they evaluated eight pea varieties. Chlorophyll content was significantly higher in the pea plants in treatment T1 than in the other treatments. This behaviour could be attributed to greater nutrient availability and improved soil structure promoted by biochar, which enhances root growth and, consequently, the aerial development of the plant. For most of the yield components evaluated, the incorporation of WSB was found to have a positive influence (T1 and T2) compared to the control (T0), suggesting that its application may affect the reproductive development of the plant, except for pod length, which varied between 6.86 and 7.42 cm, and grain diameter, which ranged from 1.10 to 1.11 cm. The results obtained are consistent with other previous studies [63,65], which evaluated yield components in pea plants with and without biochar amendment. The fresh weight of pods and the number of seeds per pod were significantly higher in the biochar treatments, suggesting that its incorporation into the soil has a positive effect on crop productivity. This can be explained by the improvement in water and nutrient availability, as well as the stimulation of microbial activity in the soil, factors that directly influence the reproductive development of the plant. The weight of 10 grains also showed a significant increase in T2 compared to T0, suggesting that biochar not only favours seed quantity but also seed filling, increasing their individual weight. Pod yield (tonnes ha⁻¹) increased significantly with the application of biochar, reaching higher values in T2. This is consistent with previous studies indicating that improving nutrient availability and soil structure through organic amendments can result in higher grain weight and quality [66,67].

3.3. Drying Performance

A preliminary evaluation of different drying techniques—solar, convective, and far infrared—was conducted to identify the most suitable method for pea grain dehydration. Appendix B.1 details the evaluation of different drying technologies (solar, convective, and far-infrared) for pea grains, including the modeling of drying kinetics to identify the most efficient method, which was far-infrared drying at 70 °C. The purpose of this assessment was to compare drying performance and quality preservation to select the most reliable technique for subsequent analyses and flour production. Although solar drying (SolD) represents a low-cost and environmentally friendly approach, its major limitation lies in its strong dependence on weather conditions, which restricts process control and standardization [25,38]. In contrast, both convective drying (ConD) and infrared drying (InfD) systems allowed better regulation of temperature and air flow, ensuring reproducible conditions [26,68]. Among them, infrared drying at 70 °C showed the best overall performance, achieving faster and more uniform moisture removal while minimizing color degradation and nutrient losses (see Appendix B.1). These results demonstrated that InfD provides a more efficient and consistent dehydration process in terms of time, making it the most suitable method for obtaining pea flours with preserved physicochemical and functional quality.

Although solar drying (SolD) resulted in slightly lower color variation (ΔE) and marginally higher protein retention compared to far-infrared drying (InfD), InfD was selected as the most suitable method for producing pea flours in this study. The primary rationale is based on process reliability, scalability, and operational efficiency. SolD is inherently dependent on ambient weather conditions, which limit process control, reproducibility, and suitability for year-round industrial applications. In contrast, InfD operates under controlled temperature and airflow, ensuring consistent drying conditions and uniform product quality. Additionally, InfD significantly reduced drying time from 330 min (SolD) to 240 min, resulting in higher throughput and lower labor requirements. These attributes make InfD a more robust and scalable technology for the standardized production of functional pea flours, aligning with the development of sustainable and reproducible food ingredients.

The validation of the established model was carried out by comparing the experimental values of the moisture ratio with the predicted values (Figure 2). The results showed that the Midilli model could be used to explain the drying behaviour of peas in a thin-layer model. The high coefficient of determination (R² = 0.9941) and the low error values (RMSE = 0.0276) confirm the model’s excellent fit. From a practical standpoint, this reliable model enables accurate prediction of drying times under far-infrared conditions, which is essential for scaling up the process, optimizing energy efficiency, and ensuring consistent product quality in industrial applications. The results of the evaluated parameters X², SSE, RMSE, and R² were 0.0008, 0.0007, 0.0276, and 0.9941, respectively.

