Calcium Preparation Aided Bioremediation of Fluoranthene-Contaminated Soil

Since natural bioremediation is a lengthy process, new bioremediation techniques should be developed to accelerate and optimize the removal of soil contaminants such as polycyclic aromatic hydrocarbons (PAHs). One of the substances that can be used to enhance PAH removal is calcium peroxide, applied as an oxidizing agent to improve soil aeration. Here, we investigated the bioremediation of soil contaminated with fluoranthene. Two doses of calcium preparation (CP), 0.29 and 0.58 g/kg, composed of calcium peroxide and calcium hydroxide, and increasing the soil pH by one and two units, respectively, were used. Fluoranthene decline was 83% and 95% for CP-aided soils, and 80% for soil without calcium preparation. During the remediation process, dynamic changes in the sorption complex of soils were found, mainly due to changes in the sum of base exchangeable cations. These changes in the first experiment period were significantly influenced by the presence of calcium preparation while in the second period by the presence of fluoranthene. The presence of calcium preparation caused changes in the microbiocenosis, especially at the higher dose. There was no effect of calcium preparation on plant growth.


Introduction
Polycyclic aromatic hydrocarbons (PAHs) are toxic and persistent pollutants that have been released into the environment for decades. Their presence in soil is one of the factors leading to soil degradation (deterioration of both biological and physiochemical properties). The possibility of PAH migration from soil to water poses a serious threat to human health. Fluoranthene is one of PAH that has been classified by the US Environmental Protection Agency as one of the 16 priority pollutants (Maliszewska-Kordybach et al., 2009;Rabodonirina et al., 2019;WHO, 1983) and used as a model pollutant for research on the biodegradation of highmolecular-weight PAHs Małachowska-Jutsz & Niesler, 2015). Its molecular structure is made up of naphthalene and benzene units connected by a five-member ring. It is cytotoxic, mutagenic, and potentially carcinogenic Rabodonirina et al., 2019).
Physicochemical and biological technologies have emerged and evolved to remediate soil contaminated with PAH , among which bioremediation is recommended as a technique for PAH removal from the soil due to sufficient efficiency and low costs. The rate and extent of PAH biological degradation depend on the number of aromatic rings in the particular PAH structure. The higher the molecular weight of PAHs (the more aromatic rings), the more pronounced the hydrophobic properties, resulting in lower bioavailablility and therefore lower susceptibility to microbial degradation (Doyle et al., 2008;Hernández-Hernández et al., 2016;Kuppusamy et al., 2017).
Efficient bioremediation generally relies on effective oxygenation, which is currently achieved via oxygen-releasing compounds (ORC). Among them, sodium percarbonate and inorganic peroxy compounds are widely used (Lemaire et al., 2013;Walawska et al., 2007). The decomposition of ORC provides hydroxyl radicals (OH*), which may be a substrate for the direct oxidation of PAHs due to the high reactivity of E° = 2.72 V (Schwarz & Dodson, 1984). These radicals react with almost all organic pollutants, facilitating decomposition (Lemaire et al., 2013).
Calcium peroxide (CP) is one of the most versatile and safest ORC (Turek-Szytow, 2016;Wang et al., 2016). Lime (CaO), hydrated lime Ca(OH) 2 , and hydrogen peroxide (H 2 O 2 ) are the by-products of slurring CP in water; the released hydrogen peroxide decomposes, resulting in the release of active oxygen. The following reaction (1) is as follows: Calcium peroxide is orders of magnitude less water-soluble than sodium percarbonate (Dugan & Pratt, 2013), which allows it to release oxygen over prolonged periods. The amount of released oxygen depends on the chemical and physical properties of the surrounding environment. However, the effect of calcium peroxide should not only be examined for its impact on pollutant removal but also regarding the impact on the environment in which it is used, including biocenosis. The presented research aimed to determine the ability of calcium preparation, containing calcium peroxide and calcium hydroxide, to release oxygen and alkalize the soil environment as well as assess the influence of the preparation and its decomposition products on biological properties of water-saturated soil, especially fluoranthene-contaminated soil.

Microcosm Configuration and Soil Samples
The soil selected for the study was non-contaminated meadow soil from the vicinity of Lake Pławniowice, Poland (50°23'15.8"N 18°30'40.1"E). It was chosen because of its low pH and typical management. It was collected for research in late autumn, probably after organic fertilization (high content of organic matter). It belongs to rusty soils, as evidenced by a small layer of humus, sandy grain size, very acidic reaction, and a brown-yellow color. The following soils may occur in this area: podzolic, rusty, tan, brown leached, and acid brown. These are typical agricultural soils in Poland (Polish Soil Science, 1989). The sampled soil was homogenized, and the larger particles were removed through sieving (2 mm). The basic physiochemical properties of the soil were as follows: pH (in KCl) of 5.5 ± 0.1 and organic carbon content of 20 ± 3%. The water-holding capacity (WHC) of the soil mixture was about 680 ml/kg of soil. The soil microcosm system consisted of six cylinder-shaped plastic tanks with a capacity of 6 L with a perforated foil top cover to protect the contents from light and at the same time to allow gas exchange. The containers (C, R, P1F, P2F, P1, P2) were filled with 5 kg of soil each. During the experiment, the contents of the tanks were periodically mixed by hand after adding water for adjustment of moisture. The scheme of the system and the main differences among the tested microcosms are presented in Fig. 1a.
