Statement of Novelty

We comprehensively assessed the primary properties and N content and pools of biochars obtained from contrasting feedstocks pyrolyzed at 300 °C and 750 °C. Our research has revealed that the choice of feedstock and, notably, the pyrolysis temperature play pivotal roles in determining the total and mineral N content within the final biochar. Biochars produced at 750 °C exhibited low levels of N, with most of them existing in aromatic forms. In contrast, biochars that underwent pyrolysis at 300 °C effectively preserved a significant portion of the N, whose content can be predicted by the N content of its precursor feedstock. As pyrolysis conditions intensified from 300 to 750 °C, C compounds were selectively preserved within the biochar. At the same time, N was lost at higher rates, the magnitude of which relies on the type of feedstock. The preservation of total N and mineral N in biochars derived from castor oil cake, chitosan, chicken manure and shrimp carcass pyrolyzed at 300 °C holds promise for providing N for crops and reducing reliance on fertilizers in the formulation of biochar-N composites. Biochars derived from woody, bamboo, and sugarcane feedstocks, pyrolyzed at 750 °C, are potential sources for storing aromatic C in soils.

Introduction

Brazil heavily depends on imported N fertilizers to support its agricultural practices [1]. The significant amounts of industrial waste, manure, and crop residues generated in the country are important for nourishing crops and formulating fertilizers. Some organic wastes have elevated levels of sodium, heavy metals (Cu, Zn, As, Cr, etc.), salts, pathogenic bacteria, and antibiotics. Thus, they require chemical and biological stabilization before being utilized as soil conditioners or sources of nutrients for crops [2,3,4,5,6]. Composting is used for processing organic waste, but it has drawbacks, including high N volatilization, an extended period to produce humified compost, emissions of greenhouse gases and volatile organic compounds, and the production of toxic slurry that can contaminate soil and water bodies [6,7,8]. Pyrolysis is employed to stabilize organic waste by rapidly charring feedstocks at temperatures ranging from 300 to 750 °C in an oxygen-depleted atmosphere [4, 9, 10]. The composition and properties of biochar are influenced by pyrolysis conditions (temperature, heating rate, retention time, etc.) and type of feedstock [11, 12]. When these factors are considered together, they create a complex interplay that significantly affects the properties and ability of biochar to interact with soil, biota, plants, water and air [11,12,13]. Different feedstocks, such as wood, crop residues, manures, fish wastes or algae, possess distinct chemical composition and structural properties [13, 14]. The choice of feedstock determines the biochar's elemental composition, potential contaminants and capacity to deliver nutrients or work as a soil conditioner [11, 13].

Meanwhile, the pyrolysis temperature sets the stage for the transformation of these feedstocks, influencing the extent of carbonization, the intensity of nutrient volatilization and the presence of organic functional groups in the biochar [13]. High-pyrolysis temperatures generally decrease yield, polarity, CEC, etc. At the same time, it increases pH, C content, surface area, ash, and porosity, leading to the formation of stable aromatic compounds in wood, sugarcane, and plant residue-derived biochar [13, 15]. In contrast, some high-ash and lower-temperature biochars contain more polar organic functional groups, preserve non-volatilized nutrients in available forms, have high CEC, and have higher levels of bioactive water-soluble C [11,12,13]. The combined effect of pyrolysis temperature and feedstock serves as a toolkit for tailoring biochars with varying liming values, capacity to supply K and P to crops, N content, CEC, and ability to regulate the dynamics, forms, and availability of C and N in soil [11, 13, 16]. For high-ash feedstocks, the rise in pyrolysis temperature converts C into more recalcitrant and aromatic compounds while increasing the volatilization of O, H, S, and N chemical species, subsequently enhancing ash content, liming value, pH, oxides and carbonate content for some biochars [13, 17].

Feedstock selection and pyrolysis temperature are key factors regulating the content, pools, chemical composition, and availability of N compounds in biochar [11, 16, 18, 19]. During pyrolysis, a significant portion of the N in feedstocks is lost through volatilization, resulting in biochars with low N levels primarily existing in heterocyclic forms within high-temperature biochars [16, 20]. When biomass undergoes pyrolysis at lower temperatures, typically around 300 °C, a substantial amount of N remains in forms prevalent in the precursor feedstock [16]. According to Liu et al. [21], an increase in pyrolysis temperature promotes the conversion of labile N compounds (such as protein-N, free amino acid-N, and alkaloid-N) in rice straw into inorganic N forms (NH4+-N, NO2-N, and NO3-N) and recalcitrant organic N compounds, including nitrile, pyridine, amino acids, and pyrrole groups, within the resulting biochars. Compared to biochars produced at 800 °C, pyrolyzing rice straw at 400 °C reduces volatilization and preserves N to a greater extent in the derived biochar. In an evaluation of N retention in wood, flyboard, plant residues, and coal-derived biochars, Lang et al. [18] confirmed that all biomasses released at least half of their N as volatile compounds by 400 °C, with additional N release occurring up to 750 °C. According to Liu et al. [21], the optimal conditions for producing high-quality biochar for soil application and enhancing lettuce growth involve pyrolyzing rice straw at 400 °C in a CO2 atmosphere. With an increase in the pyrolysis temperature, hydrolysable N decreases, while the formation of stable N in aromatic structures increases in biochar produced from dairy manure and biosolids. Wang et al. [16] noted that pyrolysis induces N loss in the oil and gas streams and reduces N availability in biochar as charring conditions intensify. The presence of aromatic-like and heterocyclic N structures in high-temperature biochars demonstrates a greater N persistence in the soil compared to the more labile N forms present in the feedstock [16, 22, 23]. The rise in pyrolysis temperature and the use of extended residence time reduce the concentrations of NH4+ and NO3 in biochars, with ammonium loss due to thermal degradation being more pronounced than nitrate [12]. A lower N availability in high-temperature biochars is attributed to the non-labile character of the remaining N, primarily in the form of pyridine-N or pyrrole-N within the charred matrices [4]. Fast pyrolysis is a viable method for converting biomass into bio-oil, although it significantly reduces biochar yield. Conversely, the combined use of moderate heating rates and slow pyrolysis leads to the breakdown of compounds with weaker bonds while preserving compounds with stronger bonds and high chemical stability [12]. Heating rate and the type of feedstock have only a slight influence on the properties of biochars charred at temperatures exceeding 700 °C [24].

