Impact of Biosolids-Derived Biochar on the Remediation and Ecotoxicity of Diesel-Impacted Soil

Abstract

This study aimed to investigate the impact of biosolids-derived biochar on the remediation of Australian soil contaminated with diesel while investigating the role of biochar in the remediation. To achieve the latter aim, sodium azide (NaN₃) was added to a separate biochar treatment (BN) to alter the bacterial community structure. Biochar (B) reduced detectable hydrocarbons by 2353 mg/kg compared to the control (C) treatment at week 24. However, the BN treatment reduced the hydrocarbon concentration by 3827 and 6180 mg/kg, relative to B and C, respectively. Soil toxicity significantly decreased at week 24 compared to the start of the remediation in B, but not in the control. Biochar and control treatments generally showed a similar bacterial community structure throughout the incubation, while the bacterial community structure in BN differed significantly. Biodegradation was found to play a significant role in hydrocarbon removal, as the variation in the bacteria community coincided with differences in hydrocarbon removal between B and BN. The increased removal of hydrocarbons in the BN treatment relative to B coincided with increased and reduced relative abundances of Gordonia and JG30-KF-CM45 genera, respectively. This study showed that NaN₃ led to a transient and selective inhibition of bacteria. This study makes an important contribution towards understanding the use of NaN₃ in examining the role of biochar in the remediation of diesel-contaminated soil. Overall, we conclude that biochar has the potential to enhance the remediation of diesel-contaminated soil and that biodegradation is the dominant mechanism.

1. Introduction

Soil contamination by petrogenic hydrocarbons represents one of the major legacies of the widespread use of crude oil throughout the 20th century. It is a major global environmental problem worldwide, including in Australia, Canada, and many other countries [1,2,3]. Exposure to this contaminant poses a threat to biota, from microbes to humans [4]. For example, humans and animals can be exposed to teratogenic, immunotoxic, carcinogenic, and other toxicological impacts when they come into contact with this class of contaminant [5]. These negative impacts and the global reach of this environmental problem confirm the urgent requirement to remediate petroleum hydrocarbon-contaminated soils. Different techniques exist to remediate hydrocarbon-contaminated soil [5]; the most desirable remediation approach should cost-effectively remove the contaminant without causing secondary pollution. Some techniques fail in this regard; for example, thermal methods like incineration generate secondary waste products after remediation [6]. These shortcomings have created the opportunity to develop new or improve existing approaches. One amendment that has gained attention in the remediation of petroleum hydrocarbon-contaminated soil is biochar.

Biochar is a carbon-based product derived from the thermal decomposition of various organic biomass (including wastes) at high temperatures in an oxygen-limited environment [7]. Various studies have reported that biochar enhances the remediation of hydrocarbon-contaminated soil [8,9,10,11,12,13,14]. For example, Aziz et al. [10] reported a 110% increase in the removal of hydrocarbons from soil amended with biochar compared to the control treatment. In another study, biochar application led to an increase of 27% in hydrocarbon removal compared with the control treatment [14]. The source of the biochar that has been used in the remediation of petroleum hydrocarbon-contaminated soil differs but includes crop residues and biosolids (sewage sludge) [10,15,16,17]. The use of biosolids or sludge digestate in biochar production and subsequently in the remediation of contaminated soil is valuable from a circular economy perspective [9]. The sustainable management of sewage sludge (biosolids) represents one of the significant challenges of today’s world due to increasing sludge generation [18]. In 2017, global biosolids production was 100–125 million tons; in 2025, production is estimated to reach 150–200 million tons [19]. Transforming biosolids into biochar reduces the amount of waste and biosolids-derived greenhouse gas emissions [20]. While efforts have been made to understand the potential of biosolids-derived biochar for the remediation of diesel-contaminated soil, no study has focused on soils with a total petroleum hydrocarbon (TPH) contamination greater than 3% [9,10]. There is a need to examine the effect of biosolids-derived biochar on the remediation of more contaminated diesel-contaminated soil (>5%); the efficacy of biochar in terms of the remediation of hydrocarbon-impacted soils has been shown to be influenced by different factors, including contaminant concentration [9,12,21].

