Environmental risks and agronomic benefits of industrial sewage sludge-derived biochar

Sewage sludge origin and biochar production

A sample of the distillery’s WWTP sludge was obtained after centrifugation of the return sludge from the collection tank of Enviral, a.s. (Leopoldov, Slovakia), a company involved in bioethanol production. The sludge obtained was oven-dried at 50 °C to constant weight. Part of the dried sludge was thermochemically treated by a slow pyrolysis process at 400 °C in a continuous pyrolysis reactor PYREKA (Pyreg, Grevesmühlen, Germany) under strictly anoxic conditions. Slow pyrolysis was carried out at a heating rate of 30 °C/min and after reaching the maximum temperature the feed material was kept in the reactor for 20 min. The use of gaseous N₂ with a flow rate of 2 L/min of the reactor exchange volume ensured that a strictly anoxic atmosphere was achieved. The sewage sludge- derived biochar (BC) and dried sewage sludge (SS) were sieved to a fraction of 0.25–2 mm using standard sieves.

Physico-chemical characterization of sewage sludge-derived biochar

The SS and BC samples were exposed to basic physicochemical characterization. All determinations were carried out in triplicates. For pH determination the KCl method was used. The BC and SS samples were mixed with 1 mol/L KCl at volume ratio 1/10 (v/v). The pH was measured using a pH/EC/TDS multimeter HI5521 (Hanna Instruments). The determination of electrical conductivity (EC) was carried out by mixing SS and BC samples with deionized water at ratio 1/10 (v/v) and measured using a pH/EC/TDS multimeter HI5521 (Hanna Instruments). To quantify the carbonate content in the BC and SS samples, the Jankov calcimeter was used. The analysis of cation exchange capacity (CEC) was carried out using the method modified by Frišták et al. (2013) in three repetitions. The Ba²⁺ as a loading cation and Mg²⁺ as an exchanging cation were used. A modified toluene extraction procedure according to Hilber et al. (2017) was used to determine the total concentration of 18 extractable structures of PAHs (polycyclic aromatic hydrocarbons). The obtained extracts were concentrated using a Hei-Vap vacuum evaporator (Heidolph, Scwabach, Germany). For the quantification of 18 structures of PAHs compounds namely naphthalene, acenaphthylene, acenaphthalene, fluorene, phenanthrene, anthracene, fluoranthene, pyrene, benzo(a)anthracene, chrysene, benzo(b)fluoranthene, benzo(a)pyrene, dibenzo(a,h)anthracene, benzo(g,h,i)perylene, ideno(1,2,3)pyrene, 1-methylnaphtalene and 2-methylnaphthalene, high-performance liquid chromatography HPLC (1260 Infinitym AGILENT Technologies) was employed. Total concentrations of carbon (C), hydrogen (H) and N in BC and SS samples were quantified by CHNS-O elemental analyser (CHNS-O EA 1108; Carlo Erba Instruments, Milan, Italy). For the determination of total concentrations of arsenic (AS), cadmium (Cd), chromium (Cr), copper (Cu), nickel (Ni), lead (Pb), Zn, iron (Fe), and mercury (Hg), the method of X-ray fluorescence spectrometry (RFA) was used by Spectro Xepos-XRF meter (Spectro AnalyticL Instruments GmbH, Kleve, Germany).