The N content in biochar is determined by the feedstock, following this decreasing order: wood residues < crop residues < manure and biosolid-derived biochars [11, 18, 19]. The N content in biochar depends on the N content and pools in the original feedstock [25]. For instance, the mean N content is, on average, 0.95% for wood residues, 1.54% for crop residues, 1.8% for grasses, and 2.42% for manure-biosolid wastes [11]. In addition, the N content in feedstocks can vary significantly across different categories of organic wastes and residues, ranging from 0.7 to 1% for woody biomasses, 0.2% to 1.4% for agricultural and crop residues, 4% to 12.2% for animal and human wastes, 0.4% to 8% for industrial wastes, and 8.2% to 10.6% for seaweed materials [26]. However, not all N sources have the same capacity to generate high content and available N chemical species in the final biochar [27]. Organic wastes enriched with proteins and amino acids often release NH3 and other nitrogenous gases during pyrolysis [28], with less N available for crops in the final biochar. Regardless of the N source, significant amounts of various N chemical species (e.g., NO, N2O, NO2, NH3, and N2) are lost as the pyrolysis temperature increases [11, 18]. Hydrolysable and labile N in biochar decreased as the pyrolysis temperature increased from 250 to 550 °C [16], while recalcitrant, aromatic, heterocyclic and graphitic N forms become dominant in biochars produced at temperatures above 600 °C [16, 29, 30]. The retention of N in biochar decreases from 82 to 10% for feedstocks charred at temperatures from 300 to 1100 °C [18]. Nitrogen losses typically begin at temperatures above 150 °C due to the thermal degradation of hemicellulose and cellulose fractions. As the thermal degradation process intensifies from 250 to 500 °C, the volatilization of N compounds is greater [30]. In the temperature range of 300 °C and 500 °C, the prevalent N chemical species in the feedstocks and charred matrices are proteins, pyridinic, pyrrolic, amino acids, and amine forms, while in the range of 500–800 °C, inorganic forms (NO2, NO3, and NH4+) and highly recalcitrant heterocyclic N forms predominate in the final biochars [16, 29, 30]. As pyrolysis conditions become more intense, losses of N in the form of oil and gases increase, leading to a sharp decrease in N availability in the biochar Nitrogen compounds and functional groups (such as nitric oxide, pyridinic N, pyrrolic N, and graphitic N) present in biochars derived from different feedstocks and pyrolysis conditions have been comprehensively reviewed by Baser et al. [27]. Consequently, the content and nitrogen (N) chemical species present in the charred matrices depend on the feedstock, although higher pyrolysis temperature plays a prominent role in determining the final content, pools and availability of N in biochars. Overall, the feedstock and pyrolysis conditions, particularly the intensity of charring conditions, are key factors controlling the elemental composition and the availability of N chemical species in the resulting biochar [16, 18, 19, 31, 32].

Organic waste materials such as manure, fish, sewage sludge, oil cakes, and shrimp shells can be effectively utilized in producing biochar-based fertilizers. After pyrolysis, these wastes can serve as short-, medium-, and long-term N sources for crops [16, 33]. It is essential to investigate how the choice of feedstock type and pyrolysis temperature influences biochar properties, besides assessing its potential as a reliable N source for crops or as raw material for the synthesis of N fertilizers. Nitrogen-rich feedstocks decompose rapidly in soils, potentially releasing CO2, NH3, and N2O into the atmosphere. Converting these valuable wastes into biochar is an innovative strategy for sequestering C in the soil [11], besides enhancing the supply of N throughout the crop growth cycle. However, limited information is currently available regarding N content and changes in N availability when high-N feedstocks undergo pyrolysis. Both pyrolysis temperature and the type of feedstock significantly influence the N content, pools, and availability in biochars [16, 34]. In this study, we assess the N content and forms in biochars derived from wood, agricultural, food, and fishery industrial wastes produced on a large scale in Brazil. These feedstocks underwent pyrolysis at temperatures of 300 °C and 750 °C. The aims of this study were as follows: (i) Investigate the properties of biochars manufactured from contrasting feedstocks charred at 300 °C and 750 °C, including yield, pH, EC, C and WSC content. (ii) Identify the effects of feedstock type and pyrolysis temperature on total and mineral N content, C and N retention, and aromatic character of biochars. (iii) Determine the optimal pyrolysis temperature for maximizing the retention and the availability of N in biochars while considering the feedstock type. We hypothesized that by carefully selecting feedstock and choosing the appropriate pyrolysis temperature, we can produce biochars with high N content and availability.

Materials and Methods

Feedstock and Pyrolysis Conditions

The Laboratory for the Study of Soil Organic Matter (LEMOS) at the Department of Soil Science, Federal University of Lavras (UFLA) in Lavras, Minas Gerais State, Brazil, conducted the production and characterization of feedstocks and biochars. We used eleven organic wastes with varying nitrogen (N) contents and chemical compositions to produce biochars with different properties. The selected feedstocks included residues with:

  • Low N content (pine bark, eucalyptus sawdust, bamboo, and sugarcane bagasse),

  • Intermediate N levels (coffee husk and olive cake),

  • High N contents (such as chicken manure, shrimp carcass, chitosan, sunflower, and castor oil cakes).

Table 1 displays the main nutrient pools, C and N content, and the C/N ratio of these feedstocks. We dried these feedstocks at 60 °C in an air circulation oven until they reached a constant weight. To preserve bioactive components such as vitamins, polyphenols, flavonoids, glycosides, N volatile compounds, and low molecular weight organic acids, it is strongly recommended using a convective, long-term drying process at relatively low temperatures (< 70 °C) [35], even though, in this study, the biochar yield calculation is based on the mass of feedstock subsamples dried at 105 °C. After drying, we macerated and sieved the feedstocks using a 2 mm mesh before storing them for further analysis.

Table 1 The main nutrient contents and pools, and C/N ratio of feedstocks utilized to produce the biochars

We used stainless steel cylinders to compact the feedstocks, which were then pyrolyzed in a Jung (N1200) muffle furnace at two target pyrolysis temperatures: 300 °C and 750 °C. Depending on the type and bulk density, we used 150–320 g of feedstock to produce the biochars. The slow pyrolysis process involved a heating rate of 10 °C min−1, and we maintained the feedstocks at the target temperature for 60 min. Subsequently, we cooled the resulting biochars to room temperature and stored them in a desiccator for subsequent analysis. Both the feedstocks and biochars underwent maceration in a mortar and sieving using a 0.5 mm mesh before full characterization in the laboratory. The biochar yield was calculated using the following formula [13]:

$${\text{Biochar}} \;{\text{yield}} \left(\%\right)=\left[\left({\text{MfMi}}\right) \times 100\right],$$

where Mf is final biochar mass, Mi is feedstock initial mass at 105 °C.

Electrical Conductivity and pH

The pH and EC were determined using a biochar (or feedstock) to water ratio of 1:10 (w/v). To achieve this, 1 g of feedstock or biochar was added to a Falcon tube and mixed with 10 mL of deionized water. The tubes were then shaken on a horizontal shaker at 100 rpm for 1 h. The pH measurement was conducted using a bench-top pH meter (Hanna Instruments, HI2221), while the EC measurement was performed using a conductivity meter (Mettler Toledo, FE30). All measurements were carried out in triplicate, following the ASTM analytical protocols [36].

Biochar N Content and C and N Retention

The Kjeldahl method was employed to ascertain the total N content in the biochar, involving three stages: digestion, distillation, and titration. During the digestion phase, 0.1 g of either feedstock or biochar was placed into digestion tubes, followed by the addition of a 1 g catalyst mixture (K:Cu:Se at the ratio of 100:10:1) and 3 mL of sulfuric acid. The digestion process commenced with the block digestion system gradually reaching temperatures of 200 °C over 30 min, followed by an additional 30 min at 200 °C and a final 60-min digestion at 350 °C. After completing the digestion of both feedstock and biochar samples, they were cooled to room temperature, and each tube received 20 mL of deionized water.

Subsequently, nitrogen distillation equipment (TECNAL-036/1 model) was utilized until 50 mL of distillate accumulated in the Erlenmeyer flask. Titration was performed to determine the N content, present as ammonium borate, in both biochars and feedstocks. A 0.07143 mol L−1 HCl solution was used in the titration phase to convert borate into boric acid.

