Physico-chemical characterization of walnut shell biochar from uncontrolled pyrolysis in a garden oven and surface modification by ex-situ chemical magnetization

All glassware was washed in distilled water. About 50 g of BWS0 was first oven dried at 105 °C for 24 h and then treated with FeSO₄⋅7H₂O, Fe₂(SO₄)₃⋅6H₂O and distilled water at room temperature (between 21 and 23 °C). The detailed scheme is shown in Fig. 1. By altering the amount of incorporated iron salts, three different magnetic WS biochars were prepared (Table 1).

Mid-IR spectra between 4000 and 400 cm⁻¹ were collected with Nicolet iS50 (Thermo Fisher Scientific Corporation) having a KBr beam splitter, diamond crystal, and deuterated triglycine sulfate (DTGS) detector. The measurement parameters of optical velocity, gain, and aperture size were 0.1581, 2.0, and 87, respectively. A total of 32 scans at resolution 2 cm⁻¹ were collected for each sample and averaged. The ATR peak intensities were dependent on the refractive index of the crystal and sample (Volkov et al. 2021), which could be changed after magnetization. Hence, peak intensities were not used for semi-quantitative estimation of the amount of each functional group present.

The elemental concentration and fiber content in the walnut shells (WS), Juglans regia L., are shown in Table 2. WS is primarily composed of lignin and α-cellulose, and only 12 wt.-% of hemicellulose. Similar compositions of WS is also reported in refs (Yuan et al. 2020; Kambarova and Sarymsakov 2008). However, in some other investigations (Altun and Pehlivan 2012; Senol 2021), higher hemicellulose concentrations similar to that of α-cellulose are found. Such differences arise from climatic and plantation conditions, walnut cultivars, and, to an extent, differences in measurement methodologies. In general, WS has one of the highest lignin concentrations (30 to 50%) compared to other crop residues and woods (Pirayesh et al. 2012) due to the lignification of the walnut endocarp. The presence of higher lignin and cellulose contents indicates a higher quantity of sclereids and increased shell thickness during fruit development. In our material, the shell thickness of WS is ~ 1–1.2 mm. As a pyrolysis feedstock, WS differs from the mineral- and ash-rich waste biomass (MWB) that the present authors have previously investigated (Nair et al. 2020) because WS has a) higher O/C molar ratio due to the phenolics in lignin and cellulose, b) higher organic carbon content, c) smaller concentration of AAEM, silicates, and ash, which lower the chances for catalytic pyrolysis, and d) trace amounts of Fe (0.014 wt.-%).

Pyrolysis of WS has reduced the VM and increased the FC in BWS0. Compared to biochar from mineral-rich feedstocks (Isaeva et al. 2021), BWS0 has only 3.41 wt.-% ash (AC). The results from proximate analysis and TOR_i of WS and its biochar are shown in Table 3 while their elemental concentrations (CHNOS and Fe) in dry wt.-% are presented in Table 4.

Magnetic biochar has higher ash concentrations due to the presence of iron oxides. The Fe content in biochar follows the same trend as the amount of iron salts induced during chemical co-precipitation—BWS1 > BWS2 > BWS3. However, the ash concentration in BWS3 is larger than that for BWS2, which indicates possible impurities in BWS2 (to be discussed later). Magnetic biochars also possess larger VM release from the biochars and are observed as the increase in the DTG peaks of magnetized biochar compared to BWS0 at ~ 800 °C (Figure S2). The DTG peak of BWS3 during combustion is smaller than that of BWS2 due to lower ash content in BWS3. FC (ash-free basis) has also comparatively reduced after the incorporation of iron oxides. Hence, as seen in other studies (Zhao et al. 2013; Leng et al. 2019; Wurzer and Masek 2021), FC cannot be used as an indicator of the degree of carbonization in biochar when it undergoes post-pyrolysis chemical impregnations. With BWS0 as the reference, the TOR_i of all magnetized biochars are near 0.45 owing to an increase in their thermal oxidization stability relative to the pristine BWS0. However, the changes in the amount of iron infused do not seem to considerably change its oxidative stability.