The main mechanisms underlying the efficacy of biochar in the remediation of petroleum hydrocarbon-contaminated soil are the subject of debate [10,15,21,22]. Whilst, Mukome et al. [21] report that biodegradation is the main mechanism of hydrocarbon removal, Wei et al. [22] conclude that adsorption is likely the main route for hydrocarbon removal. One method that has been used in determining the mechanism involves exposing biochar-treated contaminated soil to sodium azide (NaN₃), with subsequent comparison to biochar-amended contaminated soil not exposed to NaN₃ [15]. NaN₃ is a selective inhibitor that targets Gram-negative bacteria [23] due to reduced cell wall thickness. NaN₃ inhibits microbial activity [15], allowing for the determination of biotic and abiotic contributions to hydrocarbon removal. However, it remains unclear whether exposure to petroleum hydrocarbon-contaminated soils containing biochar to NaN₃ results in microbial inhibition, as no evidence has been provided in past studies. This is in part due to the fact that NaN₃ is reported to have a selective effect on bacteria [24]. However, to date, the impact of the microbial community on the efficacy of the remediation of petroleum hydrocarbon-contaminated NaN₃-amended soil has not been reported. This study assesses the impact of biochar in the remediation of hydrocarbon-contaminated soil using NaN₃, while integrating both quantitative PCR and 16S rRNA sequencing to assess the impact of the inhibition. Understanding the mechanisms associated with the efficacy of biochar can provide information that can be used to optimise biochar-based remediation.

The use of chemical analytical instruments is the traditional way of assessing how much substance has been removed during the remediation of petroleum-hydrocarbon-contaminated soil [25]. While this can accurately determine the amount of contamination removed, it does not directly indicate the impact of the remaining contaminants and intermediates on the ecological system [26]. Previous studies have shown that the best treatment in terms of hydrocarbon removal is not the treatment exhibiting the lowest soil toxicity at the end of incubation [9,27]. Further, concerns relating to biochar toxicity strengthen the need to integrate soil ecotoxicity testing into bioremediation studies [28]. Therefore, this study combines chemical analysis with ecotoxicity assessment.

This research is novel because it represents the first study to examine (i) the effect of biosolids-derived biochar on the remediation of diesel-contaminated soil with a high TPH concentration (>5%); and (ii) to integrate quantitative PCR and 16S rRNA sequencing while using NaN₃ to study the mechanism underlying the remediation of hydrocarbon-contaminated soil using biochar. This study seeks to determine: (i) if biosolids-derived biochar will enhance hydrocarbon removal and soil ecotoxicity in heavily diesel-contaminated Australian soil; and (ii) the main mechanism of biochar in the remediation of diesel-contaminated soil using NaN₃. The findings of this study will enhance the understanding of the effect of biosolid biochar on the remediation of Australian soil heavily contaminated with diesel and improve the way NaN₃ is used in studying the mechanism underpinning biochar’s enhancement of the remediation of hydrocarbon-contaminated soil.

2. Materials and Methods

2.1. Soil Sample and Characterization

Uncontaminated pasture soil collected from Whittlesea, Melbourne, Australia, was used for this study [27]. The soil has a pH, total carbon, total nitrogen, and total phosphorus content of 7.6, 2.23%, 0.22%, and 313 mg/kg, respectively [27]. The soil texture was classified as clay loam soil.

2.2. Biochar Production and Characterization

The feedstock used for biochar production was biosolids from the Mount Martha water recycling plant operated by South East Water Corporation, Melbourne, Australia. The biochar was produced in a Barnstead Thermolyne 30400 muffle furnace at 900 °C. The basic properties of the biochar, including pH, surface area, pore volume, electrical conductivity, and proximate analysis, are shown in Table S1.

2.3. Soil Contamination and Bioremediation Experiment

Uncontaminated soil from Section 2.1 was spiked with above 5% diesel (Caltex, VIC, Australia) and mixed thoroughly. Diesel fuel is composed of approximately 64% aliphatic hydrocarbons, 35% aromatic hydrocarbons, and 1–2% olefinic hydrocarbons [29]. The contaminated soil was left in a fume hood for 24 h to allow for the dispersion of the volatile fraction. Soil was dispensed into 45 different plastic pots of similar diameter. The experimental setup included three (3) treatments, comprising 15 pots each of non-amended control (C), 5% (w/w) biochar (B), and 5% (w/w) biochar + 0.2% NaN₃ (BN) treatments. For BN treatments, each pot had 30 g of soil, while other treatment pots (B and C) had 50 g. Previous studies have reported increased hydrocarbon removal with biochar when applied at 5% w/w [16,30].