N retention (Eq. 1) and C retention (Eq. 2) in biochar were calculated using the following formulas, adapted from Lang et al. [18]:

$${\text{Biochar N retention }}\left( \% \right){\text{ = }}[\left( {{\text{N in biochar }}\left( \% \right) \div {\text{N in feedstock }}\left( \% \right)} \right) \times {\text{biocharyield}}(\% )]{\text{,}}$$
(1)
$${\text{Biochar C retention }}\left( \% \right){\text{ = }}[\left( {{\text{C in biochar }}\left( \% \right) \div {\text{C in feedstock }}\left( \% \right)} \right) \times {\text{biocharyield}}(\% )]$$
(2)

Mineral N in Biochar

The determination of exchangeable N-NH4+ and N-NO3 content follows the methodology proposed by Bremner and Keeney [37]. In a Falcon tube, dried, ground, and sieved samples (1.0 g) were mixed with 10 mL of a 1 mol L−1 KCl solution, which acts as an extractant for ammonium and nitrate. The tubes were then shaken for one hour on a horizontal shaker at 100 rpm. After the extraction of mineral N in feedstock and biochar, the samples were filtered, and 5 mL of the filtrate was transferred to a distillation tube. To determine the N-ammonium (NH4+) content, 10 mL of distilled water were added to each sample along with 0.2 g of magnesium oxide (MgO). Subsequently, to convert N-nitrate (NO3) into N-ammonia (NH3), 0.2 g of Devarda's alloy was added to the same distillation tube. In both cases, ammonia (NH3) from the distilled samples was collected in an Erlenmeyer flask containing 10 mL of a boric acid solution. The ammonia was converted into ammonium borate, which was then titrated and converted into boric acid using a 0.07143 mol L−1 HCl solution. After titration, the content of N as ammonium and nitrate was determined.

Total and Water-Soluble Carbon

The total carbon content of the feedstocks and biochars was determined by combusting 0.5 g of finely ground and sieved (200 mesh) biochars in a dry combustion TOC analyzer in the solid mode (Vario TOC cube, Elementar, Germany). Water-soluble organic C (WSC) was measured by shaking a 10% (w/v) ratio of feedstock or biochar–water mixture for one hour and filtering it through a 0.45 μm pore membrane filter. The WSC contents in the liquid extracts were quantified using the liquid mode of a dry combustion TOC analyzer (Vario TOC cube in the liquid mode, Elementar, Germany). It is important to note that a single one-hour extraction may not solubilize and extract all organic C from the biochar matrix. Therefore, WSC provides only a partial indication of water-soluble C forms rather than determining all readily available and soluble C forms present in the feedstock and biochar samples [13].

Biochar Aromaticity Index

The aromaticity index of the biochar was calculated based on the spectral signatures obtained from the FTIR spectra and specific peaks. The infrared spectrum was initially recorded using the Fourier transform infrared–attenuated total reflectance (ATR–FTIR) spectroscopy technique on an Agilent® Cary 630 spectrometer. The spectrum covered the wavenumber range of 4000 to 650 cm−1 with a resolution of 4 cm−1. The normalization procedure recommended by Gautam et al. [38] was applied to pre-process the dataset from each FTIR spectrum. Subsequently, the biochar aromaticity index was determined by analyzing the areas of the peaks recorded at 1630 cm−1 and 2929 cm−1, which correspond to the ratio between the areas of the C=C and C–C functional groups in each FTIR spectrum. The same mathematical procedures and equations used by Dores-Silva et al. [39] were employed to calculate the aromaticity index of the biochars produced in this study.

Statistical Analysis

The means of feedstock and biochar attributes were grouped using the Scott–Knott test (p < 0.05) after ensuring that the basic assumptions of analysis variance (normality, homoscedasticity, additivity, and independence of residuals) were met, and the significance of the F test was achieved (p < 0.05). Principal component analysis (PCA) was conducted to assess the relationship between the attributes of feedstocks and biochar samples produced at 300 °C and 750 °C. Pearson's linear correlation matrix (p < 0.05) was also performed to validate the relationships and cluster the variables shown in the PCA diagram. To cluster the attributes of feedstock and biochar samples charred at each target pyrolysis temperature (300 °C and 750 °C), the dataset was scaled, and cluster analysis was carried out based on the matrix of Euclidean distances among samples using Ward's algorithm hierarchical clustering procedure [40]. Bootstrap support was calculated for each dendrogram branch.

All statistical analyses were conducted using R software with the following packages: stats, ExpDes.pt, corrplot, factoextra, FactoMineR, and pvclust [41,42,43,44,45,46]. The clustering procedure was based on the properties and spectral signature (aromaticity index based on ATR–FTIR spectra) of the feedstock and biochar (300 °C and 750 °C). For clustering the feedstock and biochar samples, the content or module of each attribute was normalized by dividing them by the respective C content in the sample, thereby removing the influence of the different C content of each attribute. The normalized attributes were then used to calculate a matrix of Euclidean distances among samples, which was employed for hierarchical clustering of biochar and feedstock using Ward's algorithm [40]. The pvclust package was used to calculate the bootstrap support for each branch in the resulting dendrogram, as depicted in the graphs generated in this study, following the procedures described by Suzuki and Shimodaira [45].

Results and Discussion

Influence of Pyrolysis Temperature on Biochar Properties

We assessed how two pyrolysis temperatures (300 °C and 750 °C) influence biochar properties [12, 13, 20, 35]. Higher temperatures result in increased carbon (C) content, especially in plant and woody feedstock-derived biochars, due to the removal of volatile components [11, 13]. Elevated temperatures also enhance surface area and pore volume, improving water and nutrient retention [11,12,13]. Pyrolysis temperature affects pore distribution, with higher temperatures generating more micropores and lower temperatures producing more mesopores [11, 13].

Lower-temperature biochars retain more surface organic functional groups, enhancing reactivity and cation exchange capacity, CEC [11, 13, 35]. Higher temperature biochars are more stable and less reactive, rich in aromatic C compounds [13]. Biochar pH tends to increase with higher temperatures, but liming capacity isn't solely determined by pH [13]. Higher temperatures may reduce nutrient content, especially volatile organic compounds [11, 13]. They can also increase ash content, particularly in nutrient-rich feedstocks [11, 13], while decreasing the presence of organic contaminants like PAHs [19].

Pyrolysis temperature influences surface charge and CEC, affecting adsorption capacity for pollutants, organic compounds, and nutrients, with implications for soil remediation, water treatment, and agriculture [13]. It also plays a role in biochar's suitability as an N source for crops [13, 20]. Supplementary Table 1 summarizes the overall impact of temperature and the feeds on ash contents, which will be further discussed in subsequent sections.

Biochar Yield

Biochar yield primarily relies on pyrolysis temperature and, secondly, on the feedstock type (Table 2). The composition of the feedstock, particularly ash content, substantially influences biochar yield, especially for biochars generated at high temperatures (> 600 °C) [13]. Shrimp carcass and chicken manure pyrolyzed at both 300 °C and 750 °C yield higher biochar quantities compared to other feedstocks, likely due to their elevated ash contents. Remarkably, chicken manure with high ash content produced biochar with the highest yields. Chicken manure biochar yield was only reduced by 15.8% when the pyrolysis temperature was increased from 300 to 750 °C. Inorganic constituents present in the ash of chicken manure-derived biochars act as flame retardants, reducing mass loss during pyrolysis [47]. Consequently, biochar yield is more strongly correlated with its ash content than volatile matter [48, 49], as ash catalytically affects biomass charring, retarding the rate of thermal degradation of organic compounds. During pyrolysis, the ash remains in the biochar, contributing to its mass and increased biochar yield [13, 50]. The preservation of organic compounds, even at high pyrolysis temperatures, may be associated with the elevated ash content found in coffee husk and chicken manure, where ash acts as a shield, safeguarding labile organic groups and fragments against heat-induced thermal degradation as charring conditions intensify [13, 51].