Calcium preparation (CP) with peroxide properties, which is a powdered mixture of CaO 2 and Ca(OH) 2 produced by the Institute of Inorganic Chemistry (IChN, Poland), was applied for the soil treatment. The content of CaO 2 in the CP was 17.0% ± 0.1% w/w of available oxygen. Selected doses of CP were predicted to increase the soil pH by 1 (P1) and 2 (P2) according to patent description (Turek-Szytow et al., 2018). It was assumed that the selected CP doses stimulate the activity of the indigenous microflora and neutralize the acidic nature of the environment. Soil without amendments (C) was used as a control microcosm, whereas soil without bioremediation enhancement (R) was used as a reference. The reference soil (R) was subjected only to basic bioremediation (no CP addition or other remediation treatments were applied). Basic bioremediation, alternatively called passive bioremediation, intrinsic bioremediation, or natural attenuation, should be understood as a natural degradation process that depends only on the metabolism of native microorganisms. The containers were incubated at room temperature (21 ± 1 °C) for 30 days starting from the addition of CP; they were covered with perforated foil to eliminate sudden changes in moisture levels due to evaporation. Throughout the research period, the soils were regularly watered to keep the moisture at a constant level of 65% WHC. Samples were taken during the experiment for determination of (a) fluoranthene concentration (days 0, 8, and 30); (b) soil physiochemical properties, bacteria, and yeast count and plants growth (days 0, 8, and 30); and (c) mutagenicity of soil (days 1, 8, and 30). The analyses were carried out in triplicates at room temperature.

Chromatographic Determination of the Fluoranthene Concentration
The fluoranthene was extracted from soil by cyclohexane, using a continuous extraction (4 h, 9 extraction cycles) performed in an automatic Soxhlet extractor (Buchi, type B-811) (Małachowska-Jutsz & Niesler, 2015). The extracts were evaporated to constant weight, dissolved in 2 ml of methanol and purified on glass fiber filters with a pore size of 0.2 microns (Supelco). The extraction efficiency of FLN from the sand was 76.5%. Determination of FLN concentration was carried out via high-performance liquid chromatography (HPLC), using an autosampler ASI-100 (Dionex) equipped with a pressure pump P580LP and a detector UVD 340 (Gynkotek) for the observation of spectra at four selected wavelengths and the absorbance spectrum in 3D. A guard column LiChroCART®100RP-18 (5 μm) with a length of 25 mm and a chromatography column LiChroCART®100RP-18 (5 μm) with a length of 250 mm (Merck) were used. The mobile phase was composed of a multi-step gradient of a mixture of methanol and water at a flow rate of 1 ml/min. Analysis time was 70 min, and the detection of FLN followed at a retention time of about 38 min and a wavelength of 276 nm. A liner-type standard curve was used (correlation coefficient 99.2%) in the range of 0.4-60.0 ug FLN/ml. For each sample, the extraction was carried out once, while the analysis of the concentration of fluoranthene was carried out 2-4 times. The FLN removal efficiency was calculated as a percentage of FLN removed during 30 days in relation to its concentration at the start of the experiment and expressed as a percentage. The relative FLN concentration was calculated as the ratio between the concentration on a given day to the concentration at the beginning of the experiment.

Counting of Bacteria and Yeasts
Inoculations of soil samples were performed to assess changes in the amount of the soil microflora during the treatment. Bacterial and yeast counts were determined by plating at nutrient agar and Czapek agar, respectively. For this purpose, 10 g (wet weight) of soil sample was suspended in 100 ml of 0.85% NaCl and homogenized for 60 min on a shaker. Samples (0.1 ml) of several dilutions were spread on the used culture medium, and the nutrient agar plates and Czapek agar plates were incubated at 22 ± 2 °C for 24 h and 5 days, respectively. After the incubation time, colony-forming units (CFU) were counted and calculated per kg dry wt. Relative bacterial or yeast counts were calculated as the count on a given day concerning the count at the beginning of the experiment.

Seed Germination/Plant Growth
Seed germination and elongation tests were performed to assess the toxicity of soil during the treatment. Sinapis alba (mustard) and Lepidium sativum (garden cress) were used for seed germination assays and root elongation tests, respectively (PN-ISO 11269:2001). For this, 25 mustard seeds were sown separately in glass jars containing 10 g of soil re-wetted to 80% WHC (in triplicate). Lids were loosely screwed on to reduce evaporation but allow aeration, and the seeds were left to germinate at 20 °C for 24 h. At the end of the incubation, the number of germinated seeds in all soil samples was recorded. The same procedure was applied for cress seeds, but the length of the root and the hypocotyl were measured. Seed germination was calculated as a ratio of the number of grown seeds to the number of seeds used. Cress root elongation was calculated as a ratio of the length of the hypocotyl and the length of the root. The L. sativum seedlings growth was calculated as the ratio of the whole plant length and the length of the plant in the control sample (microcosm C).