3.4. Flour Composition and Functional Properties

The values for the proximal composition and characterisation of the flours obtained from the different treatments are detailed in

Table 3. Characterization of pea flours of the Onward variety on a dry-weight basis according to the treatments used. Values are shown as mean ± SD.

Variable	Treatments
	T0	T1	T2
Moisture (%)	10.23 ± 1.00 a	9.62 ± 0.78 a	10.24 ± 0.64 a
Protein (%)	34.55 ± 1.25 a	35.11 ± 1.00 a	35.41 ± 0.95 a
Lipids (%)	1.49 ± 0.17 b	2.66 ± 0.11 a	1.99 ± 0.05 ab
Ash (%)	3.69 ± 0.21 b	3.87 ± 0.09 a	3.72 ± 0.18 ab
Crude fiber (%)	8.10 ± 0.40 b	7.31 ± 0.36 a	7.23 ± 0.41 a
Carbohydrates (%)	49.53 ± 1.23 a	48.74 ± 1.86 a	49.13± 2.25 a
WHCf (g water g flour−1)	2.52 ± 1.00 a	2.38 ± 0.78 ab	2.33 ± 0.64 b
OHC (g oil g flour−1)	0.76 ± 0.02 b	0.85 ± 0.01 a	0.84 ± 0.02 a
SC (mL g flour−1)	0.21 ± 0.01 b	0.23 ± 0.01 ab	0.26 ± 0.01 a
aw	0.40± 0.05 a	0.39 ± 0.11 a	0.42 ± 0.17 a
Lysine (g 100 g protein−1)	2.42 ± 0.03 a	2.46 ± 0.00 a	2.47 ± 0.05 a
Methionine (g 100 g protein−1)	0.32 ± 0.00 a	0.33 ± 0.00 a	0.33 ± 0.01 a
Cysteine (g 100 g protein−1)	0.42 ± 0.00 a	0.43 ± 0.00 a	0.43 ± 0.00 a
Tryptophan (g 100 g protein−1)	0.28 ± 0.00 a	0.28 ± 0.00 a	0.28 ± 0.01 a
Threonine (g 100 g protein−1)	1.21 ± 0.01 a	1.22 ± 0.00 a	1.23 ± 0.02 a

Values followed by different superscript letters in the same row are statistically significant, p < 0.05. WHCf (water holding capacity flour), OHC (oil holding capacity), SC (swelling capacity), a_w (water activity).

. The flours obtained are shown in Figure 3. The ANOVA results showed that the ash, lipids, and crude fibre content variables were influenced by the addition of biochar (F = 6.408, p value = 0.0324, F = 14.25, p value = 0.0049, and F = 16.33, p value = 0.0037, respectively). The flours analysed were found to consist mainly of carbohydrates, close to 50%, followed by proteins with values of 34–35%. In addition, they had lower proportions of crude fibre (7.23–8.10%), ash (3.69–3.87%), and lipids (1.49–2.66%). The results obtained are consistent with previous research on flours from the same matrix, as reported by Wani and Kumar [19], Millar et al. [20], González et al. [69], and Pedrosa et al. [18].

González et al. [69] studied pea flours (P. sativum L. cv. Warindo) dried at three temperatures (50, 60, and 70 °C) and observed that the chemical composition of the flours did not vary, while the in vitro digestibility of proteins and starch did increase at 70 °C. This effect is desirable in proteins, but not in starch.

The flours had low a_w values (0.39–0.42). In flours, both moisture and water activity are fundamental parameters for defining their stability and behaviour during storage and processing [70]. Although both properties are related, water activity does not depend exclusively on total water content, but rather on the fraction available to participate in physicochemical and microbiological reactions. In general terms, it is expected that as the moisture content decreases, a_w will also decrease, as the amount of free water is reduced. However, this relationship can be altered by the nature of the food matrix, as components such as starches, proteins, or crude fibre can retain water structurally, reducing a_w even when the moisture content is not extremely low [71]. Therefore, the joint analysis of both parameters is essential to interpret the functional behaviour and shelf life of the flours obtained.