Table 2 Biochar yield as affected by feedstock and pyrolysis temperature

The elemental composition of shrimp carcass likely contributed to the elevated yields of biochar generated at 300 °C despite the shrimp carcass-derived biochar experiencing considerable thermal degradation when the temperature was raised to 750 °C. Pyrolysis at both 300 °C and 750 °C resulted in the lowest biochar yields for sugarcane bagasse, bamboo, and eucalyptus sawdust, possibly explained by their composition, such as low nutrient, salt, and ash content. These feedstocks produce biochars with lower yields at both temperatures. Further studies should comprehensively assess both the inorganic components and the organic matrix, including those functional groups of biochars derived from chicken manure and sugarcane bagasse charred at 300 °C and 750 °C. In contrast, shrimp carcass exhibited the highest mass loss (− 41%) when the pyrolysis temperature increased from 300 to 750 °C.

Wood feedstocks, such as tree residues, typically have lower sulfur (S) and nitrogen (N) contents but higher lignin content compared to crop residues [50]. Lignin, a complex aromatic polymer, is more resistant to thermal degradation than cellulose and hemicellulose, which are the predominant components of crop residues. Consequently, when rich-lignin biomass undergoes pyrolysis, a significant portion of the feedstock may not be fully converted into biochar, leading to reduced biochar yields, although with no net emissions of CO2 [50]. Rich-lignin biomass may necessitate even higher temperatures to achieve complete carbonization, which can decrease the biochar yield and result in more residual ash in the final biochar [13].

pH and EC Value

The carbonization process induces alkalization in the charred matrix, leading to significant pH differences between biochar and its precursor feedstock, depending on the pyrolysis temperature (Table 3). In this study, some biochars produced at 300 °C and 750 °C exhibited pH variations exceeding five units compared to their respective feedstocks. This rise in biochar pH with higher pyrolysis temperatures is attributed to increased ash content and the enrichment of basic cations [11, 13]. Most biochars pyrolyzed at 300 °C had pH values above 6, except for eucalyptus sawdust-derived biochar (pH 4.3) and sugarcane bagasse-derived biochar (pH 5.7). Biochars pyrolyzed at 750 °C typically had pH values greater than 8. Biochar pH is expected to increase with thermal degradation of organic compounds and the enrichment of inorganic constituents as the temperature rises [13]. The prevalence of acidic functional groups in biochar decreases as charring conditions intensify [52,53,54]. In high-temperature (> 600 °C) biochars, ash content, particularly from inorganic constituents, plays a more crucial role in determining pH than aromatic compounds [55]. For feedstocks rich in salts, carbonates, potassium (K), calcium (Ca), and magnesium (Mg), increased ash content occurs with more intense charring conditions, impacting pH changes during pyrolysis and the biochar liming capacity [13].

Table 3 Values of pH and EC (dS m−1) of feedstocks and their derived biochars produced at 300 °C and 750 °C

As pyrolysis temperature rises, the loss of weakly bonded hydrogen and oxygen linked to organic compounds leads to an increase in carbonates, especially those containing Ca and Mg, sometimes originating from feedstock inorganic constituents or CO2 released and trapped in the alkaline charred structure [53, 54]. The content and type of carbonates in biochar influence its liming capacity and the biochar organic matrix stability [56].

Biochar's surface functional groups are primarily influenced by the precursor feedstock and pyrolysis temperature [57]. Increasing pyrolysis intensity raises biochar pH, ash content, surface area, and porosity while reducing the presence of bioactive, polar compounds [58, 59]. Oxygen-containing acidic functional groups contribute to biochar acidity at lower pyrolysis temperatures (< 400 °C), while higher temperatures lead to increased biochar pH due to reduced oxygen content [59]. However, pH alone does not reliably predict biochar's capacity to neutralize toxic Al and correct soil acidity [13]. Even though wood residues (PB, ES, and bamboo) produce biochars with higher pH values at 750 °C, they exhibit reduced liming value [13], as liming equivalence depends on feedstock type and ash chemical species [13, 57]. The biochar pH is weakly correlated with CaCO3 equivalent (1.5% to 20%) [58], and properties such as CaCO3 content, cation types; the amounts in the ash rather than pH are primary indicator of biochar's ability to correct soil acidity and promote plant growth [13, 57].

Electrical conductivity (EC) varies among different feedstocks and biochars, ranging from 0.1 to 31.3 dS m−1 (Table 3). Pine bark, eucalyptus sawdust, and sugarcane bagasse exhibit the lowest EC values, below 1.0 dS m−1, due to their high lignin content and low salt concentrations. In contrast, sunflower cake and its biochar pyrolyzed at 300 °C have high EC values, while chitosan-derived biochar (pyrolyzed at 750 °C) displays the highest EC among charred samples. Feedstocks with initially high EC values, such as chicken manure, show the least variation in EC as charring intensifies. Decreased EC in sugarcane bagasse-derived biochar produced at 750 °C results from the thermal degradation of inorganic components and the preservation of aromatic C compounds. The addition of high-ash biochars to the soil may increase ion concentration and soil EC, potentially impacting salt-sensitive crop growth [60]. EC reflects the quantity and nature of salts in the soil solution [61], with variations attributed to calcite, carbonate, and silvite in different manures and their corresponding biochars. Potassium (K), calcium (Ca), magnesium (Mg), and sodium (Na) salts in biochar ash strongly influence EC, especially in biochars produced at temperatures exceeding 650 °C [62]. Biochar EC values reported in the literature range from 0.04 to 54.2 dS m−1 [63, 64], primarily reflecting feedstock composition and pyrolysis conditions. Ash-rich biochars produced at higher temperatures typically exhibit higher EC compared to low-nutrient feedstock-derived biochars [25, 65, 66]. In contrast, wood and paper waste-derived biochars tend to have a lower EC compared to nutrient-rich manure-derived biochars [50, 63, 67]. For example, biochars from milk and feedlot manure have lower EC values than those derived from poultry litter [25]. The electrical conductivity of biochar exhibits a complex interplay influenced by various factors encompassing feedstock composition, pyrolysis conditions, and resulting biochar characteristics. Different feedstocks comprise varying proportions of organic and inorganic materials, which exert an influence on the electrical conductivity of the resultant biochar. The ash content found within biochar, stemming from the mineral content of the feedstock, can significantly impact its electrical conductivity. Elevated ash content may contribute to heightened conductivity, as ash frequently contains minerals that possess electrical conductivity properties. Diverse feedstocks may necessitate specific pyrolysis conditions to achieve optimal biochar production, and deviations in these conditions can yield disparities in electrical conductivity. Furthermore, the surface area and pore structure of biochars play pivotal roles in determining electrical conductivity. The type of feedstock can affect the development of pores and the overall surface area, thereby affecting the biochar's capacity to conduct electricity. Some feedstocks may inherently contain elements that enhance electrical conductivity. The degree of carbonization or charring during the pyrolysis process can also fluctuate among feedstocks, resulting in biochars with varying electrical conductivity values. Notably, electrical conductivity serves as a valuable tool for evaluating nutrient release kinetics from nitrogen mineral fertilizers and predicting the readily available nitrogen (N) content in biochar-based fertilizers [16].