Short-term ecotoxicological tests

The short-term germination test (7 days) with cress seeds (Lepidium sativum) was performed according to the scheme described in our previous work (Frišták et al., 2022). Nine testing containers were prepared for the test, three serving as a control, three containing 1% (w/w) amendment of BC and three containing 1% (w/w) amendment of SS. Washed and chemically pure sand served as a medium. The containers were divided into three levels and the effects of contaminants (gaseous, water-soluble, and bound to BC and SS) on the germination of cress seeds were studied. At the first level, the effect of volatile compounds released from BC and SS materials on germination was studied. At the second level, the effect of gaseous contaminants, water-soluble contaminants, and direct contact of seeds with BC and SS materials was studied. At the third level, the effect of volatile compounds and water-soluble compounds from BC and SS materials on seed germination was studied. To assess the suitability and toxicological safety of the analyzed BC and SS samples as soil additives, a long-term ecotoxicological test (120 days) using selected individuals of the California earthworm (Eisenia fetida) obtained from a vermicomposting station with certified origin was performed. Containers with conventional commercially available garden substrate (Natura, Agro CS, Slovakia, pH 6.8, made only from natural raw materials: peat, coconut peat and clay, with a modified limestone reaction) were amended with 1% (w/w) addition of SS and 1% (w/w) addition of BC. Containers with non-amended substrate represented control. In each container 10 adult earthworm specimens were added. The soil moisture was monitored by moisture sensors EcoWitt WH51 (EcoWitt, Shenzhen, China) and adjusted between 50 and 60% during experimental period. Once a month the counting abundance of earthworms, measurement of size and application of a feeding ration consisting of coffee grounds, carrots and leaf lettuce (3:2:1 ratio) were done. Feed rations were always applied in equal amounts to each experimental container and manually incorporated to a maximum of two cm depth of substrate. After 120 days the specimens were left emptied and executed by ethanol method. Tissues were digested by a microwave assisted digestion system (Multiwave GO PLUS; Antoon Paar, Graz, Germany) with HNO₃ and characterize by spectral analysis for the determination of selected heavy metals by atomic absorption spectrometry with electrothermal atomization (ET—AAS) and with flame atomization (F—AAS) using an Agilent 240FS AA spectrometer (Agilent).

Pot experiment with Zea mays L.

To assess how the BC and SS samples as soil additives affect plant growth and metal accumulation in the plant, a long-term outdoor pot experiment with maize (Zea mays L.) was carried out. The arable soil used as a substrate in the experimental pots was collected at the site of the village of Zákamenné in North of Slovakia (49°24′44.3″N19°17′14.6″E) in April 2023 where the experiment was set up. After collection, the substrate was air-dried, homogenized, stripped of plant and animal residues and sieved to a fraction >two mm using standard sieves. Subsequently, nine containers were applied, three serving as control (K1-3), three containing substrate with 1% (w/w) addition of BC (BC1-3) and three containing substrate with 1% (w/w) addition of SS (SS1-3). Preparation of the containers consisted of creating a drainage layer consisting of mesh and gravel, weighing the substrate and 1% additions, which were thoroughly mixed into the substrate. Maize seeds were soaked in rainwater for 3 days before being placed in the containers, to identify damaged seeds and to accelerate germination. Subsequently, into each pot were placed 6 seeds, which were allowed to germinate, and after germination, just one individual plant in each pot was left. On the day of planting, the substrates were moistened with rainwater to ensure sufficient moisture for germination. The randomized experimental set-up of pots has been used. External environmental conditions such as daily temperature, humidity and volume of rainfall were monitored throughout the experiment. The experiment was finished after twenty-two weeks by harvesting the plant objects as well as taking representative soil substrate samples for subsequent chemical and microbiological analysis.

Physico-chemical and microbiological analysis of plant and soil samples

Obtained maize plant material was separated into root system, stems, leaves, flowers and fruits, which were washed with deionized water, oven-dried and homogenized. The plant materials were digested by Multiwave Go Plus microwave digestion system (Anton Paar) with HNO₃ and analysed for the determination of As, Cd, Cr, Cu, Ni, Mn, Pb, Zn, Fe, Hg, P concentrations by inductively coupled plasma optical emission spectrometry (ICP-OES) method as well as inductively coupled plasma mass spectrometry (ICP-MS). The ICP-OES analysis was performed using an ICP-OES 5100SVDV spectrometer (Agilent) in axial view. The ICP-MS analysis was performed using an ICP MS 7900 spectrometer (Agilent) in both gas-free and He-mode.