Relative seed germination, relative root elongation or relative seedlings growth was calculated as the value on a given day to the value at the beginning of the experiment.

Mutagenicity of Soil Samples
Three bulbs of Allium cepa (onion) were cultured in each of the sampled soil. Plants were cultivated at a 16 h/8 h photoperiod and irradiance of the summer sun (PN-ISO 11269:2001). The soil in the containers was rinsed daily with tap water to keep the moisture level at 80%. After 5 days of cultivation, bulbs were withdrawn and carefully cleaned. Subsequently, they were washed, and root pieces of 20 mm in length were cut, fixed in Carnoy solution (glacial acetic acid/ethanol 1:3), and incubated for 24 h at 4 °C. Roots were then hydrolyzed for 10 min in 1 M HCl at 60 °C, and the growth cones of three roots were sectioned and stained with 1% aceto-orcein under a microscope slide. After staining, the interphase cells inside the root sections were observed under a microscope. Dividing cells and cells with micronuclei were searched and scored. In each of the three repetitions, 1000 cells were examined. The mitotic index (MI) was expressed as the number of dividing cells per 100 scored cells (%); hence, the micronuclei frequency (MN) was defined as the number of cells with micronuclei per 1000 cells scored. Relative mitotic index was calculated as the ratio of the value on a given day to the value at the beginning of the experiment. Negative and positive controls were prepared using chlorine-free tap water and maleic hydrazide (MH, C 4 H 4 N 2 O 2 ) at a concentration of 0.45 g/l (Wiszniowski et al., 2009).

Data analysis
Results are presented as a mean ± standard deviation of the mean. Shapiro-Wilk tests were employed to determine whether datasets were normally distributed. Analysis of variance (ANOVA) tests was used to determine if samples differed significantly. These tests were performed to determine significant differences between the tested soils and between the chosen sampling points. If not indicated in the text, statistical significance was accepted when the probability of the result assuming that the null hypothesis (p) was less than 0.05.
Principal components analysis (PCA) was used to assess the changes occurring in the soil during amendment with calcium preparation and/or fluoranthene (Güler et al., 2013;Thyne et al., 2004). Kaiser's criterion (Kaiser, 1960) was used to choose the number of PCs. All analyses were performed using MS Excel and XLSTAT (Adinsoft, 2010).

Fluoranthene Concentration During Soil Bioremediation
The dose of fluoranthene that was added to the soil to imitate contamination was 5 mg/kg dry wt. The FLN concentration measured 7 days from contamination (day 0) in fluoranthene-contaminated soils was up to 20% lower than the calculated beginning concentration (the concentration of FLN from 3.95 to 4.84 mg/kg dry wt soil). This may result from the sorption of this component and the efficiency of the fluoranthene determination method. The reduction of the initial FLN concentration is beneficial due to the avoidance of its probable toxic effects on soil microorganisms. However, the immobilization of FLN on soil particles can inhibit or prolong its transformation (Medina et al., 2018). The soils not subjected to FLN (C, P1, and P2) also contained small amounts of FLN of 0.32 ± 0.01 mg/kg dry wt. FLN in soil may naturally originate from combustion of fossil fuels, peatlands, or nearby grasslands burning, or even from the degradation of natural organic matter. Zeng et al. (2010) reported that the concentration of FLN in naturally contaminated soils was 1.2 mg/kg dry wt; in this sense, our values obtained in excavated soil are relatively low. For this reason, soils C, P1, and P2 were considered non-contaminated (or not artificially contaminated). The concentration of the FLN measured on day 1 (7 days after contamination and one day after adding CP) was 4.40 ± 0.50 mg/kg dry wt soil. The higher FLN concentration compared to 24 h from adding CP might be a result of desorption processes initiated by the presence of calcium preparation. It is possible that calcium ions from CP displaces fluoranthene into the soil solution, makes it available for biota, and allows its determination (Lu et al., 2017;Strawn, 2021).
During the 30 days of the experiment, both in control soil and CP-treated soil, the concentration of FLN decreased to a level presented in Table 1. Considering non-artificially contaminated soils, a higher FLN removal efficiency was observed for soils C and P1. The FLN removal was significantly lower in P2 than in C and P1, revealing the toxic effects of a higher dose of CP on biological and chemical processes involved in FLN decomposition. Among the soils treated with FLN (P1F, P2F, and R), the FLN removal was higher (P2F) or slightly higher (P1F) in soils treated with calcium preparation than in reference soil that was not subjected to CP. In the FLN-contaminated soils, the highest removal was obtained for the soil that received the higher CP dose (P2F).