Total C and N Content

The total carbon (C) contents in biochars derived from woody, bamboo, and sugarcane feedstocks increased as the pyrolysis temperature rose (Fig. 1a). In contrast, the N concentrations in these biochars remained relatively stable (Fig. 1b). Biochars produced at 750 °C from pine bark (PB), eucalyptus sawdust (ES), bamboo, and sugarcane bagasse (SB) exhibited the highest C content, reaching up to 96% C (Fig. 1a). This increase in C concentration with rising pyrolysis temperature is attributed to several factors, including enhanced polymerization, losses of hydrogen (H), oxygen (O), nitrogen (N), and sulfur (S) during the pyrolysis process, as well as an enrichment of aromatic and more condensed carbon structures within the biochar [11]. A similar relative enrichment of carbon has been reported in biochars derived from lignified and low-N feedstocks, such as pine straw [60], peanut shells [61], sugarcane bagasse [62], and wheat straw [63]. As the degree of formation of aromatic structures increases, so does the resistance of the biochar to decomposition by microorganisms, resulting in longer persistence of biochar carbon compounds in the soil [11, 64, 65].

Fig. 1
figure 1

Total content of carbon (a), total content of nitrogen (b) and C/N ratio (c) of biochars as related to feedstock and pyrolysis temperature. Means with standard error of the biochars derived from different feedstocks, within the same pyrolysis temperature, with the same uppercase letter, and those for the two pyrolysis temperatures, within the same feedstock, followed by the same lowercase letter are not statistically different, based on the Scott–Knott test (p < 0.05)

Conversely, the carbon concentration in chicken manure (CM) biochar decreased as the pyrolysis temperature increased, which is consistent with findings reported by Domingues et al. [13]. This observation suggests that the organic matrix of animal waste may be more labile and rapidly lost as the pyrolysis temperature rises, occurring before the formation of biochar with recalcitrant carbon compounds [13]. The carbon concentration in CM biochar was lower than that in wood and SB derived biochars (Fig. 1), consistent with the results reported by Ippolito et al. [11]. The analyzed feedstocks exhibited varying nitrogen (N) content, spanning from 0.2 to 8.9%. Notably, olive and castor oil cakes, chicken manure, shrimp carcass, and chitosan had the highest N levels. In contrast, pine bark, eucalyptus sawdust, bamboo, and sugarcane bagasse had N contents below 1%, with coffee husk at around 3.5% (see Table 1). Pyrolysis caused significant N losses, particularly at 750 °C, where biochars had N contents 64% to 89% lower than their feedstocks. Regardless of the feedstock, biochars pyrolyzed at 750 °C had higher aromatic character but were less suitable as nitrogen sources for crops due to low N content and limited mineralization potential of heterocyclic N compounds. Nitrogen losses during pyrolysis exceeded carbon losses, primarily due to the formation of gases. Biochar carbon content increased with pyrolysis temperature due to various reactions. Some feedstocks show total N content reductions exceeding 50% in biochars produced at 750 °C, unlike carbon enrichment, which was observed.

Most feedstocks show an increase in their total nitrogen (N) content when biochar is pyrolyzed at 300 °C (as depicted in Fig. 1b). At 300 °C, the extent of nitrogen (N) losses is strongly influenced by the feedstock type. Most biochars produced at 750 °C undergo a reduction of more than 50% in their total nitrogen content. In contrast to carbon (C), there is no nitrogen enrichment in feedstocks subjected to pyrolysis at 750 °C. Pyrolysis at 300 °C can relatively enrich biochars with nitrogen (as shown in Fig. 1c).

The carbon-to-nitrogen (C:N) ratio of biochar reflects the impact of both the feedstock and pyrolysis temperature on their carbon (C) and nitrogen (N) contents, highlighting the dissociation between losses of C and N compounds during pyrolysis. The C/N ratio serves as a valuable indicator for evaluating the capacity of biochar-treated soils to mineralize or immobilize nitrogen (N), especially in short-term incubation studies. Unlike carbon (C) compounds, nitrogen (N) chemical species are more susceptible to loss during pyrolysis, particularly in biochars produced at 750 °C. To optimize pyrolysis conditions for various applications, such as using biochar as a source of nitrogen (N) or for carbon (C) sequestration in soil, it is crucial to carefully select the appropriate feedstock and, specifically, the pyrolysis temperature (either 300 °C or 750 °C).

D. Pyrolysis at 300 °C preserved most N in biochars derived from chicken manure, shrimp carcass, and chitosan. Further research should thoroughly investigate nitrogen (N) losses and the factors that can minimize nitrogen (N) volatilization during pyrolysis at 300 °C and 750 °C, particularly for nitrogen-rich feedstocks.

Biochars produced at 300 °C exhibit a higher cation exchange capacity (CEC) than those charred at 750 °C. Conversely, biochars produced at 750 °C have a higher anion exchange capacity compared to biochars charred at 300 °C [13, 16].

The nitrogen content and N chemical species in biochars are dictated by the choice of feedstock and its initial N content, with a general trend of increasing N content observed as follows: wood residues < crop residues < manure and biosolid-derived biochars [11, 18, 19].

Nitrogen levels can vary considerably among different feedstocks and organic waste materials. For instance, typical average N content values include 0.95% N for wood residues, 1.54% N for crop residues, 1.8% N for grasses, and 2.42% N for biochars derived from manure and biosolid wastes [11]. When considering various categories of waste and organic residues, N content varies across a wide range: 0.7–1% for woody biomass, 0.2–1.4% for agricultural and crop residues, 4–12.2% for animal and human waste, 0.4–8% for industrial waste, and 8.2 to 10.6% for aquatic waste, primarily in the form of algae raw materials [26].

It is important to highlight that not all sources of N possess the same capacity to generate high levels of available N chemical species in the resulting biochar [27]. As previously noted by Ippolito et al. [11], the total N content (ranging from 0.94 to 2.42%) in biochars is influenced by various pyrolysis conditions such as temperature, retention time and heating rate, as well as the type of feedstock used. At lower temperatures (< 450 °C), the choice of feedstock has a significant impact on the total N content of biochar, following a decreasing order as follows: wood biochars < crop residue biochars < grass biochar < biochars derived from manure and sewage sludge [11]. Biochars with higher N content are typically obtained from feedstocks rich in amino acids and proteins [67]. Consequently, biochars enriched in nitrogen, such as those derived from manure and sewage sludge, are more effective as nitrogen sources for crop applications [25]. In summary, biochars produced at high temperatures (> 600 °C) exhibit low N content, making them unsuitable for crop nitrogen sources. As the pyrolysis temperature increases, substantial N losses (> 60%) occur in gaseous compounds (e.g., NO, N2O, NO2, NH3) [11], primarily due to the prevalence of aromatic chemical structures and N-heterocyclic rings in matrices charred at temperatures exceeding 650 °C [16, 20]. Regardless of the feedstock type used, pyrolysis at 750 °C is not conducive to producing biochars with the potential to supply significant nitrogen to crops, mainly because of the considerable reduction in both total N content and biochar yield. The findings also indicate that the total N content of biochars produced at 300 °C and 750 °C can be reasonably estimated based on the N content of their precursor feedstock (Fig. 2).