Physicochemical characterization of soil samples after the experiment and substrate was carried out to determine pH, EC and carbonate content. The methodology is described above. Microbial analysis of soil samples after maize harvesting was performed using ARISA (Automated ribosomal intergenic spacer analysis), in which total metagenomic DNA (mgDNA) is isolated from control soil substrate as well as soil substrates with 1% addition of SS and BC after the end of the maize sowing experiment. For analysis, DNA was eluted in 50 µL of sterile water and then the concentration and purity of the DNA samples were measured spectrophotometrically using a NanoDrop One instrument (Thermo Fisher Scientific). Later, the samples were diluted to a uniform concentration of 25 ng/µl and stored at −20 °C. ITSF and ITSReub primers (Cardinale et al., 2004) were used for PCR (polymerase chain reaction) amplification of the bacterial 16S-23S rRNA variable ITS (intergenic transcribed spacer) region, and the reverse primer was fluorescently labeled at the 5′-end with 6-FAM. The length-variable DNA region of fungal ITS1—5.8S rRNA—ITS2 was amplified using the 2234C and 3126T primer pairs (Sequerra et al., 1997), and in this case the forward primer was fluorescently labelled at the 5′-end using 6-FAM. PCR amplification was performed in a GeneAmp PCR System 9700 (Applied Biosystems) under the following conditions: initial denaturation for 3 min at 94 °C; followed by 35 cycles of denaturation for 45 s at 94 °C, annealing of the primers for 1 min at 60 °C, and polymerization for 2 min at 72 °C; and final polymerization for 10 min at 72 °C. The PCR products were subsequently precipitated with ethanol and dissolved in 10 µL of sterile water. To 1 µL of purified PCR product was added 9 µL of formamide containing LIZ1200 size standard (Applied Biosystems), the mixture was denatured for 3 min at 95 °C and separated by capillary electrophoresis in an ABI 3100 Prism Avant instrument (Applied Biosystems). Data obtained from fragment analysis with a fluorescence value greater than 30 were analysed using Peak Scanner 2 (Applied Biosystems) software, and PCR product sizes ranging from 200 to 1,000 bp were used for statistical evaluation.

Experimental design and statistical analysis

All physicochemical characterizations were performed in a minimum of three or four repetitions. In the case of elemental analysis, this involved the repeated determination of a single representative sample with a standard error of measurement not exceeding 5%. The data obtained were subjected to basic statistical processing to obtain the mean and standard deviation values. For statistical assessment and graphical processing of data from characterization and cultivation experiment the ANOVA with Tukey post-hoc test (significance level α = 0.05) has been used. All analyses were performed with SPSS v. 26 (IBM Corp., Armonk, NY, USA) and OriginPro 2016 (OriginLab Corporation, Northampton, MA, USA). For microbial analysis, Venn diagrams were created using an online tool at: https://bioinformatics.psb.ugent.be/webtools/Venn/. Standardized skewness and standardized kurtosis were used to determine the normal distribution of the data. As their values were not outside the expected range, statistically significant differences between variants were assessed using ANOVA at the 95% significance level followed by a post hoc LSD test for pairwise comparisons with a 95% confidence limit. Simpson, Shannon, Evenness and Chao-1 indices were used to detect alpha diversity of the microbial community using the PAleontological STatistics (PAST) software version 3.19. This statistical software was also used for principal component analysis (PCA). Principal component (PC) scores with a variability of over 1% from the PCA analysis were used for ANOVA analysis and to detect statistically significant differences in microbial diversity among treatments in the multivariate PCA analysis.

Physico-chemical characteristics

Based on the determination of the potential pH of the BC and SS samples (Table 1), it can be concluded that the pyrolysis treatment of the distillery sludge at 400 °C increases the pH of the material. The electrical conductivity (EC) is a parameter that characterizes the representation of the total amount of water-soluble salts in the studied materials. In our case, the pyrolysis treatment of the distillery’s sewage sludge at 400 °C resulted in a significant reduction of the EC value by more than 50% (Table 1), indicating that the representation of water-soluble salts in the BC was reduced. In the SS sample as input material, the ash content was 23.59% and after pyrolysis treatment at 400 °C, the ash content in the biochar obtained increased to 50.99%. In the analysis of our samples, we found that pyrolysis treatment of sewage sludge at 400 °C resulted in approximately twofold decrease in CEC in BC (55.656 cmol/kg) compared to SS (118.66 cmol/kg) (Table 1). Pyrolysis treatment of the centrifuged sewage sludge enriched the obtained BC solid material in total C, with a significant increase in the C_org (organic carbon) form (36, 83% for BC and 28.86% for SS) and a decrease in the C_inorg (inorganic carbon) form (1.73% for BC and 2.65% for SS). The pyrolysis treatment of the sludge reduced the % of bound H in the obtained biochar (Table 1), and the H/C ratio also decreased significantly (from 0.234 for SS to 0.051 for BC). After pyrolysis treatment, the C/N ratio also increased in BC compared to SS. RFA analysis of both samples determined the highest concentration of Fe (135,000 mg/kg for BC and 107,000 mg/kg for SS) and Zn (1,656 mg/kg for BC and 593 mg/kg for SS). Other identified heavy metals and potentially toxic elements were represented in the samples in the following order in terms of decreasing concentration Cu >Ni >Cr >Pb >Hg >Cd >As. The RFA analysis values obtained for the total concentration of metals and metalloids present in BC and SS (Table 1) indicate that pyrolysis treatment at 400 °C results in approximately a twofold increase in Cr, Pb, Cu and a threefold increase in Ni and Zn in BC. For BC and SS samples, the concentrations of As and Cd were below the detection limit of the analytical method (0.1 mg/kg). The amount of Hg in the BC sample was reduced by pyrolysis treatment to below the detection limit (0.2 mg/kg).