To achieve an initial moisture content of 13–15% (w/w) in all pots at week 0, Milli-Q deionised water was added to the B and C treatments, while a 0.2% NaN₃ solution was added to the BN treatment. The pots were incubated for 24 weeks at room temperature, and the contents were mixed at least once every week. The moisture content was maintained every week by adding 0.2% NaN₃ solution for the BN treatment and Milli-Q deionised water to treatments C and B, where necessary.

Three pots from each treatment group were destructively sampled at weeks 2, 4, 7, 13, and 24 for molecular and total petroleum hydrocarbon (TPH) analysis. Part of the sample used for TPH analysis was used for the determination of the soil physicochemical properties and ecotoxicity analysis for week 24. Samples were stored at 4 °C and −20 °C.

2.4. Total Petroleum Hydrocarbon (TPH) Analysis

RemScan technology (Ziltek, Edwardstown, Australia) was used to determine the total petroleum hydrocarbon content (C₁₀–C₄₀) of the soil [31]. Prior to analysis, soil samples were air dried and sieved using a 2-mm sieve. The RemScan has been previously shown to correlate with traditional analysis (GCMS) [31].

2.5. Bacterial Community Analysis

2.5.1. DNA Extraction

DNA was extracted from 0.25 g of soil using DNeasy PowerSoil Pro kits (Qiagen, Venlo, Limburg, The Netherlands). DNA quality (A₂₆₀/A₂₈₀) and quantity were assessed using a Nanodrop Lite spectrophotometer (ThermoScientific, Waltham, MA, USA) and Qubit 4.0 fluorometer (ThermoScientific, Waltham, MA, USA), respectively. Quality assurance on the extracted DNA samples showed that the A₂₆₀/A₂₈₀ concentrations were between 1.41–1.87 and 5.22–141 ng/μL, respectively. The extracted DNA samples were stored at −20 °C until further analysis.

2.5.2. Quantification of Total and Alkane-Degrading Bacteria

The abundance of genes encoding for the total (16S rRNA) and alkane-degrading (alkB) bacteria was determined using quantitative PCR in a QIAGEN Rotor-Gene machine (Qiagen, Venlo, Limburg, Netherlands). Quantitative PCR for both genes involved a 20 µL reaction volume, which comprised of KAPA SYBR^®FAST qPCR Mastermix (10 µL), forward primer (0.4 µL), reverse primer (0.4 µL), PCR-grade water (8.2 µL), and DNA template (1 µL) [32]. More details are discussed in Text S2.

2.5.3. Bacterial Community Analysis via Next-Generation Sequencing

The V4 region of the 16S rRNA gene was amplified prior to sequencing, using primer pairs 341 F and 806 R as the forward and reverse primers, respectively [33]. Libraries were prepared using Nextera XT indices following Illumina guidelines [34], pooled to 6 pM, including 20% phiX, and sequenced on an Illumina MiSeq (2 × 300 bp) (Illumina Inc., San Diego, CA, USA).

Amplicon size was verified on a Bioanalyzer 2100 (Agilent, Santa Clara, CA, USA) and via gel electrophoresis using amplicons and indexed samples. Following completion of the sequencing run, FastQ files were processed and denoised with DADA2 in Qiime2 as previously described, resulting in a median of 27,550 quality-filtered reads and a total of 3871 amplicon sequence variants (ASV) across 49 samples [35].

The 16S rRNA gene amplicon sequences obtained from this study were submitted to the NCBI Sequence Read Archive (SRA). The BioProject accession number is PRJNA990956.

2.6. Ecotoxicity Testing

Ecotoxicity assessment was carried out at the start and end of incubation using the Microtox test. A luminescent marine bacterium (freeze-dried Aliivibrio fischeri) obtained from Streamline Hydro Pty Ltd., Capalaba, (QLD, Australia) was used for ecotoxicity testing [33]. The test organism is a nonpathogenic bacterium that is responsive to a large number of toxic substances [36]. The freeze-dried bacteria were reconstituted using the medium provided by the supplier [37]. It was then exposed to an aqueous extract of diesel-contaminated soil [33]. Further details are discussed in Text S3.