Fig. 2
figure 2

The linear relationship between the total N content in the feedstock with the total N content in the biochars pyrolyzed at 300 °C and 750 °C. *This symbol means the equation coefficient is significant at p < 0.05

Biochar C and N Retention

The retention of carbon (C) in biochars depends on the feedstock used and is notably influenced by the pyrolysis temperature (Fig. 3a). Generally, biochars pyrolyzed at 300 °C retain more C compared to those pyrolyzed at 750 °C. Interestingly, biochars derived from feedstocks enriched with nitrogen (N), such as chicken manure (CM), shrimp carcass (SCa), chitosan (Chi), and coffee husk (CC), are highly effective at retaining C, preserving around 80% of it when pyrolyzed at 300 °C. At 300 °C, nearly all of the C present in pine bark is preserved in its resulting biochar. In general, C retention falls within the range of 61% to 73% for eucalyptus sawdust (ES), bamboo (Bam), sugarcane bagasse (SB), chicken manure (CH), olive cake (OC), and shrimp carcass (SC)-derived biochars when pyrolyzed at 300 °C. However, as the pyrolysis conditions intensify, the C retention sharply decreases in biochars pyrolyzed at 750 °C. Among the feedstocks, shrimp carcass (SCa)-derived biochar is the least effective in retaining C at 750 °C. At 750 °C, pine bark (PB), olive cake (OC), and chicken manure (CM)-derived biochars are the most effective matrices in retaining around 60% of C as the charring process intensifies. Further studies are needed to elucidate the mechanisms involved in C preservation from nutrient-poor and lignified feedstocks (PB) and those with high N levels and ash content (CM and OC). More detailed information on the mechanisms and processes related to C losses during pyrolysis can be found in the work conducted by Lang et al. [18].

Fig. 3
figure 3

Biochar carbon (a) and nitrogen (b) retention as affected by feedstock and pyrolysis temperature. Means with standard error of the biochars derived from different feedstocks, within the same pyrolysis temperature, with the same uppercase letter, and those for the two pyrolysis temperature, within the same feedstock, followed by the same lowercase letter are not statistically different, based on the Scott–Knott test (p < 0.05)

Similarly, in line with the results for C retention, nearly all the nitrogen (N) contained in pine bark is preserved in its biochar when produced at 300 °C (Fig. 3b). Most feedstocks and their resulting biochars retain N in the range of 60% to 75% when pyrolyzed at 300 °C. However, at 300 °C, biochars derived from sugarcane bagasse (SB) and olive cake (OC) are less effective at preserving N in their charred structures. The differences in the resistance of C compounds to thermal degradation, the selective preservation of C compounds over N chemical species (which are more susceptible to volatilization) during the pyrolysis process, the presence of C in aromatic structures, and the relative enrichment of C over hydrogen (H), oxygen (O), sulfur (S), and N compounds in charred matrices are the main mechanisms and explanations for the greater preservation of C than N in biochars, particularly in those pyrolyzed at 750 °C.

Biochar Mineral N Content

The total nitrogen (N) content in biochar varies depending on the feedstock and, more importantly, the pyrolysis temperature. Total N contents can range from 0.1 to 9.5% across different temperatures and feedstocks. The levels of ammonium-N in biochar vary based on the feedstock, with chicken manure and sunflower cake exhibiting the highest levels (as shown in Fig. 4a). However, pyrolysis significantly reduces ammonium-N levels, especially at a temperature of 750 °C, resulting in lower availability of ammoniacal N in the biochar. Interestingly, for shrimp carcass, pyrolysis at 300 °C retains more ammonium-N than the precursor feedstock. Similarly, feedstocks like pine bark and eucalyptus sawdust show higher N-nitrate levels in biochar pyrolyzed at 750 °C, compared to 300 °C.

Fig. 4
figure 4

Nitrogen as ammonium (a) and nitrate (b) contents in feedstock’s and their respective biochars produced at 300 °C and 750 °C. Means with standard error of the biochars derived from different feedstocks, within the same pyrolysis temperature, with the same uppercase letter, and those for the two pyrolysis temperature, within the same feedstock, followed by the same lowercase letter are not statistically different, based on the Scott–Knott test (p < 0.05)

Nevertheless, there is a substantial reduction in N-nitrate levels with increasing pyrolysis temperature for feedstocks charred at 750 °C. Overall, the total N content in biochar decreases due to significant thermal degradation, and mineral N availability in biochar decreases as pyrolysis conditions intensify. Both N-ammonium and N-nitrate contents are greatly diminished in biochars compared to the precursor feedstock when pyrolyzed at 300 °C and 750 °C.

The proportion of mineral N (N-ammonium + N-nitrate) in total N undergoes changes with pyrolysis temperature, but the extent of thermal degradation of mineral N relies on feedstock type (see Fig. 5). Increasing the pyrolysis temperature from 300 to 750 °C enriches mineral N in woody-derived biochars relative to the total N lost, indicating that the losses of available N forms are relatively smaller compared to the overall N pools in the feedstock. In most cases, the fraction of N in the form of ammonium and nitrate remains relatively consistent as the pyrolysis temperature is increased (as depicted in Fig. 4b). However, proportionally, the remaining N in bamboo, sugarcane bagasse, and, particularly, shrimp carcass-derived biochars is significantly reduced, although a portion of total N remains in biochars with high N content charred at 750 °C. This reduction in readily available mineral N (ammonium + nitrate) in biochars with increasing pyrolysis temperature is consistent with findings reported by Wang et al. [16].

Fig. 5
figure 5

Percentage of total N as mineral N (ammonium + nitrate) as a function of feedstock and pyrolysis temperature. Means with standard error of the biochars derived from different feedstocks, within the same pyrolysis temperature, with the same uppercase letter, and those for the two pyrolysis temperature, within the same feedstock, followed by the same lowercase letter are not statistically different, based on the Scott–Knott test (p < 0.05)

Water-Soluble Carbon

Water-soluble carbon (WSC) levels varied significantly among different feedstocks, with coffee husk, sugarcane bagasse, and sunflower cake reaching up to 100 g kg−1, 90 g kg−1, and 160 g kg−1, respectively (refer to Fig. 6). When considering increasing WSC concentrations among feedstocks, the order is as follows: eucalyptus sawdust (ES) < chitosan (Chi) < pine bark (PB) < bamboo (Bam) < chicken manure (CM) < shrimp carcass (SC) < coffee husk (CC) < olive cake (OC) < sugarcane bagasse (SB) < coffee husk (CH) < shrimp carcass (SCa). In most cases, biochars, except those derived from shrimp carcass (SC), exhibited a significant reduction in WSC content when pyrolyzed at 300 °C compared to their precursor feedstock. Biochars derived from nitrogen-rich feedstocks (SCa, SC, CC, CM, and SC) yielded higher WSC concentrations (at 300 °C) than those produced at 750 °C.

