3.1. Biochar Characterisation
The results of proximate analysis, elemental analysis and pH are presented in
Table 2; The lignin, cellulose and hemicellulose of the raw material were 31.8%, 50.6% and 80.2%, respectively [
21]. All biochar types can be classified in class 1, as the organic carbon concentrations were ≥60% and ≥50% following IBI (2015) [
22] and EBC (2022) [
23] standards, respectively. The ash content in the samples increased with the working temperature in the semi-pilot reactor, but it did not follow the same trend with the scaling-up process, i.e., it did not increase either in the pilot reactor or in the industrial pyrolyser. On the other hand, the concentration of volatiles, which is related to the labile organic fraction of the biochar [
24], decreased in the biochar samples when the temperatures increased for all scale levels: semi-pilot, pilot and industrial. Thus, the great influence of temperature on biochar characteristics was proven [
25].
All the biochar samples were rich in carbon (>70%), oxygen (>10%) and hydrogen (>3%) because they were obtained from a woody feedstock. It had been previously reported that woody materials yield high contents of the mentioned elements [
26]. The concentration of nitrogen (N) was between 0.5% and 2%, which falls within the range previously reported for biochar from woody materials [
26,
27]. The N content in biochar decreased along with temperatures, being <1.0% for the highest temperature. However, N content in biochar is more tightly related to the type of feedstock than the pyrolysis conditions, and during the pyrolysing process, the N concentration increases [
28]. In this case, the N content in the biochar from olive tree pruning ranged from 1.53 to 0.51%, whereas for the raw material, it was 0.39%. No relationship between S content and the temperature of the production of biochar was observed, indicating that the S content also depends mostly on feedstock characteristics [
29]. The EC values in all samples ranged between 0.32 and 1.71 dS/m. Similar results have been reported by Picca et al., 2023 [
30].
All the biochars produced in this work showed a ratio O/Corg < 0.4, fulfilling the limit established by EBC (2022). B400, B500 and BMEC showed a ratio H/Corg < 0.7, thus satisfying the limit established by EBC and IBI guidelines. Such values indicate a high degree of carbonisation and high stability and hence a high potential for C sequestration when biochar samples are added to the soil. Conversely, B600 and BF600 exceeded the limit, indicating that the obtained products either were non-pyrolytic chars or the pyrolysis process was deficient [
23].
All the samples were alkaline, and the pH value increased along with the temperature. Zhao et al., [
28] reported that the increase in pH in biochar samples due to the increase in temperature is because of the destruction of functional groups, which generate alkaline salts and alkalinising elements, especially carbonates, which for this group, is the main cause of increasing pH [
31]. For higher temperatures, the separation of the alkali salts from organic material is greater, and thus, the pH reached is also higher [
32]. Similarly, Alburquerque et al. [
33] also reported a high pH value (11.51) and high content of carbonates in biochar from olive tree pruning. Moreover, for a scheduled temperature of 600 °C, the pH increased when scaling-up the process. The reason could be that at the industrial and pilot scale, it is more difficult to control the temperature, having a larger error, i.e., the pilot reactor has a ±15°C margin and temperatures above 600 °C were reached, whilst in the industrial reactor the margin is ±50 °C because the system for temperature control is less accurate. In this case, the only way in which the temperature can be decreased when the set temperature (666 °C) is exceeded is by reducing the entry of material in the feed [
34]. Consequently, in the industrial pyrolyser, a relevant percentage of the raw material was pyrolysed at a higher temperature until it managed to decrease.
3.2. Phytotoxicity Test and Its Relationship to the Contaminants Present in Biochar
The germination test provides information about the possible toxicity of a pure material, and for this reason, it was carried out with the most concentrated solution of the different biochars obtained in this work.
Table 3 summarises the phytotoxicity test carried out. The results show that the main factors that influenced the phytotoxic effect were the temperature and type of reactor used.
At the semi-pilot scale, the biochar produced at lower temperatures showed some phytotoxicity that turned into a phytostimulant effect when the pyrolysis temperature increased. Intani et al. [
35] and Xiao et al. [
36] observed a similar effect of temperature on phytotoxicity. The effect also depended on the crop under evaluation; for tomato and lettuce, there were no significant differences between B400 and BMEC, whereas, for cress and radish, there was the same behaviour between BF600 and BMEC. At 400 °C, biochar was phytotoxic for tomato and lettuce, moderately phytotoxic for radish and phytostimulant for cress. At 500 °C, biochar was phytostimulant for tomato and radish, moderately phytotoxic for cress and phytotoxic for lettuce. Finally, the biochar produced at 600 °C was phytostimulant for all the crops. The radish crop was the least affected by phytotoxicity and benefited the most from phytostimulation. The role of temperature, time of residence, feedstock or biochar application rate in phytotoxicity has been more deeply analysed by other authors, e.g., Lehmann and Joseph [
37], Sun et al. [
27] and Xiao et al. [
36], but the influence of the scaling-up process in the phytotoxicity of the product has not been reported in previous studies to the best of our knowledge. Unfortunately, it has been found that the phytostimulant effect of the biochar produced at 600 °C on a semi-pilot scale was reproduced neither on a pilot nor at an industrial scale. In fact, at the pilot scale, biochar produced at 600 °C (BF600) was moderately phytotoxic for tomato and lettuce, whilst the biochar produced in the industrial reactor at 666 °C (BMEC) was phytotoxic for all crops. It is hypothesised that the difficulties in controlling the operating conditions of the pyrolyser at higher scales, specifically the temperature, could be the reason for the phytotoxicity problems at the industrial scale. To prove the hypothesis, we attempted to identify the phytotoxic compounds and to analyse to what extent they appeared, depending on the temperature reached with each piece of equipment.