2.7. Data Analysis

All analyses, unless otherwise stated, were carried out in replicates. One-way analysis of variance (ANOVA), with the Tukey, at p < 0.05 was used to assess the statistical significance of TPH removal, C/N ratio, PO₄, bacteria gene copies, and EC₅₀ in Minitab software (Minitab, State College, PA, USA, https://www.minitab.com/en-us/products/minitab/).

3. Results and Discussion

3.1. Effects of Biochar on Hydrocarbon Removal from Contaminated Soil

The TPH concentration of the diesel-contaminated Victoria pasture soil at week 0 was 54,147 ± 1214 mg/kg. The hydrocarbon concentration was assessed in B, BN, and C at different times over 24 weeks. The results showed that hydrocarbon removal from the contaminated soil increased in all treatments with time (Figure 1). A rapid hydrocarbon removal of 15–18% was observed in all treatments by week 2, followed by a slow but steady hydrocarbon removal. This pattern was similar to the findings of Koshlaf et al. [37], where rapid removal was observed in all treatments in the first 4 weeks, followed by a slow removal. In this present study, the hydrocarbon concentration in all three treatments (C, B, and BN) decreased from 54,147 to at least 27,820 mg/kg after 24 weeks of incubation, representing a hydrocarbon removal of at least 26,367 mg/kg from the soil (Figure 1). At the end of the incubation (week 24), the hydrocarbon removed in treatments amended with biochar (B and BN) was more than half that of the starting hydrocarbon concentration (Figure 1). This was not the case in the control, where the hydrocarbon removal was less than 50% at week 24. This suggests that biochar application alone or with NaN₃ showed higher hydrocarbon removal at week 24 than C.

Figure 1 showed that between weeks 2 and 7 of incubation, the hydrocarbon removal was higher in the biochar treatment than the control by 1422–1833 mg/kg. After 24 weeks, the differences in hydrocarbon removal between both treatments increased to 2353 mg/kg (Figure 1). Increased hydrocarbon removal following biochar application has been reported in the literature due to biochar’s positive role in improving soil properties and supporting soil microbes by providing shelter and protection [38]. For example, Wang et al. [16], who observed higher hydrocarbon removal in biochar treatment relative to the control, found that the phosphorus (P₂O₅) increased significantly at the end of the incubation compared to the start of remediation in the biochar treatment, but not in the control. Previous papers have demonstrated that the application of biochar resulted in higher hydrocarbon removal than the control [8,9,10,14]. While our current result demonstrated the effectiveness of biosolids-derived biochar in the remediation of diesel-contaminated soil, the differences in hydrocarbon removal between the biochar and treatments were five times lower compared to a previous study [10]. The TPH concentration in our current study was 1.9 times higher than the previous study [10]. However, there is a likelihood that the differences in TPH concentration may not be fully responsible for these discrepancies, as an earlier paper showed that the application of biosolid biochar resulted in a similar difference in hydrocarbon removal between the biochar and control (2287 mg/kg) to that observed in the current study (2353 mg/kg), despite differences in incubation time and initial TPH concentration [9].

In the presence of NaN₃, biochar gave rise to significantly higher hydrocarbon removal compared to the control or the sole biochar treatment (without NaN₃) after 24 weeks of incubation (p < 0.05) (Figure 1). There are two possible reasons for the higher removal observed in the BN treatment. Firstly, the contaminated soil at week 0 had a higher C/N ratio by virtue of the addition of a very high amount of hydrocarbon to the soil (Table S2). Among all treatments at week 24, only BN showed a significantly (p < 0.05) lower C/N ratio than in week 0 (Table S2). Nitrogen was likely released due to azide reduction to the ammonium ion [39], lowering the C/N ratio of the contaminated soil. In addition, the higher phosphate concentration in the BN treatment may have contributed to the higher hydrocarbon removal (Table S2). Inorganic nutrients like phosphate can be utilised quickly when there is a large amount of organic carbon in the soil [40]. This is a result of increased metabolism by soil microbes [40]. There is a likelihood that the PO₄ in the soil in this current study will be altered since a large content (>5% TPH) of hydrocarbons was introduced to the soil. Biodegradation of the contaminated soil is limited in the presence of inorganic nutrient depletion [40]. Wang et al. [16] found that TPH removal increased as the phosphate (P₂O₅) level in the soil increased. The increase in phosphate may be due to the release of phosphate from dead microbial cells [39]. In addition, the presence of NaN₃ in the biochar-amended soil resulted in the selective inhibition of non-resistant microbes and the growth of NaN₃-resistant microbes. This could result in the promotion of beneficial hydrocarbonoclastic bacteria. The details of this will be discussed further in Section 3.4. The result of this study contradicts other works that reported that the co-application of NaN₃ with biochar resulted in lower [15,21] or non-significant hydrocarbon removal [41]. Aside from the fact that one of the studies applied ethanol alongside NaN₃ [21], several factors may be responsible for the different outcomes, including incubation conditions, biochar type, hydrocarbon class, soil microbial community, and soil properties.

