Allelopathic potential of selected woody species growing on fly-ash deposits

The objective of this study was to determine the allelopathic potential of Robinia pseudoacacia L., Ailanthus altissima (Mill.) Swingle and Amorpha fruticosa L. that grow on the fly-ash deposits at the “Nikola Tesla – A” thermoelectric power plant in Obrenovac. The chemical characteristics of fly ash, such as pH, electrical conductivity (EC), content of carbon (C) and nitrogen (N), contents of available phosphorus (P2O5) and potassium (K2O), the contents of total and available Fe, Cu, Mn, Ni and Zn as well as of phenolic acids (3,5 dihydroxybenzoic acid (3,5-DHBA) and ferulic acid) and flavonoids (rutin and quercetin) were analyzed in control fly ash (bare zones without vegetation cover) and plant rhizospheric fly ash. In order to determine the allelopathic activity of phenolic compounds in fly ash, modified soil sandwich allelopathic biotests were performed, and Trifolium pratense L. (red clover) was used as the indicator species. A. fruticosa showed the highest allelopathic activity, followed by A. altissima whereas R. pseudoacacia showed the lowest allelopathic potential. Negative correlation was noted between radicle and hypocotyl growth inhibition of red clover and the pH of fly ash. Positive correlations were found between radicle growth inhibition and the content of C, P2O5, total concentrations of Cu, available concentrations of Mn and Ni, the contents of ferulic acid, 3,5-DHBA, and rutin. Our results indicate that A. fruticosa and A. altissima increased the content of phenolics in fly ash, which can act as allelochemicals leading to radicle growth inhibition of red clover in the pioneer plant community on fly-ash deposits. These woody species that colonized fly-ash deposits can initiate the beginning of pedogenetic processes altering the ecosystem processes at degraded sites.


INTRODUCTION
Allelopathy presents both inhibitory and stimulatory interactions between plants through the action of secondary plant metabolites -allelochemicals [1,2].Allelochemicals are often investigated as compounds that potentially allow the invasion of plant species in new habitats due to the lack of adaptive potential of native species to new allelochemicals originating from introduced species [3].One of the most studied groups of allelochemicals are phenolic compounds, which in plant tissue can be found in soluble form or bound to the cell wall polysaccharides [4,5].
Phenolic compounds from the plant organism are delivered into soil by leaching from the surface of the plant and fallen leaves, decomposition of litter, and active excretion from roots [6].Most of the phenolic compounds are soluble in water and rinsed from the surface of the plant body and transferred to the deeper parts of the soil by rainwater [7].In soil, phenolic compounds represent the second most widespread group of compounds (after cellulose), and occur in three different forms: free, reversibly bound and bound [8].However, the concentration of phenolic compounds in the soil is much lower than in plant organisms [9].These compounds have a short lifetime and they are susceptible to degradation by microorganisms, so their detection and identification in soil is much more difficult than in plant tissues [10].
Through allelopathic activity, phenolic compounds have significant effects on the structure and composition of plant communities [11,12].The phenolic compounds may have direct harmful effects due to their release by donor species or can be degraded or transformed by soil microorganisms.Furthermore, these compounds can have an effect on the physical, chemical and biological soil characteristics or can induce the release of biologically active substances in third plant species [13].Phenolic compounds in the soil can lead to root and hypocotyl growth inhibition, the limitation of water and mineral uptake, and inhibition of photosynthesis and enzymatic activity in acceptor plants [5,14,15].Blum et al. [16] indicated that the focus in allelopathic research should be shifted to the soil system since the phytotoxic activity of allelochemicals is a function of the characteristics of plant allelochemicals and soil [17].Additionally, adsorption and desorption of allelochemicals in soil is a dynamic process and is influenced by various physicochemical factors, such as soil moisture, pH and organic matter content [18,19].Furthermore, phenolic compounds represent an important link in the dynamics of mineral and organic compounds in the soil due to their effect on the chemical properties of the soil, availability of certain metals and the microorganism community [5,8,15,20,21].
The process of plant colonization on fly-ash deposits (revegetation) is very slow due to unfavorable conditions of substrates [22].The plants growing on the fly-ash deposits are exposed to a combination of several stress factors, such as high temperatures, increased irradiation, drought and elevated concentrations of some metal(loid)s [23][24][25].Woody plants that have a large cover and great biomass play a prominent role in the process of fly-ash revegetation.Organic matter originating from these plants is of great importance in the process of soil humus formation, which represents a mixture of various organic compounds among which phenolic compounds are very important.
In the majority of studies, the effect of phenolic compounds on vegetation is emphasized, while their effects on substrate characteristics are not fully addressed [26].In this study, we hypothesized that woody plant species, Robinia pseudoacacia L., Ailanthus altissima (Mill.)Swingle and Amorpha fruticosa L., show a great allelopathic potential due to the production of high amounts of phenolics that are released from these plants in fly ash.It is also assumed that these allelochemicals contribute to the increased availability of certain metals, which can additionally lead to inhibition of seedling (radicle and hypocotyl) growth of the indicator plant species, Trifolium pratense L. (red clover).Therefore, the objectives of this study were (i) to compare the chemical characteristics of control and rhizospheric fly ash; (ii) to analyze the total and available metal contents (Fe, Cu, Mn, Ni, Zn) in control and rhizospheric fly ash; (iii) to evaluate the contents of phenolic compounds in control and rhizospheric fly ash; (iv) to determine the allelopathic potential of the selected woody species through screening of seedling (radicle and hypocotyl) growth inhibition of red clover as an indicator plant species.