Fig. 6
figure 6

Water soluble C in biochars as related to feedstock and the target pyrolysis temperature. Means with standard error of the biochars derived from different feedstocks, within the same pyrolysis temperature, with the same uppercase letter, and those for the two pyrolysis temperature, within the same feedstock, followed by the same lowercase letter are not statistically different, based on the Scott–Knott test (p < 0.05)

Water-soluble carbon (WSC) constitutes a substantial portion of the organic content in biochar [5, 69]. The pyrolysis temperature plays a critical role in controlling the presence of water-soluble C in biochars, with a notable decrease observed as the pyrolysis conditions become more intense, as previously reported by Domingues et al. [13]. For most biochars, pyrolysis at 750 °C is not an effective method for retaining soluble C within the charred matrices. The visual appearance and color intensity of extracted WSC solutions often strongly correlate with biochar WSC concentrations [5]. Understanding the presence of water-soluble C in biochar is essential for comprehending the dynamics and transformations of biochar C in soil, as well as elucidating the complexation and transport of contaminants and nutrients throughout the soil profile [5]. Both total C content and soluble carbon are critical factors in determining whether N added to the soil via biochar will be immobilized or undergo net mineralization, converting organic N species into available mineral N. The introduction of inputs rich in soluble C and N chemical species to the soil is expected to influence soil organic matter decomposition rates (positive or negative priming effect), microbial biomass, and microbial activity [68, 69]. However, it's worth noting that Jiang et al. [70] found that biochar decomposition rates in soils can vary, and the presence of N or low molecular weight soluble C compounds in biochars is not the primary factor influencing the biochar's ability to be processed by biota or its persistence or capacity to sequester C in the soil.

Principal Component Analysis

Principal component analysis (PCA) was employed to assess the influence of feedstock type and pyrolysis temperature on biochar properties, aiming to condense the extensive dataset into a more accessible graphical representation featuring distinct clusters. The division in PCA diagrams according to pyrolysis temperature was determined by the significant impact of pyrolysis conditions on carbon (C) and nitrogen (N) content and distribution within the biochar, as well as the unique attributes of each feedstock in relation to its corresponding biochar. The primary axes of the principal component analysis (PC1 + PC2) accounted for a majority of the variation in feedstock properties (65.3%) (see Fig. 7a). Feedstocks with high lignin content and low nutrient concentration, such as eucalyptus sawdust, pine bark, sugarcane bagasse, and bamboo, exhibited elevated C:N ratios (greater than 100:1) and total carbon content. Conversely, shrimp carcass demonstrated high nitrogen and water-soluble carbon content, along with elevated pH and electrical conductivity (EC) values. Chicken manure exhibited the highest availability of nitrogen in the form of nitrate and ammonium (refer to Fig. 6a). For biochars pyrolyzed at 300 °C, the PCA components elucidated 80.2% of the variation (PC1 + PC2) in the primary biochar attributes. Shrimp carcass, chicken manure, and chitosan-derived biochars displayed high pH and total nitrogen content. Sugarcane bagasse, eucalyptus sawdust, and bamboo-derived biochars had elevated total carbon content, C:N ratios, and a predominance of mineral nitrogen in the form of nitrate (see Fig. 7b). Principal component analysis conducted on the dataset for biochars produced at 750 °C accounted for 75.3% of the variation (PC1 + PC2) in biochar properties. In the PCA analysis of biochar pyrolyzed at 750 °C, feedstocks with substantial lignin contents (pine bark and eucalyptus sawdust) yielded biochars characterized by an increased aromatic nature, C:N ratios surpassing 200:1 and low total nitrogen content, with a portion of it predominantly existing in mineral nitrogen forms (see Fig. 6c). Biochar (750 °C) derived from shrimp carcass, chitosan, and chicken manure exhibited the highest EC, pH, total nitrogen content, and water-soluble carbon (refer to Fig. 7c), highlighting their potential as nutrient sources and soil conditioners with a significant liming capacity.

Fig. 7
figure 7

Principal component of the main attributes of feedstocks (a) and biochars produced at 300 °C (b) and 750 °C (c). CSW the content of water-soluble C, C-total total C content, CSW:C-Total the % of water-soluble C as related to total C content, NO3 the content of N-nitrate, NH4+ the content of N-ammonium, N-mineral contents of N-mineral (nitrate + ammonium), EC electrical conductivity, NH4+:NO3 ratio ammonium:nitrate ratio, NH4+:N-Total the content of N-ammonium as related to total N content, NO3:N-Total content of N-nitrate as related to total N content, N-mineral:N-Total content of N-mineral as related to total N content, AI aromaticity index, C:N ratio total C and N content ratio

Elevating the pyrolysis temperature resulted in increased pH levels, higher electrical conductivity (EC) values, greater total carbon content, and higher C: N ratios. Feedstocks with a high nitrogen (N) content produced biochars with reduced levels of water-soluble carbon, mineral nitrogen, and total nitrogen. Nevertheless, biochars pyrolyzed at 750 °C still exhibited certain characteristics inherited from the original feedstock despite the predominant influence of high pyrolysis temperatures on determining their aromatic nature and chemical/physicochemical properties. The elemental composition and attributes of the biochars no longer resembled those of the initial feedstock. Both carbon (C) and nitrogen (N) contents in the biochars experienced significant declines with increasing pyrolysis temperature. However, the loss of nitrogen (low N retention in biochar) was relatively more pronounced than the thermal degradation of carbon, especially as the charring process intensified from 300 to 750 °C.

Clustering the Feedstock and Biochar Features

We conducted a hierarchical clustering analysis to identify the main characteristics of feedstocks and biochars based on their chemical, physicochemical, nutrient contents, and aromaticity index. The results are presented in the form of a dendrogram in Fig. 8a–c, where feedstocks and biochars produced at 300 °C and 750 °C are grouped.

Fig. 8
figure 8

Cluster analysis of the normalized attributes of feedstocks and biochars (300 °C and 750 °C) showing the percentage of similarity within each group formed (G I, G II, G III, G IV, G V). Feedstock dried at 60 °C (A), Biochar 300 °C (B), and Biochar 750 °C (C). CSW content of C soluble in water, C-T total C content, \(\frac{CSW}{C-T}\) the content of water-soluble C as related to total C content, NO3 the content of nitrate, NH4+ content of ammonium, N-m contents of N-mineral (nitrate + ammonium), EC electrical conductivity, \(\frac{{NH}_{4}^{+}}{{NO}_{3}^{-}}\) ammonium:nitrate ratio, \(\frac{{NH}_{4}^{+}}{N-T}\) the content of ammonium as related to total N content, \(\frac{{NO}_{3}^{-}}{N-T}\) the content of nitrate as related to total N content, \(\frac{N-m}{N-T}\) content of N-mineral as related to total N content, AI aromaticity index, C:N total C:N content ratio, SB Sugarcane bagasse, Bam bamboo, CH coffee husk, SCa shrimp carcass, PB pine bark, CM chicken manure, Chi chitosan, ES eucalyptus sawdust, OC olive cake, SC sunflower cake, CC castor cake. The radar charts at the tip of terminal branches display the ratio of value variable for each feedstock or biochar normalized by means of the maximum value of the analyzed variable. The amplitude of the normalized attributes ranges from 0 to 1 (maximum across all samples) for each target pyrolysis temperature (biochars) and feedstocks dried at 60 °C

Feedstocks (Fig. 8c) were divided into 6 clusters based on various attributes, including N content, mineral N forms, total and water-soluble C, aromaticity index (determined by FTIR-ATR spectrum peak ratio), EC, and pH. Group I (Fig. 8a) consisted of a single sample (Shrimp carcass) characterized by high EC, medium pH, high aromaticity index, and a high ratio of ammonium to nitrate. Group II included sunflower cake, characterized by high total N content, water-soluble C, and elevated pH. Group III included castor cake and chitosan, distinguished by high total C, total N, and pH. Group IV included OC, Bam, CH, and SB, with high total C content, reduced levels of both total and available mineral N, lower pH, aromaticity index, and EC. Group V consisted of chicken manure, characterized by high availability of N as ammonium and nitrate and elevated pH. Group VI included lignin-enriched feedstocks (pine bark and eucalyptus sawdust) with high total C content, high C:N ratio, lower pH, and a greater contribution of mineral N to total N in these biochars.