The pyrolysis of walnut shell yields 28.5 wt.-% BWS0 with a total carbon and organic carbon of 90.14 and 81.98 wt.-%, respectively. BWS0 also has low hydrogen and oxygen, showing a good degree of carbonization and polycondensation of the material. The higher heating value of BWS0 calculated according to elemental (Channiwala and Parikh 2002) and proximate measurements (Parikh et al. 2005) are 31.99 and 32.14 MJ/kg, respectively. This high HHV is similar to that of Anthracite despite being produced from UCPy. The intrinsic accumulation of Fe in the unmagnetized BWS0 is small (0.16 wt.-%). BWS0 has a H/C_org and O/C_org molar ratio of 0.161 and 0.06, respectively. This fits within the definition (Eq. (2)) of biochar by the International Biochar Initiative (IBI) and the European Biochar Commission (EBC) (Budai et al. 2013; EBC 2012).

The carbon sequestration potential (CSP) (Sieng-Huat 2019) of BWS0, calculated according to Eq. (4) is 0.84 kg_CO2/kg_biomass for a BC₊₁₀₀ of 98.23% (Eq. (3)). BC₊₁₀₀ is the percentage of sequestered carbon remaining in the soil after 100 years. It can mitigate the N₂O emissions from the soil by ~ 73 ± 7% (Cayuela et al. 2015). For H/C_org ⩽ 1, this molar ratio can be correlated with an n*n aromatic cluster size (Eq. (5)), which predicts that BWS0 can have up to a maximum of 11-ring polyaromatic compounds, which is common for cellulose and lignin-rich materials at high HTTs (Xiao et al. 2016). It is not possible to draw such inferences from H/C_org molar ratio for magnetized biochar because of the changes in elemental concentrations (Fe, S, and H) introduced by the chemical co-precipitation. BWS0 has 92% aromatic carbon as measured from ¹³C-NMR CP spectra between 165 and 110 ppm (Figure S3). Details of this ¹³C-NMR analysis are in supplementary information.

The O/C is inversely proportional to hydrophobicity and (N + O)/C is the polarity index, which increases with surface polarity (Zhao et al. 2019; Fan et al. 2018). These are shown in Fig. 2. Hydrophobicity is an important parameter relevant to biochar application. Adding hydrophobic biochar to soil can reduce its wettability, and water retention capacity, creating a preferential flow of contaminants to groundwater sources (Mao et al. 2019). Usually, low-temperature biochars (< 500 °C) are hydrophobic due to the presence of aliphatic functional groups, while at temperatures > 500 °C, hydrophobicity reduces due to the loss of these functionalities and the polarity arising from recalcitrant C-O, C = O and nitro groups on the turbostratic carbon, which assists in the formation of hydration clusters with water molecules. At higher HTTs, these oxygen and nitrogen functionalities also decompose due to polycondensation, and the biochar’s affinity for water is limited to the adsorption on the well-developed micropores. BWS0 is more hydrophobic with a low O/C ratio compared to other pristine WS biochars produced through controlled pyrolysis at 600 and 700 °C as described in (Gupta et al. 2019; Lahijani et al. 2018). Magnetization resulted in a general increase in the hydrophilic nature due to the O-functionalities incorporated during the chemical co-precipitation. Hence, if land-applied, the magnetized variants may increase the water retention capacity of sandy and drought-prone soils (Suliman et al. 2017).

Unlike organic litter, a significant portion of the carbon, nitrogen, and phosphorus in biochar is recalcitrant. Some microbes, depending on conditions, can also mine energy and nutrients from such recalcitrant aromatic forms (Zhang et al. 2020a). Long-term incubation experiments would be required to quantify the fraction of recalcitrant carbon that is accessible by various types of soil microbes. Hence, in this study, to determine soil applicability from biochar properties, we consider that only the C, N, and P in the VM of biochar are immediately accessible to microbes. Therefore, labile C, N, and P in biochar were evaluated. There is no labile and readily plant-available phosphorous (P) in these biochars since P usually forms stable aromatics at HTT nearing 700 °C (Xu et al. 2016). However, BWS0 surface is negatively charged (seen in "Physical and Structural Properties of Biochar" section) at a pH of 9.45. Thus, it can adsorb Fe and Al ions, if present in the soil, to reduce their complexation with soil P; ergo not influencing the labile pool of P that may be present in the soil prior to biochar application. Under the assumption of homeostasis and carbon-controlled microbial communities, a higher soil loading rate of BWS0 can raise the net soil C/P (usually ~ 290) (Ch’ng et al. 2019) above the threshold elemental ratio of P (TER_C:P) creating a situation of P deficiency and energy (carbon) excess. TER_C:x for a nutrient x defines the elemental ratio at which microbial activity switches from C limitation to a deficiency in x (Guo et al. 2020). This can incite microbial phosphatase synthesis by species from the genera Pseudomonas, Bacillus, and Rhizobium, and consequently improve P-mineralization (Pérez et al. 2007). In the long term, while C leaves the soil in the form of CO₂, mineralized P remains, and/or is turned into a plant-available form through microbial necromass. The absence of labile P is also beneficial to prevent runoff and subsequent eutrophication.