Study area
The study was performed on the fly-ash deposit at the "Nikola Tesla -A" thermoelectric power plant (TENT-A) in Obrenovac (N 44º30', E 19º58'), which is located on the right bank of the Sava River 40 km upstream from Belgrade (Serbia) (Supplementary Fig. S1).The climate is semi-continental with local mean annual rainfall and temperature of 647 mm and 11ºC, respectively.As the largest producer of electricity in Serbia, southeastern Europe and the Balkans, the six blocks of TENT-A with a total power of 1820 MW annually produce about 8 billion kWh of electricity.Annually, TENT-A burns 12-14 Mt of lignite, which is supplied by the Kolubara-Tamnava surface pits.The chemical analysis of electrofilter ash at TENT-A showed that it consists of SiO 2 (54.21%),Al 2 O 3 (24.98%),Fe 2 O 3 (6.13%),CaO (5.89%), MgO (3.15%),K 2 O (1.12%), Na 2 O (0.29%), TiO 2 (0.69%), P 2 O 5 (0.07%) and SO 3 (0.96%) (data obtained from the Vinča Institute for Nuclear Sciences, Belgrade, Serbia).
To date, the landfill of TENT-A has disposed of more than 66×106 t of fly ash that occupies 400 ha of agricultural land of fluvisol type.Fly ash is hydraulically transported in a suspension with water at ratios of 1:10 or 1:20.The disposal of fly ash is carried out in three lagoons, one is active (L 2 ), and the other two are in a temporary mode for the technical consolidation of fly ash and drainage (L 1 and L 3 ) (Fig. S1C).To reduce the negative impact of fly ash on the environment, TENT-A has provided biological recultivation, i.e. the sowing of legumes (Medicago sativa L., Lotus corniculatus L., Vicia villosa Roth., Trifolium pratense L.) and grasses (Secale cereale L., Lolium multiflorum Lam., Festuca rubra L., Dactylis glomerata L.), as well as planting of trees (Robinia pseudoacacia L., Ailanthus altissima (Mill.)Swingle) and shrubs (Tamarix sp.).The established herbaceous and woody vegetation cover on fly-ash deposits over time promote natural succession and revegetation.In the flat part of L 1 , some spontaneous herbaceous plant species, shrubs and trees were noted: Calamagrostis epigejos L., Oenothera biennis L., Sorghum halepense (L.) Pers., Erigeron canadensis L. and Amorpha fruticosa L.