Complying with the regulatory guidelines is important in any commercial remediation. At the end of the incubation, the three treatments achieved the EPA Victoria Category B waste requirement of 40,000 mg/kg [42], with BN being the first to fulfill this requirement. However, during the incubation, none of the treatments achieved the fill material maximum concentration of 1000 mg/kg. Kinetic analysis was used to estimate when the fill soil requirement would be achieved for the different treatments. Kinetics provides an avenue to predict the remediation time, as well as evaluate the contaminant concentration at a given time [43]. The kinetic curve, equation, and half-life of the reaction are shown in Figure S1. The data for the three treatments fit first-order kinetics (R² = 0.96–0.97). Other studies have also utilised the first order for their bioremediation kinetics [43,44]. The first-order kinetic curves showed that the biodegradation half-life (t_1/2) was lower in BN, followed by B and C. This implies that, based on the model, treatment BN will achieve the removal of half of the amount of hydrocarbon in the soil in a faster manner than other treatments. Based on estimation from the first-order kinetic equation [45], B and BN treatments would reach 993.3 mg/kg, a concentration lower than the maximum threshold of the EPA Victoria fill material maximum concentration of 1000 mg/kg, 16 and 44 weeks ahead of the control, respectively (Table S3) [42]. Achieving the EPA Victoria fill soil waste category earlier will save cost and time of remediation, suggesting that applying biochar to the soil would benefit remediation.

3.2. Bacterial Gene Copies

The alkB and 16S rRNA gene copies were determined for all treatments at different sampling times to assess the population of the total bacteria and alkane-degrading bacteria, respectively (Figure 2). Figure 2a shows changes in the copy number of the alkB gene with time for all treatments. This gene is generally used to assess the potential for the biodegradation of alkanes [46]. The results indicated that the copy number of this gene was significantly higher (p < 0.05) at weeks 2 and 24 in all treatments, relative to week 0, except BN at week 2. The lack of changes in the abundance of alkB in treatment BN at week 2 relative to the start of remediation was due to the short-lived bacterial inhibition resulting from the presence of NaN₃. This observation aligns with the result of the bacterial community structure, where the community structure was similar in BN at week 2 compared to the soil at week 0 (Figure S3). Overall, this finding suggests that the hydrocarbons could have been degraded by the native hydrocarbonoclastic bacteria present in the soil. The strong positive correlation between the number of alkB gene copies and hydrocarbon removal further confirms the role of biodegradation in the removal of the contaminant in this study (r = 0.74, n = 6). A previous study also found a strong positive correlation between hydrocarbon removal and the number of copies of this gene [9].

The results showed that the number of 16S rRNA gene copies in all treatments did not significantly increase at weeks 2 and 24 compared to week 0 (Figure 2b). At week 2, BN was significantly lower than the B or C treatments because of the effect of NaN₃. In comparison to the control, biochar addition did not lead to an increase in bacterial gene copies, which demonstrates that biochar was not beneficial in increasing the bacterial population. This observation may be linked to the lack of differences in the soil properties between biochar and the control treatments (Table S2). Table S2 showed that important soil properties, soil pH, and C/N ratio did not differ between both treatments at week 24. Generally, soil microbes are sensitive to changes in soil properties [47], and if changes do not occur, there is a likelihood that the bacterial population will not change. Previous studies have also reported that biochar application did not alter the number of 16S rRNA gene copies [9,48]. The implication of this is that biochar addition to the soil, unlike other remediation techniques, did not cause disturbance to the bacterial population relative to the unamended control.