The total concentration of polycyclic aromatic hydrocarbons (PAHs) present was reduced by pyrolysis treatment of the industrial sewage sludge to 3.38 mg/kg in the final BC material from 4.03 mg/kg in SS (Table 2). In general, the concentration of the individual PAHs of the structures was reduced by pyrolysis treatment at 400 °C, with naphthalene being the most abundant in both materials (1.82 mg/kg for SS and 1.68 mg/kg for BC). In the toluene extracts of the analysed materials, other 4-ring and 3-ring PAHs structures were also identified above the detection limit of the assay, as well as the structure of a single 5-ring PAH, namely benz(b)fluoranthene. Benzo(b)fluoranthene was also the only PAH structure detected whose total extractable concentration was increased by pyrolysis treatment in the BC compared to the input precursor SS.

Short-term ecotoxicological tests

Graphical representations of the results of the three-stage ecotoxicological germination test are shown in Fig. 1. In all three levels, i.e., when considering I. the influence of volatile contaminants, II. the influence of volatile contaminants, water-soluble contaminants and direct contact with BC and SS, and III. the influence of volatile and water-soluble contaminants, we did not observe a statistically significant difference between the additions (α = 0.05). Comparing the level II control group and with the group with 1% addition of BC, we found that seed germination did not increase statistically significantly with the addition of biochar. Even, substrate enrichment with 1% addition of BC inhibited germination most at level II, when seeds are in direct contact with biochar while being exposed to volatile and water-soluble contaminants from BC. However, even in this case, no statistically significant difference was confirmed at the α = 0.05 level of statistical significance. In the case of 1% addition of untreated SS, seed germination is most inhibited at level III, where seeds are exposed to volatile and water-soluble contaminants.

We also assessed the ecotoxicity of BC and SS in terms of their impact on soil biota by conducting a 4-month incubation test with soil giant nematodes, specifically California earthworms (Eisenia fetida). During the exposure period, 120 days, we monitored earthworms every 30 days, and mortality of 10 individuals in all groups began to be observed after three months. The mortality of earthworm individuals after 90 days of exposure was 30% in the control sample (without BC and SS addition), 60% in the sample with 1% SS addition, and 20% in the sample with 1% BC addition. We did not observe cocoon formation in either sample for the duration of the experiment. After the end of the experiment, an exposure period of 120 days, there were seven original individuals in the control sample, two individuals in the substrate with 1% SS addition and four individuals in the substrate with 1% BC addition. Earthworm mortality decreases in the order SS >BC >CONTROL, while the average weight of surviving individuals increases in the reverse order. The total concentration of selected heavy metals (Fe, Zn, Cu, Pb and Cd) in the earthworm digests obtained from control, 1% SS and 1% BC additions were determined using both F-AAS and ET-AAS, and all the values obtained did not exceed 1 mg/kg. The Fe value in individuals from the 1% SS exposure was almost threefold higher compared to individuals from the 1% BC exposure, and almost sixfold higher compared to the control group. Similarly, the amount of Cu and Cd was lower in individuals from the 1% BC exposure compared to individuals from the 1% SS exposure. There was also no statistically significant difference between the amount of Zn in individuals from the 1% BC exposure compared to individuals from the 1% SS exposure. In general, total heavy metal levels decreased in the control and 1% BC groups in the order Zn >Fe >Cu >Pb >Cd and in the 1% SS group in the order Fe >Zn >Cu >Pb >Cd.