Sampling of fly ash
The field research into fly-ash deposits on L 1 was carried out during August 2016.The fly ash was collected at a depth of 0-30 cm on bare zones without vegetation cover represented control fly ash (C FA ), whereas the fly ash taken up in the root zone of R. pseudoacacia, A. altissima and A. fruticosa was marked as plant rhizospheric fly ash (RP FA, AA FA, and AF FA , respectively).Fly-ash samples were collected in the range of 5-10 m from the embankment of L 1 .Collected fly ash was packed into plastic bags and brought to the laboratory for analysis.After the removal of visible plant remains, samples were dried at room temperature (25ºC) and sifted through a sieve (0.5 mm mesh).For chemical, elemental and biochemical analyses, five composite samples of fly ash were used (n=5).

Chemical analysis of fly ash
Fly-ash pH was measured in water (pH H2O ) and 0.1 M KCl solution (pH KCl ) with a pH meter (PHT-026 Multifunction meter).Electrical conductivity (EC[dSm -1 ]  in the extract (fly ash:water=1:5) was measured with a conductometer (PHT-026 Multi-function meter).Organic carbon (C) was measured by the method of Tyurin [27], and the total nitrogen content (N) was determined by Kjeldahl digestion [28].The C/N ratio was calculated.Available forms of phosphorus (P 2 O 5 ) and potassium (K 2 O) were analyzed using the standard AL-method [29].

Elemental analysis of fly ash
Total concentrations of chemical elements (Cu, Fe, Mn, Ni, Zn) in fly ash were determined according to the modified method 3051A (EPA SW-846 test methods) [30] as follows: 2-3 g of fly-ash sample were oven-dried for 1 h at 105ºC.The sample was dissolved in a mixture of 25 mL of HNO 3 and HClO 4 for 12 h at 40ºC.Concentrations of available chemical elements in the fly ash were determined according to Zemberyova et al. [31].Extraction was performed with 0.05 mol/ dm 3 EDTA at pH 7. The sample of dried fly ash (2-3 g) was added in 25 mL of 0.05 M EDTA and mixed on a magnetic stirrer for 1 h at room temperature (20±4ºC).Flam atomic absorption spectrophotometer (FAAS, "Perkin Elmer 3300") was used for analyzing the concentrations of chemical elements.Standard solutions were used for the preparation of calibrated diagrams.The range of concentrations of test elements of the standard solutions was 0.5-2.0mg/dm 3 for Cu, Zn and Ni, and 1.0-5.0mg/dm 3 for Mn and Fe.The measured values of the element contents in fly ash are expressed in µg/g dry matter.

Determination of phenolic acids and flavonoids by HPLC
Phenolic acids and flavonoids were extracted by dissolving 10 g of dry fly ash in 30 mL of pure methanol (99.8%) in an ultrasonic bath for 15 min and then left to dissolve another 24 h.Subsequently, samples were centrifugated 20 min at 10000×g and supernatants were filtered through 0.2-μm cellulose filters (Agilent Technologies, Santa Clara, CA, USA) and stored at 4°C until use.
The methanolic extracts were analyzed by a HPLC system (Shimadzu, Kyoto, Japan) which consisted of a degasser DGU-20A3, analytical pumps LC-20AT, 7125 injectors and SPD-M20A diode array detector and CBM-20A system controller.Separation was achieved on a Luna C18 column at 30ºC, 250×4.6 mm I.D., 5 µm (Phenomenex, Torrance, CA, USA) with a flow rate of 1.0 mL/min.The injection volume was 20 µL.The chromatographic data were processed using LC Solution computer software (Shimadzu).Gradient elution was used (5% B 0-5 min, gradient 5-60% B during 5-30 min, 60% B held for 5 min, then ramped from 60% to 90% B for 2-3 min and equilibrated for a further 5 min; mobile phases: A, water acidified with formic acid, pH=3; B, acetonitrile).The identity of compounds was determined by comparing the retention times and absorption maxima of known peaks with pure standards (Sigma) at 290 and 245 nm.Two phenolic acids were used as phenolic acid standards: ferulic and 3,5-dihydroxybenzoic acid (3,5-DHBA); for identification of flavonoids, two different standards were used: rutin and quercetin.The concentrations of phenolic acids and flavonoids are expressed in μg/g of extracts.