Overall, the results showed that the bacterial inhibition induced by NaN₃ is only short-lived, raising a question about one of the methods used in the literature to assess the major mechanism of biochar [15,41]. In such studies, one treatment is coapplied with biochar and NaN₃ (BN), while a separate treatment is amended with biochar alone (B). The assumption in these past studies is that NaN₃ inhibits bioactivity, but no evidence was provided to verify if the application of NaN₃ in hydrocarbon-contaminated soil amended with biochar was successful in causing complete inhibition. It would be more valuable to validate microbial inhibition before drawing any inferences about the mechanism in order not to exclude important information. This current study has addressed this gap by integrating quantitative PCR, which provides an avenue to check for bacterial inhibition.

3.3. Bacterial Diversity

The diversity of the bacteria within a community was studied using alpha diversity [22]. The dynamics of bacterial richness, Shannon diversity, and evenness in the contaminated soil as a function of time and treatments are shown in Figure S2. The result of this study showed that all three alpha diversity indices were lower at week 24 relative to week 0. As an example, the species richness in B, BN, and C treatments at week 24 was approximately 310, 140, and 300, respectively (Figure S2). At week 2, only BN showed similarly or close similarity to the initial-contaminated soil. The observation of the BN treatment at week 2 agrees with the result of the 16S rRNA gene copies, where no difference was found between week 2 and week 0 (Figure 2b). This may have been due to the fact that at week 2, NaN₃ acted as a preservative and thus prevented the overgrowth of any bacterial group [49,50]. At Week 24, treatment BN generally had a lower richness, diversity, and evenness compared to B, presumably because of NaN₃ activity as a selective inhibitor [51]. Previous studies have found that the NaN₃ addition resulted in lower richness (Chao1 estimator) [52] or Shannon index [53]. Figure S2 shows that the application of biochar resulted in higher species richness and Shannon diversity at both sampling times than the control. This observation suggests that biochar benefited a wider range of bacteria relative to the control. For example, in week 24, only 123 and 158 genera were present in the control and biochar treatments, respectively (Table S4). This implied that the addition of biochar made the soil conducive to more bacterial genera than the control without biochar. An earlier study reported higher Shannon diversity, ACE, and Chao in biochar treatments relative to the control at days 45 and 120 [8].

3.4. Bacterial Community Structure and Abundance

The results of the sequencing revealed that Actinobacteriota (23–81%), Firmicutes (8–32%), Proteobacteria (3–26%), Chloroflexi (1–20%), Planctomycetota (0.08–5%), Verrucomicrobiota (0.03–5%), Bacteroidota (0.008–5%), Acidobacteriota (0.17–4%), Sumerlaeota (0.014–3%), Patescibacteria (0.01–2%), Myxococcota (0.05–1%), Gemmatimonadota (0.008–1%), and Nitrospirota (0.01–1%) had a relative abundance of 1% and above in at least one of the treatments (Figure 3). The relative abundances of these phyla accounted for at least 99.5% of the total bacterial community. Among them, Actinobacteriota was the most dominant in all treatments at different sampling times, except for the contaminated soil at week 0 and the BN at week 2. Actinobacteriota is one of the largest and most diverse bacterial phyla and has many genera with hydrocarbon-degrading abilities [54,55,56]. This phylum was identified as the most dominant phylum in hydrocarbon-contaminated soils in a previous study [12]. It is apparent from Figure 3and Figure S3 that changes in the community structure were observed in all treatments after week 0 at both phylum and genus levels.

At the genus level, the relative abundances of some of the genera generally increased after week 0 (Rhodococcus, JG30-KF-CM45, and Norcardiodes), whereas others decreased (Bacillus, Rhizocola, and Candidatus_Udaeobacter) (Figure S3). The number of genera decreased from 281 in the contaminated soil at week 0 to 80–252 after week 0, depending on treatment and time (Table S4). The dynamics of the bacterial community structure can be due to the changes in hydrocarbon concentration and type over time [57]. Another reason, which is particular to the amended treatments only (B and BN), is the introduction of amendments to the soil. The soil bacterial community is sensitive to changes in the soil’s physicochemical properties [47], which could take place due to the above mentioned scenarios. As can be seen in Table S2, the C/N ratio was different between the contaminated soil at week 0 and the treatments at week 24, though only statistically significant in the BN treatment. The soil C/N ratio controls important soil parameters, nitrogen immobilization and mineralisation [58].