The differences in CP effect on non-FLN-contaminated and contaminated soils are most likely a result of the nature of hydroxyl radicals that were released to the soil matrix during CP decomposition. In contaminated soils, the radicals were focused on the oxidation of organic compounds, i.e., soil organic substance including FLN, that occurred in the soil in excess. However, in the non-contaminated soil treated with a higher dose of CP (P2), the radicals affected rather the biological activities, thus resulting in a lower FLN decline (Lu et al., 2017). It should, however, be mentioned that for all tested soils, the decrease in FLN concentration after both 8 and 30 days of the experiment was statistically significant (p < 0.05). The loss of fluoranthene could have been caused by biodegradation. If so, the decomposition potential of the microbes of the tested soil seems to be high.
However, the determined lower concentration of FLN may be due to chemical oxidation and/or sequestration in the organic or mineral fractions of the soil.
A part of soil organic matter which is mineral-associated organic matter is not permanently sorbed to minerals. It may de-sorb and enter again soil solution as dissolved organic carbon, be transported to another parts of the soil or used by microorganisms. It may also be again released (Dynarski et al., 2020). However, the dynamics of FLN removal in the reference soil were different compared to that in the soils amended with CP (Fig. 2).
The rate of FLN removal assisted by CP was higher at the beginning of the experiment. Thus, the time to reach the selected FLN level was shorter when CP was applied. This means that in cases where it is necessary to remove FLN relatively quickly from the soil, the addition of CP can be a good solution for such a challenge. Although the concentration of FLN after 30 days of the experiment was similar in the considered microcosms, it is clearly visible that in soils with calcium preparation, a given level of FLN concentration was obtained at least 10 days earlier than for soil without CP addition. This observation is important in a light of the time necessary for the bioremediation of contaminated sites. However, a lower concentration of FLN in soils amended with calcium preparation could have been caused also by an enhanced sequestration of organic compounds in the presence of calcium ions (Dynarski et al., 2020). In soils, calcium co-associates with organic carbon and ferrous oxides, though the bonding mechanism (e.g., cation bridging). A clear indication of the main causes of the loss of fluoranthene would require additional and more in-depth research. Moreover, an important analytical aspect should also be noted. In Table 1 Fluoranthene concentration and its removal efficiency obtained after 30 days of the experiment (C, control soil, i.e., one that did not contain any additives; R, reference soil, i.e., one to which only fluoranthene was added; P1, P1F, non-contaminated or fluoranthene-contaminated soil, respectively, adjusted with 0.29 g of calcium preparation per kg dry wt; P2, P2F, non-contaminated or fluoranthene-contaminated soil, respectively, adjusted with 0.58 g of calcium preparation per kg dry wt; different letters indicate significant differences between soils (p < 0.05)) the fluoranthene determination method, the extraction efficiency is 76%. This means that only 76% of the component present in the soil can be determined. Depending on the method used, these efficiencies of PAH extraction range from 65 to 99% (Lau et al., 2010;Maliszewska-Kordybach & Oleszek, 1994). Taking into account the dynamics of changes taking place in the soil, the lability of various forms of compounds, including organic ones, and the multitude of biochemical and chemical reactions, it is probable that the extraction efficiency may also change. If the extraction efficiency decreases, there is a risk that the determined lower concentration of the compound does not result from decomposition processes. For this reason, the results obtained in this section should be considered illustrative.

Fluctuation of Physio-chemical Soil Parameters During the Experiment
The soil properties have been analyzed in all sampling points. However, the study aims to focus on the biological side of conducting and supporting bioremediation; therefore, the physicochemical changes in the soil will be discussed in general terms, focusing on the significant changes observed during the experiment. Moreover, all results are used for the factor analysis presented in Sect. 3.5. The results carried out before the addition of calcium preparation to the selected microcosms are presented in Table 2. Most of the presented parameters are the same or similar for non-contaminated and FLN-contaminated soil, e.g., water-holding capacity, moisture, hygroscopic moisture content, pH, and sum of base exchangeable cations. The organic substance content was between 16 and 25%. These differences are large, considering the fact that the same soil was found in the individual microcosms. The organic carbon content was about 4% in most of the samples, but incredibly high in microcosm P2 and P2F. The addition of fluoranthene should not change the organic content indicators significantly or noticeably. These differences may result from the heterogeneity of soil samples that were previously fertilized organically (hence the high value of this parameter). The analyses were most likely performed in too few repetitions in relation to the heterogeneity of the sample (variation coefficient was up to 20%). Soil organic matter is not evenly distributed within soil; the distribution and identity of soil organic matter compounds vary even at micro-and