Biochars produced at 300 °C were categorized into four groups (Fig. 8b). Group I predominantly included lignified materials such as ES, PB, Bam, and SB, characterized by high C:N ratios, elevated total C content, and a high proportion of total N as nitrate. Group II consisted of a single sample (SC-derived biochar) with low total C content, high EC, elevated availability of mineral N, and high total N content. Group III included a single sample (SCa) enriched with water-soluble C, total N, predominance of nitrate as mineral N, high pH, aromaticity index, and medium total C content. Group IV included five samples (CH, CM, OC, Chi, and CC) distinguished by higher pH and medium levels of total N and C compared to other feedstock-derived biochars.

Biochars produced at 750 °C were separated into three clusters (Fig. 8c), indicating that the main features of biochars are more dependent on the pyrolysis temperature rather than the feedstock type. Group I included lignified and nutrient-poor biochars, such as PB, with high total C content, elevated C:N ratio, low total N content, and high aromaticity index. Group II included five samples (SC, OC, CC, Bam, and SB-derived biochars) characterized by intermediate levels of C and N, aromaticity index, and pH values. Group III included biochars produced from SCa, CM, CH, and Chi, mainly distinguished by high EC, higher content of water-soluble C content, total N, pH, and the proportion of total N as nitrate-N.

In our study, we focused on certain N-functional groups present in biochars without fully investigating the specific N-functional group or organic compounds that influence the dynamics and fate of other nutrients found in poor- and N-enriched biochars. Further research should also evaluate the complexation and adsorption of polar and apolar compounds by biochars added to soils [28].

Biochars produced at 300 °C exhibit a higher cation exchange capacity (CEC) and a prevalence of N in the ammonium form within the charred matrices. In contrast, biochars charred at 750 °C have nitrate as the dominant form of mineral N. Considering that nitrate is a stable and non-volatile form of nitrogen at the high pH of biochar produced at 750 °C, this type of biochar appears suitable for synthesizing biochar-based N fertilizers (BBNFs). Further investigation into the diffusion of nitrate into biochar micropores as a mechanism for nitrogen retention at high temperatures is needed. On the other hand, biochar charred at 300 °C, with its lower pH, higher CEC, and greater density of polar organic functional groups, is suitable for composites containing N in the ammonium form. These composites can be indirectly supplied by urea, MAP, ammonium sulfate, and similar N sources. Biochars produced at high temperatures, such as 750 °C, have the ability to persist and remain stable in the soil for an extended period. Consequently, they can immobilize and sequester carbon and nitrogen while also retaining more water, thus minimizing the off-site environmental impacts associated with N loss in the soil–water–plant system [20]. Further research should investigate the patterns (kinetics) of N release from biochars produced at 300 °C and 750 °C to classify these novel N sources as fast- or slow-release fertilizers capable of adequately nourishing crops in the short, medium, and long terms.Analyzing biochar properties reveals their potential environmental and agronomic applications, especially their role in providing nitrogen (N) for crops. We focused on diverse feedstocks, finding that low-N wood-based feedstocks pyrolyzed at 300°C are unsuitable N sources, while nitrogen-rich feedstocks pyrolyzed at 300°C are effective in supplying N to crops. The ratio of N in feedstock to biochar remains 1:1 at 300°C, but only around 50% of the N in feedstocks is retained in biochars generated at 750°C. Biochar N retention depends on yield, not N content. Woody biochars at 750°C may be useful as adsorbents for nonpolar soil compounds. However, low-ash and high-aromatic biochar from lower temperatures are not suitable to produce biochar-N composites. Biochars derived from nutrient-rich feedstocks also possess greater polarity, rendering them suitable for adsorbing ammonium. Biochars produced at high temperatures and with low ash content are less effective at adsorbing ammoniacal nitrogen compounds due to a reduction in organic functional groups and polarity. Eucalyptus sawdust, pine bark, and sugarcane bagasse biochars charred at 750°C demonstrate potential for adsorbing nonpolar compounds. The water-soluble carbon and mineral nitrogen contents are influenced by the feedstock, and the pyrolysis temperature impacts the degradation of water-soluble carbon and mineral nitrogen availability. Additionally, the pyrolysis process enriches biochar with nitrate over the ammonium form.

Conclusions

In our study, we observed significant changes in biochar properties as the pyrolysis temperature increased from 300 to 750 °C. Most biochars showed reduced yield, decreased N content, lower C and N retention, and a decline in water-soluble C. In contrast, pH, electrical conductivity, and the aromaticity index increased with higher pyrolysis temperatures. Biochars produced from woody sources such as pine bark and eucalyptus sawdust, as well as bamboo and sugarcane bagasse, were found to be unsuitable as N sources for crops.

Regardless of the pyrolysis temperature used, biochars with low ash content and lignified characteristics exhibited high concentrations of carbon, making them potential contributors to aromatic carbon storage in soils. On the other hand, biochars derived from chicken manure, shrimp carcasses, chitosan, and castor oil cake, when produced at 300 °C, displayed moderate to high N content (ranging from 4 to 9.1%), retained over 60% of their N, and provided adequate available N (NH4+ + NO3: 190 to 470 mg kg−1). These biochars have the potential to serve as valuable N sources for crops.

The agronomic value of biochars produced at 300 °C in supplying N to crops is closely linked to the N richness of the feedstock used for biochar production. While the N content in biochar produced at 300 °C closely resembles that of its precursor feedstock, N retention varies from 33 to 77%, and it is lower than the C retention in both low- and high-temperature charred matrices. The ratio of feedstock N to biochar N content is approximately 1:1, with roughly half of the feedstock's N content being lost when pyrolyzed at 750 °C. Therefore, we can reasonably predict the N content in biochars, both at 300 °C and 750 °C, based on the N content of their precursor feedstock. As pyrolysis conditions become more intense, from 300 to 750 °C, C and N losses are not directly correlated, and N chemical species are lost at a greater rate than C compounds. This shift is accompanied by increases in pH, electrical conductivity, C content, and the aromatic character of biochars. Woody feedstocks, particularly pine bark, and low-nutrient feedstocks such as eucalyptus sawdust, bamboo, and sugarcane bagasse, pyrolyzed at 750 °C with C retention exceeding 50%, are better suited for long-term aromatic carbon storage in soils. Future research should concentrate on evaluating the effectiveness of nitrogen-enriched biochars in delivering N to crops throughout their growth cycle. Additionally, investigating N pools within woody, sugarcane bagasse, and bamboo biochars can be essential in assessing their persistence in soil and their capacity to supply long-term N to crops. Elucidating the mechanisms associated with the thermal resistance of specific N pools present in chitosan-derived biochar pyrolyzed at 750 °C is also recommended. The results presented in this study offer valuable insights for designing biochars tailored to meet specific crop N requirements and for the development of N-based biochar fertilizers. These objectives can be achieved through careful feedstock selection and adjustments in pyrolysis temperature as needed.