The data in Figure 3 indicates that the application of biochar in most cases did not cause a significant change in the bacterial community structure relative to the control at the different sampling times at the phylum level; even at the genus level, both treatments were similar (Figure S3). These findings corroborated the lack of differences observed in the Pielou evenness between both treatments (Figure S2). Other reports have shown that the application of biochar did not cause a significant difference in the bacterial community relative to the control [22,59,60]. For example, Bao et al. [59] found that the relative abundance and community structure of bacteria in the control and biochar treatments were similar. This absence of major differences in the bacterial community structure may be because both treatments did not differ significantly in most soil properties (pH, C/N, PO₄) (Table S2). Soil pH has been reported in previous studies to influence soil bacterial diversity and community composition [61,62]; one way that pH may influence the microbial community is by modulating soil nutrient availability [63]. Another soil property, the C/N ratio, affects nitrogen mineralization and immobilisation [58].

Figure 3 showed that, apart from week 2, the bacterial community differed from the initial contaminated soil at all sampling times, which demonstrates the lack of complete inhibition of NaN₃. The bacterial communities of BN and B were dissimilar at the different sampling times, especially after week 4; a lower number of bacterial phyla accounted for the top 1% bacterial phylum in the BN treatment (Figure 3). Similarly, compared to BN, the total relative abundances of the Gram-negative (GN) and Gram-positive (GP) bacteria were higher and lower in the biochar treatment after week 4, respectively. The relative abundance of Actinobacteriota, a GP bacterial phylum, was higher in BN than in B after week 4 (Figure 3). Conversely, the relative abundance of Chloroflexi, a GN bacterial phylum, was lower in the BN treatment than in the B treatment after week 2. At the genus level, the relative abundance of JG30-KF-CM45, a Chloroflexi phylum, was higher in B than in treatment BN after week 2 (Figure S3). In contrast, the Gordonia genus from the Actinobacteriota phylum was seen in treatment BN after week 2 but was found in B only at week 24, with a relative abundance of 0.01% (Figure S3). This genus has been reported in previous hydrocarbon-contaminated compost amended with biochar [64,65] and soil with bacteria-immobilised biochar and soil amended with digestate [66]. The presence of sodium azide in treatment BN was likely responsible for the apparent differences between both treatments. Sodium azide is a selective bacteria inhibitor that targets GN bacteria [23], although some can survive in its presence, as seen in this study (Figure 3 and Figure S3). This is the first time the effect of NaN₃ on bacterial community structure in petroleum hydrocarbon-contaminated soil amended with biochar has been reported. However, a previous non biochar study reported that NaN₃ had a selective effect on bacteria [51].

Another aim was to determine how differences in the autochthonous bacterial community between B and BN treatments could have been responsible for the variation in hydrocarbon removal. Although there could be other explanations, two genera could provide insight into how the differences in the bacterial community structure affected hydrocarbon removal (Gordonia and JG30-KF-CM45) (Figures S3 and S4). At week 2, when the genus Gordonia was absent in treatment BN, there was no discernible difference in hydrocarbon removal between both treatments (Figure S4a). As the difference in the relative abundance of this genus increased with time after week 4, the difference in hydrocarbon removal between these two treatments increased (Figure S4a). A strong positive correlation was found between the difference in the relative abundance of this genus and the difference in hydrocarbon removal between both treatments (r = 0.89, n = 6). This genus contains several species that are reported to degrade various hydrocarbons and are present in contaminated sites across the world, such as Korea, Mexico, China, Russia, Kazakhstan, and Indonesia [54,67].

Another genus that deserves attention is the Gram-negative JG30-KF-CM45. As shown in Figure S4b, there was no difference in the relative abundance of the JG30-KF-CM45 genus between B and BN treatments at week 2, which coincides with the lack of difference in TPH removal between both treatments at the same sampling time. As the difference in the relative abundance of JG30-KF-CM45 between both treatments increased over time, the difference in hydrocarbon removal increased (Figure S4b). A strong negative correlation was found between the difference in hydrocarbon removal and the difference in the relative abundance of JG30-KF-CM45 (r = −0.96, n = 6). Although in a previous study involving a phytomicrobial electrochemical system, norank_f__JG30-KF-CM45 was associated with hydrocarbon degradation [68], there is a scarcity of literature that implicates JG30-KF-CM45 in hydrocarbon removal. A previous study found that PAH removal was lower in treatments with a higher relative abundance of Chloroflexi phylum and Chloroflexi_unclassified genus [59]. JG30-KF-CM45 belongs to the Chloroflexi phylum [68], and the Chloroflexi_unclassified genus may be related to JG30-KF-CM45.