nanometer scales (Dynarski et al., 2020); thus, soil analysis requires usually more than traditional three repetitions. As an excuse for less accurate results, the illustrative nature of the physico-chemical analyses should be assumed, which were only to provide a background for the biological part, which was the core of the presented experiment. Nevertheless, the analyses indicate the organic nature of the soil, which is related to the use of the oxidizing nature of the preparation used and its decomposition products (active oxygen, hydroxyl radicals, or hydrogen peroxide). Oxidizing compounds will oxidize not only the supplied FLN, but also other components of soil organic matter. The organic nature of the soil will therefore result in an increased demand for the preparation used and thus also increased costs of the applied treatment. Free radicals, especially hydroxyl radicals that are produced during calcium peroxide decomposition, have been reported to have strong chemical activity. This leads to a nonselective degradation of both persistent organic pollutants and organic substances (Miksch et al., 2015). Moreover, organic substances as organic acids, amino acids, and others are natural chelating agents which may increase hydroxyl radical production by enhancing the presence of transition metals in aqueous phase (Zhang et al., 2015). Additionally, results of Bao et al. (2022) show significant enhancement of organic matter degradation and promotion of humic substances formation when calcium peroxide was added to the soil. All these issues show how many different changes of a chemical nature can affect the disappearance of organic compounds, including FLN, apart from biodegradation. All of them could have been affect the biological properties of the soil of interest. The soil buffering capacity was around 21% lower in FLN-contaminated soil in comparison to non-contaminated soils which results from the organic nature of the contamination. FLN could saturate the exchangeable ions present in the soil sorption complex, thus lowering the soil buffering capacity. The lower value of Kjeldahl nitrogen (the sum of organic and ammonium nitrogen) may result from the microbiological activity aimed to oxidize these nitrogen compounds, which is reflected in the increased content of nitrates in samples contaminated with FLN.
After 24 h from CP addition, pH increased by one unit in a lower dose of calcium preparation soils (P1, P1F), i.e., pH of 6.5. Meanwhile, adding a dose twice as high (P2, P2F) resulted in a pH increase of 1.3-1.5 units, i.e., pH of 6.9 and 6.7, respectively.
In the consecutive measurement days, a decrease in pH was observed in all CP-adjusted soils, especially in the first 7 days. Changes in pH in the presence of calcium peroxide influences hydrogen peroxide generation while calcium peroxide oxidation according to the reaction (3), which may affect the decomposition of organic compounds and soil composition.
The changes observed among the physicochemical parameters of the studied soils concern, in particular, the sorption complex. These changes appear to be strongly related to the presence of fluoranthene and/or calcium peroxide. On day 8, the FLN concentration in soil amended only with the lower dose of CP (P1) was 15% higher than that in the control sample, whereas that in soil amended with a higher dose (P2) was about 70%. This might have been caused by the separation of this compound from the soil sorption complex due to the saturation of the complex with calcium ions (Ca 2+ ). In this case, fluoranthene was more readily available and can be determined at higher concentrations. Cation exchange capacity measures the ability of a soil to absorb cations in exchangeable forms and increases as the pH increases (Chesworth and (Ed.)., 2008). This parameter was slightly increased throughout the experiment and stabilized at 234 ± 6.8 mmol H + /kg dry wt on day 30, regardless of the soil modification. As noted earlier, 24 h after adding the calcium preparation, soil pH increased. In the following measurement days, a decrease in pH was observed in these samples. After 30 days of testing, the pH in soils with a lower dose of the preparation was only 0.3-0.4 times higher (pH 5.7 and 5.9 for P1 and P1F soils, respectively) than the initial level. Meanwhile, in the samples with a higher dose of CP the pH was 0.8 times higher than the beginning (pH equal to 6.3). Soil enriched with CP or FLN revealed a pH of 5.7. The changes in cation exchange capacity seem to be the most associated with changes in hydrolytic acidity. HA increased throughout the experiment in each of the soil, but in the soils C, P1, P1F, and R, the level of hydrolytic acidity was similar, while in the P2 and P2F, it was lower than in other soils by approximately 20% at days 1 and 8, and 5 and 15% lower at the end of the experiment, respectively for P2 and P2F. After 30 days of the experiment, the hydrolytic acidity was around 4-4.5 times higher than at the beginning of the experiment.
There were no significant differences between the individual microcosms in terms of base exchangeable cations, cation exchange capacity, and base cation saturation ratio values. The base cation saturation ratio was comparable between the samples, although it decreased in all soils by the 8th day of the experiment and stabilized at the level of 53 ± 3%. Buffering slightly decreased during the experiment in all samples, however, two groups of soils can be distinguished according to the buffering value-C, P1, P2, and P1F, P2F, and R. This means that the presence of fluoranthene not only after the addition of FLN but also in the course of the experiment lowered the buffer values of contaminated soils.