Pot experiment with Zea mays L.

Throughout the outdoor experiment, we recorded the height of the plants in the pots every 7 days, based on which we established a trend of plant growth. From the data obtained, we conclude that the growth of the above-ground part of the maize was slower in all pots for the first 30 days. After one month, the growth rate of the plants increased, and we observed the greatest increase 60-80 days after planting. After about 80 days from planting, we observed a decreasing trend in plant growth. From the data obtained, it is evident that the plant height was highest in the substrate enriched with 1% addition of SS followed by plants with 1% addition of BC and the lowest plants were in the control group. The average values of the dimensions (height and length, respectively) and weight after harvesting (22 weeks from sowing) and drying of the underground and aboveground parts of the plants are given in Fig. 2. From the graph (Fig. 2A), it can be seen that the weight of the above and below ground parts of the control group plants is almost the same, while the height of the above ground part is three times higher. In the case of the plants enriched with 1% SS addition, it can be seen from the graph that the difference in the weight of the biomass of the below ground part is statistically significant compared to the weight of the biomass of the below ground part of the plants grown in the substrate with 1% BC addition. In the case of comparison of above-ground part biomass weight between plants grown in 1% SS substrate and 1% BC substrate, statistical significance was not confirmed (α = 0.05). We did not observe statistically significant differences between the biomass of the underground part and the aboveground part of SS and control. The highest plant biomass yield was obtained from plants from the group with 1% SS addition. Comparison of the height and length of above-ground plant individuals grown in the control (no addition), 1% SS and 1% BC addition environments (Fig. 2B) showed a statistically significant difference between plants grown in the SS and BC environments, respectively (Fig. 2B). However, both treatments did not show statistically significant difference with control environment. The lengths of the underground parts (root systems) were without statistically significant difference among all groups analyzed.

Based on the experimental data obtained, we found that the soil substrate used has a pH in the acidic region (6.4), and after cultivation of maize, the pH increased in all soils, in the order of CONTROL <SS <BC. The most significant increase in soil pH occurred in the case of 1% addition of BC to the soil (7.21). In both groups (BC, SS), we observed an increase in pH and a decrease in C_anorg, with the largest increase and decrease occurring simultaneously in the case of soil with 1% BC addition. In the case of the group with 1% addition of SS, we observed an increase in the amount of C_anorg in the soil after the end of the experiment, which may be due to the higher amount of C_anorg in SS compared to BC. In terms of soil EC, we found that after cultivation of maize, EC decreased in the order CONTROL >SS >BC.

Total microbial biomass activity was assessed as the amount of DNA isolated per 1 g of soil substrate. The DNA isolated represents DNA from both living and dead microorganisms. Using the applied least significant difference (LSD) statistical test, there was a statistically significant difference between the total microbial biomass in the BC amended soil and the control (LSD, α = 0.05), with the BC amended soil showing the statistically lowest microbial biomass (Fig. 3). The representation of microbial biomass in the soil substrates decreased in the order CONTROL >SS >BC. Venn diagrams were used to evaluate the common and dissimilar bacterial (B-ARISA) and fungal (F-ARISA) species among the samples (Fig. 4). Soil with 1% BC addition contained the lowest number of bacterial species but on the other hand the highest number of fungal species. The number of bacterial species decreased in the order CONTROL >SS >BC, with 44 fewer bacterial species in the soil with 1% BC compared to the soil with 1% SS addition and up to 50 fewer species in the soil with 1% BC addition compared to the control. For fungi, this variation in species numbers was not as pronounced and the soil enriched with 1% BC had six more fungal species than the other samples. Of all the bacterial and fungal species detected, eight bacterial species and 15 fungal species were identified in all three samples.