The molar labile C/N is 24.8, 62.3, 24.0, and 32.7 for BWS0, BWS1, BWS2, and BWS3, respectively. The soil molar C/N ratio for soil ranges from 10.2 to 30.2 (Tate 2020), with 24 considered a “balanced diet” (Brust 2019). Hence, BWS0 addition will not decrease the carbon use efficiency (CUE) as compared to other amendments rich in organic C. Magnetic biochar, on the other hand, is acidic with more Fe in it for soil P complexation. Thus, its application can make the soil P unavailable to plants. The C/N of BWS2 and BWS3 is also near the aforementioned optimal range for soil and is more hydrophilic than BWS0. However, their iron concentrations of > 5 dry wt.-% warrant further investigation into Fe speciation, and microbial toxicity.

WS shows strong signals in the IR region between 3500 and 2500 cm⁻¹ (Fig. 3a). As seen in ref (Shah et al. 2018), these peaks mainly include the –OH stretching from lignin and carbohydrates near 3400 cm⁻¹, and –CH stretching vibrations at ~ 2930 cm⁻¹. These features disappear in the BWS0 (Fig. 3b) sample due to a reduction in ν_(OH), ν_(NH), and ν_(CH) vibrations (Singh et al. 2017) after carbonization. Figure 3c shows the comparative fingerprint region of IR spectra (between 1800 and 400 cm⁻¹) for the four biochars. After carbonization, the samples have lost most of the bands due to ν_(C–O) at ~ 1030 cm⁻¹ that stem from polysaccharides. The peaks near 550 cm⁻¹ and 634 cm⁻¹ are from Fe–O stretching (Wan et al. 2020; Jack et al. 2019) of maghemite. Their relative peak heights change in proportion to the Fe content seen in elemental analysis—BWS1 > BWS2 > BWS3. These peaks are absent in unmagnetized BWS0. Since the biochars are hydrogen-poor carbons (H/C < 0.3) (Russo et al. 2014) the presence of aliphatic carbon is quite weak and is only seen as a broad ridge between 1400 and 1000 cm⁻¹. The small peak at 670 cm⁻¹ is from the Fe–O bond in magnetite. Trace amounts of silicates and phosphates do not contribute much to the region around 1000 cm⁻¹. The peak at 1600–1500 cm⁻¹ is from ν_(C=O) of oxygenated functional groups (OFGs) such as carboxylic and carbonyl (Singh et al. 2017) that are acidic. The reactor environment during pyrolysis did not possess an active N₂ purge. This supports cracking reactions of tar and char in presence of the evolved CO₂ and CO at high temperatures to remove more OFGs as compared to pyrolysis in a thermogravimetric reactor with N₂ purge (Hong et al. 2019). This would also explain the low O/C of BWS0 described above. The peaks at 1100 and 610 cm⁻¹ present in BWS3 correspond to asymmetric stretching and bending vibrations of SO₄, respectively. As also later seen in PXRD, Na₂SO₄ is present as an impurity in BWS3. The peaks between 900 to 700 cm⁻¹ arise from substituted aromatics.