Growth inhibition test
The allelopathic activity of selected plant species grown on fly ash was assessed by a modified "soil sandwich method" [32].In the experiment, 5 mL of agar (0.5%) cooled to 42ºC were added into a multi-dish (6 dishes) plate containing 3 g of dried fly ash.After solidification, 3.2 mL of agar (0.5%) was added on a fly ash-agar layer.After 1 h, 5 seeds of T. pratense L. were added to the gelled agar culture medium in one dish (30 seeds per multi-dish plate).Control plates contained only agar medium (without fly ash).The multi-dishes were incubated at 25ºC in the dark.After 7 days, the lengths of the radicle and hypocotyl were measured and the percentage of growth inhibition was calculated (compared to control).The bioassays were done in 5 replicates (30 seeds per replicate, n=150).

Statistical analyses
Statistical analyses included determination of the mean (M) and standard deviation (SD) for each of the analyzed parameters.Differences between groups in terms of chemical properties of fly ash, total and available concentrations of chemical elements, content of phenolic acids and flavonoids in fly ash as well as the inhibition of radicle and hypocotyl growth of indicator species, were determined by multivariate variance analysis (MANOVA) and Scheffé's post-hoc test.Pearson's correlation coefficients (r) between the allelopathic activity of selected plant species growing on fly ash and the chemical characteristics, total and available concentrations of elements and content of phenolics in fly ash were determined.Statistical analysis was performed using the package Statistica 10.0.

Chemical properties of fly ash
The chemical properties of the control (C FA ) and rhizospheric fly ash (RP FA, AA FA, and AF FA ) are shown in Table 1.The results showed that the pH (H 2 O) and pH (KCl) in RP FA had higher values than the C FA (p<0.001),AA FA (p<0.001) and AF FA (p<0.001).EC values were higher in AF FA compared to C FA (p<0.001),RP FA (p<0.001) and AA FA (p<0.001).In comparison to C FA the results showed statistically significant higher values of C, N and C/N in RP FA (p<0.01),AA FA (p<0.01) and AF FA (p<0.01),except for the values of N in the RP FA and the C/N values in the AA FA .The highest values of C and N were detected in AA FA .Also, statistically significantly higher values (p<0.001) for both P 2 O 5 and K 2 O were detected in all rhizospheric fly ash in comparison to control fly ash.The highest values of P 2 O 5 were detected in AA FA , while the highest values of K 2 O were detected in the RP FA .

Total and available elements in fly ash
Total and available concentrations of Cu, Fe, Mn, Ni and Zn in C FA , RP FA , AA FA and AF FA are shown in Table 1.The results showed statistically significantly lower concentrations of Cu, Fe and Ni in the C FA compared to all plant rhizospheric fly ash (p<0.001),except for Fe and Ni in the AA FA .However, statistically significantly higher concentrations of both Mn and Zn were found in the C FA (p<0.001) compared to all other plant rhizospheric fly ash, except for Zn in the AF FA .In the control fly ash and all other rhizospheric fly ash, total metal concentrations decreased in the following order: Fe>Mn>Ni>Cu>Zn.
The available Cu, Fe and Ni concentrations in the C FA were lower in comparison to all plant rhizospheric fly ash (p<0.001),except for Fe in the AA FA and Ni in the RP FA .However, the available concentrations of Mn and Zn in the C FA were higher versus all the plant rhizospheric fly ash (p<0.001).The content of the available elements in the C FA , RP FA and AF FA showed the same decreasing order: Fe>Mn>Cu>Ni>Zn, whereas the content of available elements in the AA FA decreased in a slightly different order: Fe>Mn>Ni>Cu>Zn.Furthermore, the percentage of available concentrations of Fe, Mn and Ni in fly ash relative to the total content of the elements was the highest in the AA FA (0.42%, 4.07%, 4.39%, respectively).The percentage of available Zn in fly ash relative to the total metal content was the highest in the C FA (11.31%), whereas the highest percentage of available Cu concentrations was detected in AF FA (19.6%).