One of the characteristics to which the phytotoxicity of biochar in seed germination has been attributed is due to the high concentration of soluble salts (EC > 10 dS/m) [
4]. High salinity concentrations cause yield losses in sensitive crops. Yield losses of 50% have been reported at salinity concentrations of 2.7 dS/m for lettuce, at 7 dS/m for radish, and at 10 dS/m for wheat [
38]. All samples showed EC < 1.80 dS/m, a value that is below the EC limits at which productivity losses begin to appear in the most sensitive crops. Therefore, the phytotoxicity of the samples cannot be attributed to this parameter and maybe others are responsible for this phenomenon.
Interestingly, the phytotoxic effect of BMEC seems to be mainly from the extreme pH (11.52), which was, in turn, assigned to the higher temperatures reached in the industrial pyrolyser (see
Section 3.1 for more details). However, even if biochar with high pH values is not adequate for agronomic purposes due to phytotoxicity, Chintala et al. [
39] have observed that alkaline biochars with high contents of calcium carbonates are useful as an amendment of acidic soils before putting them under cropping. Moreover, biochar with high pH has also shown great potential for alleviating Al toxicity in crops, which is the chief agronomic problem of acidic soils [
40]. Biochar has good potential for the reduction of Al toxicity in crops due to the adsorption capacity of this element and the ability to replace the monomeric Al species in soil exchange sites with more neutral Al hydroxides in soil [
41]. On the other hand, alkaline biochar is useful to bioremediate soils contaminated with heavy metals (HM) because it has been proven that highly alkaline biochar favours the adsorption of heavy metals due to their more negative charges [
5].
3.2.1. Concentration of Heavy Metals (HM) in the Biochar
The total concentration of As, Cd, Co, Cr, Cu, Hg, Mo, Ni, Pb, Se and Zn in the produced biochar are shown in
Table 4. All the samples were below the limits proposed by IBI (2015) [
22]. In contrast, the EBC (2022) [
23] limits are stricter, and the Cu content of all samples exceeded the EBC-Agro (Class III) limit.
The Cu encountered in the biochar samples may come from the application of Cu as a phytosanitary treatment in the field. Subsequently, the pyrolysing process concentrated this element, a phenomenon that has been previously reported by other authors [
12,
42]. Other regional regulations are less strict with Cu limits, e.g., the Spanish legislation for fertilisers [
43] establishing three classes of material depending on the concentration of heavy metals, namely classes A, B and C. In our case, all the samples were in class B due to the concentration of Cu, making field application possible.
As, Cd, Hg, Mo, Pb and Se showed low concentration, and their content does not seem to be influenced by the pyrolysing temperature. Conversely, the concentrations of Cr and Ni were directly related to rising temperatures, as had already been observed by other authors, such as Wang et al. [
42] and Hilber et al. [
44]. Unexpectedly, sample BF600 showed lower Cr and Ni content than sample B600, even if the temperature effectively reached was higher in BF600 than in B600; the reactor configuration could have been a determinant of this result. For the rest of the elements, no clear relationship with temperature was observed.
The total concentration of HM decreased in the following order: BMEC > BF600 > B600 > B500 > B400, coinciding with the decrease in the temperature of biochar production. Phoungthong et al. [
45] suggested that the increment of HM concentration along with the temperature is because of the decomposition of the organic matter during pyrolysis and the partitioning of heavy metals with low volatility accumulated in biochar matter.
Although there is no restriction on the application of biochar in the field, it is important to consider the heavy metal thresholds of the international guidelines when using biochar as a carrier for a possible biostimulant or for its application as an organic amendment.
3.2.2. Concentration of Polycyclic Aromatic Hydrocarbons (PAHs) in the Biochar Samples
Polycyclic aromatic hydrocarbons are one of the main contaminants in biochar. They are formed due to the aromatisation and carbonisation of organic matter [
12].
Table 5 shows the concentration of total Σ16 PAHs (US EPA). In all the biochar samples, Naphthalene (69%) was the prevailing PAH, followed by phenanthrene (9%) and acenaphthylene (6%). In general terms, the contents of lower molecular weight PAHs were more abundant than the higher molecular weight compounds.
According to IBI’s guidelines, the permitted range of Σ16 PAH concentrations in biochar samples ranges from 6 to 300 mg/kg, whereas EBC’s recommendations are more rigorous and differentiate between various categories depending on the content. All biochar samples fulfil the IBI specifications. Following the EBC-Agro (Class III) classification, the B600 and BMEC samples were below the limit (6 mg/kg), the B400 and B500 samples were just above the limit by 2% and 7% respectively, and the BF600 sample exceeded the threshold set for Class III. However, only part of the PAH content is bioavailable and potentially risky [
44,
46]. Even the bioavailable part of the PAHs that are directly responsible for the toxicity is very low compared to the total content [
46,
47]. De la Rosa et al. [
48] suggested that PAHs are not the most reliable indicator of toxicity. Our results indicate that pH could be more relevant for phytotoxicity than PAHs. In this light, the BMEC sample was the most phytotoxic sample with the lowest PAH concentration and the most alkaline pH. Therefore, the elevated pH may have contributed more to the inhibition of seed germination than the PAH concentration. However, for the rest of the samples with pH closer to neutrality, there was a strict correlation between the phytotoxicity and the total PAH concentration, both decreasing as follows: BF600 > B400 > B500 > B600. Thus, this study confirmed that PAHs are an important source of biochar toxicity, as previously reported [
49], but they cannot be regarded as the main cause of phytotoxicity.