The Fate of Soil Microorganisms
At 1 week after the contamination of the microcosm P1F, P2F, and R with FLN, the bacterial count in soil decreased from 83 ± 38 × 10 8 CFU/kg to 42 ± 38 × 10 8 CFU/kg dry wt that is by 50%. However, there were no statistically significant differences at the confidence level of 95% (p < 0.05). At the same time, the yeast count remained unchanged (from 159 ± 24 × 10 8 CFU /kg dry wt to 158 ± 54 × 10 8 CFU /kg dry wt). On day 1, all soils showed a low bacterial content of no more than 30 × 10 8 CFU/kg dry wt. Not taking into account the incomprehensible decrease in the number of bacteria on day 1, the number of bacteria in the soil C and P1 remained at a similar level throughout the experiment, while the other CP amended soils revealed a higher number of bacteria on day 8. Especially in soils P2 and P2F, the count was higher by about 3-3.5 times compared to the other microcosms (344 and 304 × 10 8 CFU/ kg dry wt, respectively). On the last day of the experiment, soils C, P1, and R were inhabited with 43-105 × 10 8 CFU/kg dry wt, while P1F, P2, and P2F with 5-8 × 10 8 CFU/kg dry wt. The results from the 30th day presented as relative values are shown in Fig. 3a,b. In summarizing, compared to the control microcosm, the application of a lower dose of CP to non-contaminated soil (P1) significantly stimulated bacterial growth. The addition of the higher dose (P2) resulted in the short-term stimulation of bacterial growth (day 8, data not shown) but in a further drastic decrease in the number of viable cells, resulting in the relative bacterial count after 30 days much lower than 1. Application of a higher CP dose resulted in short stimulation of both communities (P2, P2F), but after another 3 weeks of the experiment, inhibition of bacterial growth was observed in both cases. The use of a double dose of CP or a single dose but in the presence of a contaminant caused more havoc within the bacterial community than the contamination itself, which can be seen from the statistical significance analysis.
A similar tendency as for bacteria was observed for the yeast count, especially at the end of the experiment (Fig. 3b). The changes in the yeast count on days 1 and 8 do not allow any link to be made with the presence of contamination or calcium preparation. Only the addition of a single dose of CP caused shortterm stimulation of yeast observed on day 8, as in the case of bacteria. Ultimately, both doses of CP influenced negatively microbial biocenosis; however, the second dose gave a stronger inhibitive effect.  Fig. 3 Effect of CP on microorganisms in non-contaminated and FLN-contaminated soils after 30 days of the experiment. Values are mean (n = 3) ± SD. Statistical significance of the difference between day 0 and day 30 (no(n)/(yes(y)) given at p < 0.05. Different letters indicate statistically significant differences between microcosms/plots according to ANOVA results The observed changes in bacterial and yeast count in non-contaminated soils may result from physiochemical changes in the soil-especially an increase in the pH proportional to the CP dose observed on days 1 and 8, which weakened between 8 and 30 days of the experiment. This may be explained by the alkaline character of CP, which loses its alkalization properties during decomposition. Peroxide decomposition was most pronounced during the first 8 days. At the same time, free radicals, which have strong oxidative and biocidal effects, are increasingly generated (Walawska & Gluzińska, 2006). The rapid discharge of the oxygen in the first period (up to 8 days) affected the microbial activity of the soil. The activity of bacteria that dominate in a neutral and slightly alkaline environment leads to the decomposition and mineralization of the pollutant. Over time, the intensity of oxygen generation decreased, and the effect of calcium peroxide was noticeable. The strong alkaline nature of CP and its sorption abilities caused an increase in the soil sorption complex capacity. The chemical oxidation of FLN resulted from the release of free radicals originating from CP decomposition, potentially resulting in the production of metabolites that are more toxic for the bacteria than CP and FLN itself. Fungi play an important role in the hydrocarbon-oxidizing activities of the soil, and they seem to be at least as versatile as bacteria in metabolizing aromatics (Mao & Guan, 2016;Quintella et al., 2019). The results obtained show that the dose of 0.58 g/kg dry wt is more toxic for yeast populations than that of 0.29 g/kg dry wt.

Plant growth dynamics
The presence of FLN itself in samples P1F, P2F, and R (i.e., before the addition of CP) demonstrated a slight simulative effect on plant growth as mustard seed germination and cress seedlings growth increased by 15 and 10%, respectively. The values for the control and reference sample at the beginning of the experiment were as follows: mustard seed germination 38.5 ± 14.6% and 44.5 ± 12.0, cress' root elongation 36.7 ± 8.1 and 33.7 ± 16.0, and cress' seedlings growth 66.3 ± 57.5 and 73.0 ± 23.4, respectively.
Mustard seed germination changed during the 30 days of the experiment, exhibiting an increase of approximately 40 and 20% in comparison to the start of the experiment for P1 and P2, respectively (Fig. 4a).
There were no significant differences between the two CP doses on mustard seed germination, although the difference between the start and the end of the experiment was significant in both cases (relative mustard germination always higher than 1). We, therefore, conclude that the changes in mustard seed germination were largely caused by other factors than the CP dose. The germination of mustard seeds in contaminated soil treated with a lower CP dose (P1F) was significantly higher in comparison to soil P1, P2, P2F, and reference. It is likely that, in contrast to the microorganisms, the metabolites of FLN decomposition, which were enhanced by the lower dose of CP, stimulated the growth of mustard (i.e., seed germination). However, the higher CP dose was excessive, suggesting that the production of free radicals was more pronounced.