Phenolic content in fly ash
The content of phenolic acids and flavonoids in C FA , RP FA , AA FA , and AF FA is presented in Table 2.The content of 3,5-DHBA in C FA was higher, whereas the content of ferulic acids was lower compared to the RP FA and AA FA (p<0.001).Results in this study showed the highest content of these phenolic acids in the AA FA .
The rutin content in C FA was lower in relation to the RP FA and AF FA (p<0.001), while the quercetin content in C FA was lower in comparison to the AA FA (p<0.001).
The highest content of rutin was noted in the AF FA , whereas the highest content of quercetin was found in the AA FA .The phenolic acid and flavonoid contents in the C FA showed the following decreasing order: quercetin>3,5-DHBA>rutin>ferulic acid.In the RP FA and AA FA , the order of phenolic content was as follows: quercetin>ferulic acid>rutin>3,5-DHBA.However, the content of phenolic acids and flavonoids in the AF FA showed a different decreasing order: quercetin>3,5-DHBA>ferulic acid>rutin.In this study, the relationship between chemical properties and the phenolic content in fly ash is presented in Table 3. Positive correlations were noted between EC, the P 2 O 5 content, the total and available concentrations of Cu, total concentrations of Fe and Ni and 3,5-DHBA, ferulic acid and rutin (p<0.05,p<0.01, p<0.001), and between the C content and ferulic acid (p<0.001), the N content, the available concentrations of Fe and Ni and quercetin (p<0.01).However, negative correlations were found between the C and K 2 O contents, the total concentrations of Fe, the available concentrations of Ni and quercetin in fly ash (0<0.05,p<0.01, p<0.001).

Allelopathic effects of plants growing on fly ash deposits
Higher inhibition of radicle growth of red clover was noted in AF FA in comparison to the C FA and RP FA (p<0.01) (Fig. 1).Higher inhibition of red clover radicle growth was found in the AA FA versus RP FA (p<0.001).The inhibition of radicle elongation decreased in the following order: AF FA >AA FA >C FA >RP FA .Hypocotyl growth inhibition in red clover was significantly lower in the rhizospheric fly ash of all plant species compared to the C FA (p<0.01,p<0.001), as well as in RP FA compared to AA FA (p<0.01) (Fig. 2).The inhibition of red clover hypocotyl growth decreased in the following order: C FA >AA FA >AF FA >RP FA .