Table 5 summarizes the concentrations of heavy metals and the amount of AAEM in the biochar and the elemental retention rate in BWS0. The concentrations of Cd, Co, Pb, and Se are negligible and below their detection limits. The concentrations of Ni, Zn, and Cu are below their respective lower guideline values (LGV) for abatement of ecological and health risks (Toth et al. 2016). Chromium in BWS0 is slightly above its LGV of 0.2 mg/g due to an enrichment of 1505%. Such enrichment of non-volatilizable elements is directly proportional to the highest treatment temperature and is unavoidable during pyrolysis. Carbonization has also increased Al, Bi, Ni, and Sr concentrations since these elements transformed into stable oxides or sulfides under 700 °C. Magnetic biochar has diluted heavy metal concentrations, ergo they lie below their LGV limits. In the case of AAEM, K is ~ 100% retained due to the lack of Cl providing a volatilization route during the active pyrolysis while a portion of Mg, Na, and S are lost. It is worth noting that the heavy metal enrichment in WS-based biochars is still considerably less than that from MWB like sewage sludge and manure (Liu et al. 2015; Zuo et al. 2020). Also, only the speciation of Cr can reveal whether it is phytotoxic or not accessible within the carbon matrix.

Figure 4 shows the Py-GC–MS chromatogram of BWS0, the mean and standard deviation of the concentration of the gases and compounds as a percentage of the total products evolved during analytical pyrolysis. Carbon dioxide and benzene constituted about 95.5 and 3.7% of the total chromatogram area, respectively. The detected PAH are naphthalene, biphenyl, and phenanthrene. These PAH, which are < 7–rings, are biodegradable (Rombolà et al. 2016) and can undergo biological or chemical oxidation as biochar ages.

The PXRDs and Raman spectra of the four biochar samples are shown in Fig. 5 a and b, respectively. The broad peaks in the PXRDs at 2θ of 23.5° (002) and 43.8 (101) indicate the presence of strongly disordered sp² amorphous carbon as that is expected from non-graphitized biochar. Diffraction peaks of γ-Fe₂O₃ can be found at 2θ of 30.1° (220), 35.5° (311), 43° (400), 57° (511), and 63° (440) (Golubev et al. 2019) in the magnetized biochar samples. For BWS3, in addition, the PXRD peaks of Na₂SO₄ are found which is in line with the FTIR results that indicated the presence of sulfate ions in this sample. Na₂SO₄ can be created during the co-precipitation process when using NaOH for initiating the precipitation process.

During carbonization, the intensity of the D band (I_D) at 1350 cm⁻¹ in the Raman spectra represents defects in aromatic ring structures (which are absent in perfect graphite) while the G band (I_G) at 1590 cm⁻¹ indicates (locally) ordered structures of carbon. The D* band arises from sp³-bonded carbon atoms and the D** band is associated with sp²-bonded carbon atoms as well as fragments or functional groups in the disordered structure (Yin et al. 2018). The I_D/I_G ratio of BWS0, BWS1, BWS2, and BWS3 are 0.79, 0.74, 0.67, and 0.71, respectively. Based on these ratios of I_D/I_G, the average size of ordered graphitic regions (“cluster size”) can be calculated as the coherence length (L_a) according to the Eq. (6).where λ_L is the Raman excitation wavelength (532 nm); C(λ_L) is a wavelength-dependent factor; and C_o and C₁ are constants (-12.6 nm and 0.0333, respectively) when 400 nm < λ_L < 700 nm (Matthews et al. 1999). The L_a of BWS0, BWS1, BWS2, and BWS3 are 6.50, 6.91, 7.69, and 7.16 nm, respectively. For nanocrystalline graphene structure, I_D/I_G increases with the introduction of defects while for sp² amorphous carbon it decreases at higher defect density (Childres et al. 2013). BWS0elongs to the latter as seen from the XRD spectra. The magnetized biochars have more defects than BWS0 owing to the introduction of maghemite. Most of the magnetite formed from co-precipitation oxidizes to maghemite during the formation of magnetic carbon nano-particles. The absence of magnetite in Raman spectra (~ 690 cm⁻¹) and FTIR spectra confirms this.

The SEM micrographs depict the porous nature of BWS0 and the impregnation of iron oxide particles in the magnetic variants (Figure S4). Low concentrations of AAEM are not distinguishable in the images. The magnetic biochars are superparamagnetic (Fig. 6) with the ratio of magnetic remanence to saturation magnetization (M_s) of less than 25% (Cazetta et al. 2016). The M_s is less than 1 emu/g for all cases. This is lower than magnetite which tends to form during such chemical co-precipitation. The M_s of BWS3 seems to be an outlier since it is independent of the amount of iron introduced during synthesis. However, the magnetic WS biochar can be still effectively separated from the aqueous solution in the presence of an external magnetic field (Figure S5).