The results of relative L. sativum seedlings growth and root length ( Fig. 4b and 4c) showed no differences between control and CP-amended soils, suggesting that CP did not impact these parameters. The observable changes were not significantly different, most likely because of the highly scattered measurements at the beginning of the experiment. The strong increase in plant growth compared to the values on day 0 (relative value considerably higher than 1 in most cases) may results from the low values obtained at the beginning of the experiment (day 0). Most likely, the quality of the distilled water during the analysis on day 0 could be inadequate due to a breakdown of the distiller, which consequently could inhibit the growth of the test plants on that day. The differences in L. sativum seedlings growth and root length between the start and the end of the experiment appeared to be positive for all soils but with statistical significance only for soil P1F and P2F ( Fig. 4b and 4c). However, the tendency was similar to that found for mustard seed germination, resulting in the same conclusions (Fig. 4c).
As presented in the previous section, the presence or absence of FLN, combined with the usage of different CP concentrations, can lead to different mechanisms and decomposition products. Thus, the soil was tested for its mutagenic potential for plants. Relative mitotic index (results for day 30 concerning day 0) is shown in Fig. 4d. For all tested soil samples, the mitotic index did not significantly differ from the negative water-based control (p < 0.05) and was significantly different from the positive maleic hydrazide-based control (p < 0.05). The maleic hydrazide is an herbicide that induces cell death and inhibition of mitosis in A. cepa root tips (Marcano et al., 2004). This increases the percentage of non-dividing cells and causes a mitotic index (MI) decline. A decrease in the mitotic index in comparison to the negative control implies an increase in toxicity (Wiszniowski et al., 2009). The inhibition of the mitotic activity in comparison to the negative control was never higher than 50%, a level that indicates sublethal effects (Khanna & Sharma, 2013). The highest differences concerning the beginning of the experiment were observed after 24 h. Later, the MI oscillated close to the beginning value, except for the soils amended with the higher dose of CP. After 30 days of the experiment, the MI value for soil P2 was 31% higher and for soil P2F 53% lower than at the beginning of the experiment. After 8 days of the experiment, all samples except P2F approached the initial value. This tendency was continued up to day 30, albeit not for soil P2F. Here, the frequency of mitosis was much lower than at day 0. As the results revealed a stimulating effect of the higher CP dose after 30 days, this confirms previous assumptions that the products of chemical and biological decomposition of FLN and other soil compounds enhanced by the higher CP dose (P2F) may be different from those in other soil samples, further confirming their inhibitive character.
No micronuclei, which is a small nucleus produced during telophase of mitosis (Wiszniowski et al., 2009), in the negative control and all tested soil samples were found, while the positive control was 0.6 ± 0.2. Micronuclei are produced additionally to the main nucleus by lagging chromosomes. Their presence is an indicator of exposure to mutagenic agents (Wiszniowski et al., 2009). Lack of micronuclei in the tested soils show a lack of mutagenic potential in the soils.
Summarizing the direct effect of FLN and/or CP on biota, CP negatively impacted bacterial biocenosis, mainly at the higher dose. The yeast population  Values are mean (n = 3) ± SD. Sta-tistical significance of the difference between day 0 and day 30 (no(n)/(yes(y)) given at p < 0.05. Different letters indicate statistically significant differences between microcosms/plots according to ANOVA results and mustard germination were stimulated by both CP doses, albeit more significantly by the lower dose (0.29 g/kg dry wt). The addition of CP had no impact on the growth of C. pratensis.

Influence of Fluoranthene and Calcium Preparation on Soil Parameters and Biota
To compare the physical, chemical, and biological parameters of the tested soils, principal components analysis (PCA) was performed on the set of results recorded after 8 or 30 days of the experiment. In both cases, more than 60% of the total variation was explained by the eigenvalues F1 and F2. After 8 days of the experiment, the soils amended with CP were similar, but they were different from the untreated soils (R and C). The influence of the dose was noticeable as the samples with the same CP dosage were closer to each other. The reference soil (R) differed from the others mainly by its chemical composition, whereas the control soil (C) exhibited specific biological parameters. After 30 days of the experiment, the effect of calcium peroxide was not as clear as at the beginning, and the samples differed due to the presence of fluoranthene. The oxidation products of various substances in the soil, including fluoranthene, caused by free radicals and oxygen from CP, differed qualitatively and/or quantitatively, depending on the presence and dose of peroxide and fluoranthene. Thus, the changes in physicochemical and biological parameters were also different. Decomposition of FLN results in the formation of acid metabolites, which activate the sorption complex. The ions are released from the sorption complex, which neutralizes the acidic pH of the soil. In our study, different exchange reactions were activated, leading to the maintenance of homeostasis. These processes occurred simultaneously, and the effect was enhanced by the chemical activity of CP. The efficiency of amendments is site-specific, depending on various factors such as pollutant type and soil properties (Kumpiene et al., 2008). In the present experiment, the water conditions were kept similar and constant for all tested microcosms. The obtained results suggest that the mechanisms of soil compound oxidation and degradation are different in different microcosms, depending on the presence or co-presence of CP and FLN and their concentrations.