In summary, while Gordonia was beneficial in promoting hydrocarbon removal, JG30-KF-CM45 was detrimental. Additionally, the results of the sequencing demonstrate that the changes in the bacterial community between B and BN coincided with differences in hydrocarbon removal and thus show that biodegradation was the main mechanism for hydrocarbon removal in this study.

3.5. Ecotoxicity

The toxicity of the soil was measured at the beginning and the end of incubation using the Microtox test. The toxicity of the soil at week 24 was at least 2.6 times lower in all treatments, relative to the beginning (Figure 4), confirming a decrease in soil toxicity following remediation, consistent with a previous study [60]. However, only treatment B showed significantly lower toxicity at week 24, relative to week 0 (p < 0.05). Although the sole biochar treatment was not the best treatment in this work in terms of hydrocarbon removal, lower soil toxicity was achieved in this treatment at week 24, relative to other treatments (Figure 4). Previous works have also shown the lack of a relationship between both parameters [9,27]. A number of reasons for these observations have been highlighted previously [9], although, in the current study, the component of one of the treatment agents (NaN₃) strongly contributed to this observation.

At week 24, the soil was two times less toxic than the control, suggesting that biochar application was beneficial in reducing soil toxicity. The reduced soil toxicity following biochar application is consistent with previous works [9,60]. During the biodegradation of hydrocarbon contaminants, intermediate metabolites like fatty acids and aldehydes are formed [69]. Biochar reduces soil toxicity because it can immobilise the toxic metabolites formed during bioremediation [60]. The result of an earlier study showed that the toxicity of the soil was lower when biochar was applied on day 80, relative to when applied at day 0 [60]. They speculated that when biochar was added at the early stage, the petroleum hydrocarbon was absorbed at the site for immobilisation, thus not leaving enough space for the metabolites. However, when it was added at later stages, there were more available sites for the toxic substance to immobilise, which could have led to lower toxicity, which suggests the role of adsorption. Biochar’s role in adsorption/immobilisation is facilitated by its unique properties, including its porous nature and surface functional groups [70]. A previous study found that soil toxicity decreased as the pyrolysis temperature used to produce the biochar increased, with the surface area also increasing with temperature [9].

In the presence of NaN₃, biochar (BN) showed higher toxicity compared to the sole biochar. The higher toxicity of BN, relative to B, was due to the selective effect of NaN₃ on bacteria [51]. As seen in Table S4, Figure 3 and Figure S2, NaN₃ addition in BN resulted in a reduction in soil bacterial diversity as well as the number of phylum and genera that can survive in the soil. Further, the test organism for the ecotoxicity testing (Aliivibrio fischeri) is a Gram-negative bacterium, and this group of bacteria is more susceptible to NaN₃. A previous study had earlier demonstrated that sodium azide showed toxicity when exposed to the test organism [71].

4. Conclusions

This study examined the impact of biosolids-derived biochar on the remediation and ecotoxicity of heavily diesel-contaminated soil while assessing the major mechanisms of biochar in remediation. The results showed that biochar enhanced hydrocarbon removal and reduced soil toxicity. This study also showed that NaN₃ results in a short-lived and selective inhibition of bacteria. The efficacy of biochar in enhancing hydrocarbon removal was influenced by the differences in the bacterial community following NaN₃ addition relative to the sole biochar treatment, thus demonstrating that biodegradation contributed greatly to the removal of the hydrocarbon. Specifically, the differences in the relative abundances of Gordonia and JG30-KF-CM45 genera were related to variations in hydrocarbon removal in BN relative to the biochar treatment. The findings of this study will help improve the way NaN₃ is used in studying the major mechanism of biochar in the remediation of hydrocarbon-contaminated soil. This study is limited in that it was carried out under laboratory conditions and thus may not wholly reflect outcomes from field studies. Future studies should integrate complete physicochemical characterisation as well as identification and quantification of degradation products during the remediation of diesel-contaminated Australian soil amended with biochar.