DISCUSSION
The phytotoxic activity of allelochemicals in soils is a function of the interaction of the allelochemicals from plants and soil characteristics [17].Allelochemicals primarily affect the soil pH, aggregation and aeration, content of N and C, plant-water relations, nutrient ion uptake, degree of decomposition, composition and activity of soil microbiocenosis [5].Phenolic compounds  have an effect on the accumulation and availability of soil nutrients, which in turn have a great influence on plant growth [17].However, phenolics can also play an important role in inhibiting the nitrification that affects a plant's nutrient status [1].Recent studies are more focused on the environmental impact of allelochemicals on ecosystems than on plant-plant interactions [11], indicating a great influence of allelochemicals on the turnover of inorganic and organic compounds in soil [33].
In this study, the pH values of fly ash ranged from 6.32-8.1, which is similar to results obtained by other authors (8.85 [34], 7.95 [25]).According to Whitehead et al. [35], soil pH can affect the concentration of phenolic compounds in the soil solution.Our research showed that as the pH of fly ash increases, the inhibition of radicle and hypocotyl growth of red clover is less pronounced.A high pH value (8.1) and low nitrogen content (0.13%) in the rhizospheric fly ash of R. pseudoacacia (RP FA ) was noted, although this is a nitrogen-fixing species.The cause of this phenomenon may be the detrimental effect of high pH on the microorganisms involved in nitrogen fixation [36].Generally, microorganisms can reduce the allelochemical activity of phenolics in the rhizosphere and reduce the inhibition of plant growth due to rapid mineralization of phenolic compounds [15].The limitation of phytotoxic activity of active substances released from the donor plants could occur through sorption to organic matter which is greatest under neutral to slightly alkaline conditions [21].However, high inhibition of red clover radicle and hypocotyl growth could be attributed to the lower pH values of the AA FA and AF FA (6.34 and 6.32, respectively), which can modify the nutrient status and element availability in fly ash [21].
In the present study, the C content in plant rhizospheric fly ash (RP FA , AA FA and AF FA ) was greater than in the control fly ash (C FA ) (2.41, 3.12 and 3.07 times, respectively), and this might be due to its origin from humus that contains humic acids, peptides and carbohydrates as compared to control fly ash where the C is derived from coal [37].Our findings show that phenolics in the AA FA and AF FA had a more negative impact on red clover radicle growth with an increasing C content in fly ash compared to phenolics in the RP FA and C FA .The C pool in soils depends on its input through leaf and root litter and root exudates and decomposition of organic matter by microorganisms [38].The high inhibitory effect of A. altissima and A. fruticosa on red clover radicle growth could be related to the high content of phenolics in fly ash which, being C-rich compounds, are released from plants to fly ash and contribute to the C-stock in fly ash.The positive correlation between ferulic acids and carbon content in fly ash (r=+0.585)points to their high content in the AF FA (4.11 µg/g, 4.43%) and AA FA (0.7 µg/g, 4.5%).Hence, it was assumed that some phenolics from these plants were less susceptible to microbial decomposition and had a high allelopathic effect.
The N content in the AA FA and AF FA was higher than in C FA (3.27 and 1.90 times, respectively) and in RP FA (2.77 and 1.61 times, respectively).The high N content in rhizospheric fly ash reveals the ability of these plant species to increase the amount of nitrate in the soil where the net nitrification rate in N-deficient substrates, such as fly ash, depends on the quality of litter [39].Furthermore, in N-limited soils, the high content of phenolics inhibits N mineralization, leads to the release of dissolved organic N from leaf litter and reduces overall ecosystem N loss [40].In our study, the highest content of quercetin (60 μg/g) and N (0.36%) in the AA FA and the positive correlation between quercetin content and N in fly ash (r =+0.945) suggest that the high polyphenol production from this plant species may present an advantage to increased plant uptake of organic N [40].
The content of available P was significantly higher in plant rhizospheric fly ash (RP FA , AA FA and AF FA ) than in the C FA (1.21, 1.51 and 2.69 times, respectively).Herr et al. [41] noted that the rhizospheric soil of Solidago gigantea had a lower pH value and a higher content of available P than surrounding soils without this species, which is similar to our observations in A. altissima and A. fruticosa.Our results showed significant inhibition of red clover radicle growth in response to the high available content of P in fly ash.According to Kafkafi et al. [42], phenolic compounds can compete with P for sorption sites on mineral surfaces and may form complexes with Al and Fe.The high content of phenolic allelochemicals (3.5-DHBA, ferulic acids and rutin) that are released from plants can increase P availability (r=+0.925,r=+0.990,r=+0.955,respectively), desorbing previously bound P, which might be related to our results obtained for A. fruticosa.The low content of P in the RP FA can be related to the inhibition of N fixation [43] as well as with the low content of phenolic acids, which can reduce solubility and availability of P [5].