The specific surface areas (S_BET) and pore volume (V_pore), as derived from nitrogen sorption measurements, of the biochars are shown in Table 6. The S_BET of BWS0 is comparatively larger than other WS biochar in literature (Lahijani et al. 2018; Kaya et al. 2020) which have been prepared through controlled pyrolysis at 700 °C. Expectedly, it is lower than those materials that have also undergone activation or washing as seen in refs (Qin et al. 2021; Wan et al. 2021).

Usually, there is an inflexion temperature (Leng et al. 2021) beyond which the porosity of biochar reduces with an increase in HTT due to pore collapse (from char melting, graphitic stacking, and pore fusion) or widening into macropores (reducing microporous surface area and increasing pore volume). The inflexion point for BWS0 obviously only occurs at temperatures > 700 °C (Lahijani et al. 2018; Qin et al. 2021). It is worth noting that this micropore collapse theory, though widely used for biochar derived from CP_y in inert environment, has its origins in activated carbon prepared under oxidizing conditions, which usually does not show an open-loop hysteresis curve during N₂ physisorption (Maziarka et al. 2021). In this study, despite the partial oxidation conditions, an open-loop hysteresis exists (seen later). Therefore, to confirm the existence of an inflection point due to pore collapse, a detailed investigation of possible adsorption-induced pore deformation is required. However, here, this would not be beneficial to the BWS0 synthesis because, in the case of an UCP_y, it is not possible to optimize biochar production based on inflection point.

The S_BET (r > 0.99) of BWS0 is higher (292 m²/g) than the corresponding values of its magnetized counterparts, probably from pores blocked by iron oxide particles in the latter. These particles are introduced into the mesopores of the biochar during chemical co-precipitation. Hence, after the incorporation of iron oxides, the contribution from mesopores to the total surface area and pore volume reduces as seen in Fig. 7. The V_pore of magnetized biochars is higher than BWS0 due to the contribution by mesopores pores (2–50 nm) in maghemite.

The IUPAC’s adsorption isotherms (Thommes et al. 2015) are used as a reference for interpreting the adsorption–desorption curves of biochar. The isotherms of the biochar in Fig. 8 are a blend of Type I (Langmuir) and Type IV. The steep rise at low relative pressures (P/P_o) for all samples indicates that during pyrolysis the micropores have developed. The adsorption corresponding to higher P/P_o shows the presence of mesopores, the amount of which varies with the amount of magnetizable material levels (Yang et al. 2016; Spokas et al. 2014). The N₂ adsorption also takes place in the larger mesopores introduced by the iron oxides and thus is higher in magnetized biochars. The larger hysteresis between the adsorption and the desorption branch introduced in magnetic biochars between 0.1 and 0.9 P/P₀ is also evidence of mesopores. The desorption time scale is much larger than that of the experiment and causes open-loop hysteresis, which may be caused by steric restrictions at the openings of micropores (Nguyen and Bhatia 2009). Such open-loop hysteresis curves are also known (Maziarka et al. 2021) to be caused by the deformation (swelling/contraction) of micropores in materials such as activated carbon during N₂ adsorption. This can be confirmed if the hysteresis disappears during CO₂ physisorption. The surface area should reduce with increasing levels of iron content due to pore blocking by iron oxides. However, BWS3 is an outlier which may be due to the presence of Na₂SO₄ in BWS3 as seen in "Elemental and Fiber Analysis" section.