WHCf and OHC constitute the abilities of a matrix to trap water and oil, respectively. WHCf values were found to be between 2.30–2.50 g g⁻¹, and T0 and T2 were found to be statistically different (F = 9.41; p value = 0.0141; ANOVA). The ability of flours to absorb and retain water is fundamental in baking processes, as it directly influences variables such as the cooking time required and the final texture of the baked product [72]. The OHC was 0.76–0.85 g g⁻¹ and was significantly different in the control group from groups T1 and T2 (F = 14.12; p value = 0.0054; ANOVA). In foods with high protein content, a marked lipid retention capacity is observed, attributed to the functional properties of proteins. This affinity for oil and water is determined by the structural characteristics of proteins, such as their configuration, the nature of their amino acids, and the balance between polar and hydrophobic regions on their surface [73]. The OHC values in this study were higher than those reported by Laing [74], who observed OHC values of 0.48–0.61 g g⁻¹ in yellow pea flour.

On the other hand, SC represents the degree to which starch granules can incorporate water and expand, reflecting in turn the intensity of the molecular interactions present in their internal structure [73]. The value presented was 0.21–0.26 mL g⁻¹, and this techno-functional property showed significant differences between the groups of flours evaluated. Flours from T2 had the highest value (F = 6.29; p value = 0.0390; ANOVA). SC and carbohydrates are highly correlated with each other. Although starch content was not determined in this study, the carbohydrate content of peas is predominantly starch. These differences in techno-functional properties are likely attributable to biochar-induced changes in soil physicochemical conditions, such as enhanced nutrient availability, improved water retention, and stimulated microbial activity, which in turn influenced nutrient partitioning during grain development. Such shifts can alter the organization and interactions of macromolecules (proteins, starches, and fibers) within the grain matrix, ultimately affecting functional attributes like WHCf and SC.

The estimated amino acid composition of the pea flours obtained showed that they have a higher lysine content, followed by threonine, cysteine, methionine, and finally tryptophan. These values are lower than those reported by Fischer et al. [75]. The differences may be related to variety, geographical area, management, among others.

4. Conclusions

This study evaluated an integrated approach linking walnut shell biochar application to soil health, pea crop performance, and the techno-functional quality of derived flours. Four key findings emerge: (1) Biochar amendment at 20 t ha⁻¹, after a 90-day incubation period, significantly improved soil health indicators, including a 9% increase in cation exchange capacity, a 47–48% increase in organic matter and organic carbon, and a 4.4-fold increase in soil respiration. These improvements are foundational to long-term soil sustainability and carbon storage. (2) Agronomic performance was substantially enhanced, with green pod yield increasing by up to 30.9% under the 20 t ha⁻¹ treatment, demonstrating biochar’s capacity to improve food production per unit of land. (3) Far-infrared drying at 70 °C was identified as the most suitable dehydration method, offering a balance of efficiency (240 min drying time) and quality preservation, making it a scalable option for producing consistent pea flour. (4) Biochar application indirectly influenced flour functionality: flours from the 20 t ha⁻¹ treatment exhibited enhanced swelling capacity (from 0.21 to 0.26 mL g⁻¹) and oil holding capacity, reflecting biochar-mediated effects on soil nutrient availability and grain development. In contrast, the techno-functional properties were more sensitive to the application of biochar as an amendment, showing differentiated responses that likely reflect indirect effects mediated by the influence of biochar on soil quality and plant nutrient partitioning, rather than direct modifications to flour components at the molecular level.

Collectively, these results validate a circular economy model in which an agro-industrial residue, walnut shell, is transformed into a soil amendment that improves soil health, crop yield, and the functional quality of a processed food ingredient, while reducing reliance on synthetic inputs and closing resource loops.