The concentration of Cu, Mn, Ni, and Zn in fly ash can be 30 times higher than in coal, making fly ash the main threat for the surrounding environment and human health [44].In our study, the total concentrations of Ni in all rhizospheric fly ash samples were toxic (12-34 μg/g, [45]), the concentrations of Cu were within the normal range (13-24 μg/g, [45]) whereas the concentrations of Mn and Zn were deficient (270-525 μg/g and 45-100 μg/g, respectively [45]).The availability of Cu, Fe, Mn, Ni and Zn in control and plant rhizospheric fly ash samples was relatively low.In addition, the concentrations of total and available Mn and Zn in the rhizospheric fly ash of all woody species were lower than in control fly ash.Higher Cu availability in AF FA in relation to the C FA can be due to low pH values (6.3), as in slightly acidic conditions Cu is more available [46].However, red clover radicle growth inhibition is positively correlated with total concentrations of Cu and the available content of Mn and Ni in fly ash.The content of heavy metals in soil can be crucial for the persistence and activity of allelochemicals [47] and may affect the production and release of secondary metabolites from the plant roots, which in turn increase the availability of nutrients or form chelates with toxic metal in the soil [48].
Phenolic compounds can increase the availability of Cu, Fe and Mn by forming organic complexes that enhance the uptake of these elements by plants [5,21].The high concentrations of 3.5-DHBA, ferulic acid and rutin in fly ash were related to the high total content of Cu (r=+0.706,r=+0.863,r=+0.798,respectively), Fe (r=+0.717,r=+0.705,r=+0.777,respectively) and Ni (r=+0.8555,r=+0.818,r=+0.888,respectively), which coincides mostly with the results obtained in A. fruticosa.The available content of Cu increased and the available content of Fe decreased with the higher contents of 3.5-DHBA (r=+0.791,r=-0.768,respectively), ferulic acid (r=+0.872,r=-0.616,respectively) and rutin (r=+0.878,r=-0.733,respectively) in fly ash mostly found in A. fruticosa, whereas the highest content of quercetin in fly ash coincided with the highest available content of Ni (r=+0.755) in the AA FA .The decrease of hypocotyl growth of red clover was lower with increasing total concentrations of Cu, Fe, Mn and Ni and with decreasing available concentrations of Mn and Zn in fly ash.Since there are no significant correlations between hypocotyl growth inhibition of red clover and phenolic compounds in fly ash, it can be assumed that these phenolics do not express an allelopathic effect on red clover hypocotyl growth.
The results of this study showed that the contents of 3,5-DHBA, ferulic acid and rutin were greatest in the AF FA where they expressed the strongest inhibitory potential on red clover radicle growth.The cause of this phenomenon may be the increased allelopathic activity of several different phenolic compounds in the combination [49].Thus, the additive and synergistic effects of several different phenolic compounds become significant and decisive in allelopathic inhibition of various ecophysiological processes in the acceptor plant [5,8].Csizar [50] showed that A. fruticosa had stronger allelopathic potential than A. altissima and R. pseudoacacia, which is in line with our results.Generally, different allelopathic compounds were found in R. pseudoacacia and A. altissima [51,52], but the data on the allelopathic activity of these woody species grown on fly-ash deposits are still lacking.
In this study, the presence of phenolic compounds in all plant rhizospheric fly ash samples indicates that R. pseudoacacia, A. altissima and A. fruticosa can affect the beginning of pedogenetic processes of a sterile substrate, such as fly ash.Phenolic compounds, as the most prevalent allelochemicals in plants, enter the soil by leaching from the surface of the plant body, decomposition of litter and active secretion from roots [6,53].Free phenolics are very important in the interactions between donor and acceptor plants because they are first available to the acceptor plant when they reach the litter and the soil [23].Therefore, the detection of allelochemicals in the rhizosphere of the donor species is very important for allelopathy [54].Further studies will determine the seasonal dynamics of phenolic compounds in rhizospheric fly ash and leaf litter of investigated plant species, the contents of these allelochemicals in roots and aboveground plant parts, as well as the effects of these compounds on germination and the seedling growth of the other herbaceous plant species that grow on fly-ash deposits.

CONCLUSIONS
The allelopathic activity of phenolic compounds in rhizospheric fly ash is strongly influenced by pH, content of C and available P, as well as the total content of Cu and the available contents of Mn and Ni in fly ash.In addition, the unfavorable conditions that prevail on fly-ash deposits promote the high production of phenolic acids and flavonoids in woody plant species such as A. fruticosa, A. altissima and R. pseudoacacia.These phenolic compounds have the properties of allelochemicals in fly ash and can lead to the inhibition of growth of T. pratense as the understory herbaceous species present in the fly-ash plant community.These woody species that colonized the fly-ash deposits may act as promotors of pedogenetic processes on fly ash, altering the ecosystem processes in anthropogenically degraded sites.

Table 4 .
Pearson's correlation coefficients (r) between red clover radicle and hypocotyl growth inhibition and chemical parameters, total and available concentration of elements and phenolic contents in fly ash.