The ζ potential of biochar is another surface property that plays an important role in soil application and adsorption. For BWS0, ζ potential increases with a reduction in pH as shown in Fig. 9. The pH_pzc of BWS0 is 2.81, corresponding to an acidic surface. This is due to the absence of an inert N₂ purge to remove evolved gases away from the pyrolyzing substrate. Consequently, the biochar can undergo oxidation to form acidic surface functional groups such as carboxyl and carbonyl groups. Their dissociation in water leads to a ζ of − 44.5. Such a steep fall in pH_pzc is also seen during the oxidization of activated carbon surfaces (Álvarez-Merino et al. 2008; Jaramillo et al. 2010). BWS0 thus shows an electrostatic affinity for cations in a wide pH range between 2.8 and 9.45. The ζ potential of magnetic variants BWS1 (− 32.17 mV), BWS2 (− 18.34 mV), and BWS3 (− 39.92 mV) have not changed to the positive range despite the introduction of iron oxides (Long et al. 2019). Iron oxides impart positive charges through the formation of complexes with OFGs increasing the disorder of the carbon structure and inducing surface porosity; to a lesser extent by physical coverage of iron oxides on the biochar surface (He et al. 2020). The electric charge of magnetic biochar can be traced back to three causes: a) deprotonation of the acidic OFGs in the magnetized biochars offsets the positively charged iron moieties; b) the conjugated π-aromatic structure can act as electron donors or acceptors due to the polarization of basal plane near the edge defects created by the OFGs and c) contribution of negative charge from adsorbed sulfates, carbonates and phosphates to a small extent (Xu et al. 2020; Verma and Singh 2019). The stability and dispersibility of colloids increase when ζ potential < − 30 mV. So, except for BWS2, there exists a potential for biochar-induced groundwater contamination (through colloid-enabled transport of contaminants) from the sub-surface depending on the net pH and EC of the soil medium. The low pH_pzc and high pH of BWS0 enable applicability in a wide range of complex dispersion mediums, including as a sink for cations when added to most soils (pH 4.7–5.5) (Zhao et al. 2015).

The result from ion chromatography, DOC, and EC measurements, reflecting the substances leaching from the biochar, are shown in Table 7. The concentrations of leaching anions PO₄³⁻ and NO₃⁻ are very low because of the formation of their recalcitrant forms such as pyrophosphates (Sahin et al. 2017) and binding to the biochar matrix, respectively, and hence below the quantification limit of the instrument. As seen earlier, the investigated biochars have no readily plant-available P. EC is directly proportional to the concentrations of monovalent and divalent water-soluble ions and is high for BWS3 (1.22 dS m⁻¹) and BWS0 (0.2 dS m⁻¹).

The EC-value below 4 dS m⁻¹ for all biochar suggests that they will not increase soil salinity after land application (Carrier et al. 2012). In BWS0, Ca²⁺ and Mg²⁺ are lower than K⁺ and Na⁺ due to the lower solubility of the salts of the former cations. For the same reason, K⁺ concentration has decreased in all magnetized biochars except BWS3 as most of it was removed during the water washing stage of magnetization. This is also evident from the low total K content (Table 5). The Na⁺ concentration in magnetized biochar is from NaOH used during chemical co-precipitation. More washing with distilled water at the end of co-precipitation is recommended to reduce the sodium salt impurities in biochars after an ex-situ chemical magnetization. Mg²⁺ and Cl⁻ leaching is not influenced by magnetization. Among the magnetized variants, BWS3 has the most Ca²⁺ leaching, and overall Ca content. This type of water removability of alkali metals is also seen in ref (Kong et al. 2014).

The phosphate ions are water insoluble due to interaction with Fe and Ca (Rajapaksha et al. 2015) and the high pH of BWS0. These biochars are thus a slow-release P- source in the soil minimizing the eutrophication of freshwater sources. Owing to a lack of water-soluble nitrates and phosphates, in the short term, BWS0 might require complimentary amenders like compost. This has been seen in other biochars derived from lignin-rich precursors (Yang et al. 2015). However, it is worth noting that, nitrogen is not completely lost during pyrolysis. Some nitrogen is bound to the biochar matrix ("Elemental and Fiber Analysis" section) and can become bio-available with the aging of the biochar (Jamroz et al. 2020). Smaller dissolution concentrations of ammonium ions can support in reducing ammonification, and autotrophic nitrification rate, and alter soil urease activities; thereby lowering N₂O emissions. Alkaline pH increases water-soluble NH₄⁺ concentration (Zhang et al. 2022). These ions can be adsorbed on biochar through electrostatic interactions with the negative functional groups (Heaney et al. 2018). The introduction of iron sulfate salts increases the SO₄²⁻ leaching from the magnetized biochars. By increasing the biochar loading rate in soil its ionic strength can be raised and promote flocculation that was adversely affected by zeta potential ζ (Bakshi et al. 2014). The low EC-value and the low concentrations of Na⁺ and Cl⁻ and SO₄²⁻ facilitate BWS0 to be applicable in saline soil in arid regions at a higher loading rate (Yang et al. 2015).