Changing water conditions may affect further the fate of the oxidized soil compounds.
The authors of a previous study (Zeng et al., 2010;Zhou et al., 2008) showed that FLN was readily degraded only during plate culturing and not in the liquid culture. As the decomposition rate of CP depends on the water conditions, such differences may be enhanced. High-molecular-weight PAHs are more resistant to biodegradation not only due to their hydrophobic nature but also because of their stronger sorption properties onto micropores of particulates in soil matrices (Ololade et al., 2017;Rabodonirina et al., 2019). Most likely, this explains why the sorption capacity was the most changeable factor observed in all tested samples.
Principal components analysis (PCA) was employed to determine the importance of the presence of fluoranthene and its metabolites as well as calcium preparation. For that purpose, data collected on days 8 and 30 were subjected to PCA, including water holding capacity (WHC), moisture (M), hygroscopic moisture content (HMC), organic substance content (OSC), absolute dry mass content (ADM), pH, organic carbon content (OC), total nitrogen (TN) and its soluble (NO 3 -N , NO 2 -N, NH 4 -N) ion forms concentration, phosphorus (PO 4 -P) concentration, base exchangeable cations (BEC), hydrolytic acidity (HA), cation exchange capacity (CEC), buffering (B), phenol concentration (FEN), bacterial and yeast count (BC, YC), mustard germination (MG), cress root' elongation (CRE) and cress seedlings growth (CE). Figure 5 presents the PCA of both physicochemical and biological features of the tested microcosms. It should be noted that on days 8 and 30, each microcosm had its own, unique characteristics and could be easily separated from the others. After 8 days of the experiment (Fig. 5a), the soils treated with calcium preparation were similar, irrespective of the presence of fluoranthene. This means that the samples with calcium preparation, i.e., P1, P2, P1F, and P2F, were more similar than the C and R microcosms without CP. As a result of moisture present during the first week after calcium preparation application, most calcium oxide and free radicals were formed. During this time, the largest changes occurred in the sorption complex. The decrease in the buffering capacity (B) and the base cation saturation ratio (BS), caused by the sum of base exchangeable cations (BEC) decrease, was observed.
Since fluoranthene has low solubility, not all of the introduced compound is available for microorganisms. The available oxygen facilitates the oxidation of organic compounds and ammonium nitrogen by nitrifying bacteria.
After 30 days of the experiment, the similarity and correlation between the samples had changed completely (Fig. 5b).
The degradation of fluoranthene was progressing, and the presence of its metabolites caused further changes in the soil sorption complex. The buffering capacity (B) and the hydrolytic acidity (HA) increased, along with the sum of base cations (BEC). At the end of the experiment, the presence of fluoranthene and its metabolites divided the microcosms into two separate groups-soils contaminated with fluoranthene (R, P1F, P2F) and non-contaminated soils (C, P1, P2). The oxidation products of various substances in the soil, including fluoranthene with free radicals and oxygen with CP, differed qualitatively and/or quantitatively, depending on the presence of peroxide and fluoranthene. Thus, the changes in physicochemical and biological parameters differed in each case.
The highest squared cosine to the factor F1 was observed in both cases for the sum of base exchangeable cations. In the case of factor F2, the highest squared cosines were observed for WHC and M. The sum of base exchangeable cations, which is a part of the cation exchange capacity, is the most important factor in the bioremediation process. The higher the CEC, the higher the number and diversity of ions, also biogenic ions, that are released into the soil solution and are then available for microorganisms and plants. Soil functional quality is dependent on bioavailable soil compounds since this fraction plays the most important role in parameters such as enzyme activities or microbial-related processes (Cipullo et al., 2019;Ololade et al., 2017;Torri et al., 2018).

Conclusions
In the soil without the addition of calcium preparation, within 30 days of the experiment, a loss of FLN of about 80% was observed as a result of autogenous processes. The addition of 0.29 g of calcium preparation per kg of soil further enhanced fluoranthene removal after 30 days by 2%, whereas the addition of 0.58 g of CP per kg of soil resulted in a 14% higher fluoranthene removal compared to the non-amended soil. However, the growth of Sinapis alba, Lepidium sativum, and Allium cepa was stimulated by the presence of calcium preparation and/or fluoranthene. Amendment of FLN-contaminated soil with calcium preparation was detrimental for microbial biocenosis, as the bacterial and yeast counts decreased over the experimental period. Thus, bioaugmentation or additional bacterial stimulation might be recommended after bioremediation with calcium preparation.