From our results (Fig. 5A), it is evident that maize is a Zn hyperaccumulator as we identified increased translocation of Zn to above ground parts in all samples. Zn accumulation by maize roots was statistically significantly reduced after 1% addition of SS and BC compared to the control. For SS plants, although there was a statistically significant decrease in the amount of Zn in the stem of maize compared to the control, based on % translocation, we found that the largest amount of Zn accumulated in the aboveground part was in SS plants. The translocation of Fe uptake in the maize specimens obtained and Fe accumulation in the different plant parts is shown in Fig. 5B, and it can be observed that the difference between the groups analysed is not statistically significant in any plant part. In all cases, individuals accumulated more Fe in the roots than in the above-ground part, with the % Fe abundance in the different parts decreasing in the order R (root) >L (leaf) >F (flower) ≥ S (stem) >C (cob-fruit). The translocation of As uptake in the maize specimens obtained and the accumulation of As in the different plant parts is shown in Fig. 5C, and it can be observed that the difference between the analysed groups is not statistically significant in any plant part. For all groups analysed, the amount of As in stems, flowers and fruits was lower than the limit of detection (LOD) of the analytical method of determination used (0.4 mg/kg). Individuals from the control and SS groups accumulated more As in the roots than in the aboveground part, and in the case of BC individuals we observed an increase in translocation to the aboveground part, resulting in the same percentages of As in roots and leaves. For Cd, similar to As, we did not expect a statistically significant effect of 1% addition on Cd accumulation and translocation in maize, since both the soil substrate and the additions contained less than 1 mg/kg Cd according to the RFA analysis. The translocation of ingested Cd in the obtained maize individuals together with the accumulation of Cd in the different plant parts is shown in Fig. 5D, and it can be observed that the differences between the amount of Cd in the root, stem and leaves are not statistically significant between the analysed groups. Compared to the other plant parts, the highest amount of Cd was accumulated in roots within all groups. A statistically significant reduction in Cd content was observed only in the BC maize flower compared to the control. Also, for BC cobs we observe a reduction in Cd content in fruits and also a reduction in % translocation of Cd to fruits. The translocation of the Cr uptake in the maize specimens obtained and the accumulation in the different plant parts is shown in Fig. 5E, and it can be observed that the difference between the analysed groups is not statistically significant in any plant part. From the results obtained, it can be assumed that in the soil substrate itself, Cr was mainly represented in the form of Cr(VI), since in the case of the control, 61% of the Cr taken up was transported to the aboveground part. After application of 1% SS and 1% BC, the amount of accumulated Cr in the roots increased statistically insignificantly. For all samples, the highest amount of Cu accumulated was in the roots, with no statistically significant difference between the groups. However, in SS plants we observed that up to 68% of Cu was accumulated in the whole aerial part (Fig. 5F). From the results obtained, it is evident that 1% addition of SS and BC did not have statistically significant effect on the amount of Cu accumulated by plant parts compared to control, but in case of 1% addition of BC, there was a change in Cu translocation in maize plant parts compared to control and SS. We observe a statistically significant increase in the amount of Ni in the stem in BC plants compared to SS and a statistically significant decrease in the amount of Ni in the flowers in SS compared to control. Thus, Ni accumulates in SS and BCS plants mainly in the non-edible parts of the plants, which protects the fruits and seeds from contamination. In the case of BC maize, we observe an overall increase in the amount of Ni uptake into the aboveground part of the maize (Fig. 5G). For our samples, we identified (Fig. 5H) that it was the roots and leaves that accumulated the most Pb, while in the case of roots we did not find a statistically significant difference between the analyzed groups and, on the contrary, in the case of leaves we identified a statistically significant decrease in the amount of Pb in SS plants compared to the control. In the case of maize plant stems, we cannot assess whether there was a statistically significant difference in accumulation, as the amount of Pb in SS plants was below the detection limit of the assay analytical method used (0.1 mg/kg). The translocation of uptake of P in the maize specimens obtained and the accumulation in the different plant parts is shown in Fig. 5I, and we found that there were no statistically significant changes in the roots, leaves, flowers and fruits due to the effect of 1% addition of BC or SS compared to the control. However, in the case of SS plants, there was a statistically significant decrease in the amount of P accumulated in the stem, which is probably the result of increased translocation of P from the stems of SS plants to the fruits, in order to maintain an optimal level of P in the leaves, consumed for photosynthesis. The translocation of uptaken Mn in the obtained maize individuals and the accumulation in the different plant parts is shown in Fig. 5J, and it can be observed that the 1% addition of BC and neither SS statistically significantly affects the amount of Mn in the stems and roots. However, 1% addition of BC statistically significantly reduced the amount of Mn in leaves compared to the control and also along with 1% addition of SS statistically significantly reduced the amount of Mn in flowers compared to the control.