About 0.4 to 0.9 × 10¹² kg of terrestrial DOC annually flows into the oceans (Nebbioso and Piccolo 2013). It affects the biosphere through chelation and solubility of metals, transport of organic pollutants, alterations in microbial growth rate, photochemical reactions in water, and nutrient cycling. Hence, the biosphere is sensitive to the DOC incorporation from anthropogenic sources such as biochar. The DOC of pyrolysis-derived biochar decreases at higher HTTs and is usually lower than that from manure (Singh et al. 2017; Peng et al. 2020). Here, DOC from magnetized biochar is considerably larger than that from BWS0 (19.75 mg L⁻¹) owing to the larger cation concentration in the latter. The low negative electric charge of BWS2 (ζ = -18.34 mV) also facilitates the adsorption of negatively charged DOC but these ionic bonds are easily disassociated in water compared to those of polyvalent cations such as Ca²⁺ (Setia et al. 2013). Hence, BWS2 leaches the most DOC among the magnetic variants. The DOC from all WS biochar is the least aromatic (SUVA₂₅₄ < 2 to 3 L mg⁻¹ m⁻¹) (Wang et al. 2013; Zhang and He 2015) compared to humic acids (Ateia et al. 2017). A decrease in SUVA₂₅₄ after magnetization confirms that the additional DOC that is leached from them is hydrophilic and easily biodegradable aliphatic carbon. This also agrees with the E2/E3 ratio, which indicates that the biochar-derived DOC from magnetized biochar is predominantly carbon of low molecular weight. BWS0 can lower the export of low molecular and humified DOC due to microporosity and π-π interactions on its carbon surface, respectively (Shimabuku et al. 2016). However, this would also lead to fouling (Greiner et al. 2018) during adsorption applications whereby the adsorption of a target pollutant in natural waters is affected by the background DOC.

Based on the investigated physico-chemical properties of WS-derived biochar, the following inferences may be drawn with respect to their applicability in environmental applications such as carbon sequestration, soil amelioration, and liquid-phase adsorption. BWS0 has a carbon sequestration potential of 0.84 kg CO₂/kg biomass. The carbon in BWS0 is 83% aromatic with a BC₊₁₀₀ of 98.23% and the PAH mainly consists of naphthalene, biphenyl, and phenanthrene, which are biodegradable. Magnetization tends to reduce the H/C_org, O/C_org, and BC₊₁₀₀, while increasing the oxidation stability.

The C/N of BWS0 is optimal for promoting microbial growth in soil. The loading rate of BWS0 should be high to counteract its dispersion in soil. It must also be mixed with N and P-rich compost or fertilizers. Heavy metal enrichment in biochar from WS is less compared to that from waste biomass from agricultural and industrial processes. The low DOC from unmagnetized biochar abates the risk of contaminant transport in soils (Sun et al. 2021) while their adsorption affinity toward DOC can reduce the eutrophication of aquatic reservoirs due to DOC leaching. These biochars poorly adsorb species like P oxy-anions and nitrates and may show relatively better adsorption capacity for cations like K⁺ and NH₄⁺. However, combined with the high porosity, the low ζ of BWS0 should improve the water holding capacity in most hydrophilic soils (Batista et al. 2018). BWS0 can lower the leaching of cationic plant nutrients and the alkalinity of acidic soils. It is beneficial to mix WS biochar with compost or other organic fertilizers since a) it is a slow-release source of N and P, b) has Cr above the LGV limit, and c) has higher colloidal stability. The π-π electron-donor interactions can improve its selectiveness for pollutants such as PAH in soils. Detailed correlation between biochar DOC characteristics and soil microbial community is still relatively unexplored in literature (Nobaharan et al. 2021) and must be investigated.

The WS biochar derived from the UCP_y in this investigation is not suitable for the adsorption of anionic adsorbates and can show preferential adsorption of DOC in aqueous solutions. Surface sorption of metal ions may be improved by washing and activating BWS0. This can remove impurities, create more oxygenated functional groups, and improve porosity. Adsorption of cationic contaminants such as As(V) or As(III) through electrostatic interaction is possible in a wide variety of solutions due to the low pH_pzc of BWS0. Compared to the magnetic WS biochar, BWS0 can show relatively good adsorption potential for hydrophobic contaminants and cations over a wide pH range, whereas the relatively acidic magnetic variants can show selectiveness for polar organic contaminants (Isaeva et al. 2021) and complex more with some metal ions due to relatively more OFGs.