Chemometric Evaluation of 16 Priority PAHs in Soil and Roots of Syringa vulgaris and Ficus carica from the Bor Region (Serbia): An Insight into the Natural Plant Potential for Soil Phytomonitoring and Phytoremediation

Abstract

The soil phytomonitoring and phytostabilization potential of Syringa vulgaris and Ficus carica was evaluated regarding 16 priority polycyclic aromatic hydrocarbons (PAHs) using a chemometric approach and the calculation of bioconcentration factors (BCFs) for each individual PAH in plants’ roots from each selected location in the Bor region. PAHs in roots and the corresponding soils were analyzed using the QuEChERS (Quick, Effective, Cheap, Easy, Rugged, Safe) method with some new modifications, gas chromatography/mass spectrometry, Pearson’s correlation study, hierarchical cluster analysis, and BCFs. Several central conclusions are as follows: Each plant species developed its own specific capability for PAH management, and root concentrations ranged from not detected (for several compounds) to 5592 μg/kg (for fluorene in S. vulgaris). In some cases, especially regarding benzo(a)pyrene and chrysene, both plants had a similar tactic—the total avoidance of assimilation (probably due to their high toxicity). Both plants retained significant quantities of different PAHs in their roots (many calculated BCFs were higher than 1 or were even extremely high), which recommends them for PAH phytostabilization (especially fluorene, benzo(b)fluoranthene, and benzo(k)fluoranthene). In soil monitoring, neither of the plants are helpful because their roots do not reflect the actual situation found in soil. Finally, the analysis of the corresponding soils provided useful monitoring information.

1. Introduction

Polycyclic aromatic hydrocarbons (PAHs) are hydrophobic, semi-volatile organic pollutants with two or more benzene rings combined in different manners [1,2,3]. According to their molecular structure and molecular weight, PAHs can be simply divided into light-molecular-weight (LMW) or high-molecular-weight (HMW, containing five or more aromatic benzene rings) PAHs [4].

Polycyclic aromatic hydrocarbons can be formed during natural or anthropogenic processes; the latter is recognized as the main source of PAHs [3]. Volcanic activities, wood fires, diagenesis, and some biogenic processes (resulting from microbial activities) can be counted as natural sources of PAHs, whereas anthropogenic sources include a range of industrial processes, traffic, incineration of various organic materials (tobacco, different organic waste materials, grilled meat), all kinds of pyrolytic processes, etc. [5].

This class of dangerous micro-pollutants accounts for more than one hundred compounds, but the following six teen of these are the most examined in different environmental studies across the world [1,6,7]: acenaphthene (Ace), acenaphthylene (Acy), anthracene (Ant), benzo(a)anthracene (BaA), benzo(a)pyrene (BaP), benzo(b)fluoranthene (BbF), benzo(g,h,i)perylene (BgP), benzo(k)fluoranthene (BkF), chrysene (CHR), dibenzo(a,h)anthracene (DhA), fluoranthene (Flt), fluorene (Flr), indeno(1,2,3-c,d)pyrene (IcP), naphthalene (Nap), phenanthrene (Phe), and pyrene (Pyr). Because of a great concern about their toxic effects and easy transmission over long distances [2,4,7,8,9], these compounds are included in the list of priority pollutants that was created by the United States Environmental Protection Agency (USEPA). Seven of them,BaA, BaP, BbF, BkF, CHR, DhA, and IcP, are USEPA-classified as probable carcinogens [10,11]. In addition, the USEPA formed the Risk-Based Screening Tables, which refer to all important pollutants together with these 16 PAHs. Practically, these Generic Tables contain limit values (LVs) related to the so-called Regional Screening Levels (RSLs) for concentrations of pollutants in tap water and in residential and industrial soil [12]. As for soil, RSLs denote PAH concentrations that do not require any additional procedures to be undertaken regarding soil protection; however, when PAH concentrations are identical to or surpass RSLs, further assessments (but not automatic remediation) are needed. USEPA RSLs are defined on the basis of the target cancer risk (TR) of 1 × 10⁻⁶ and the two values of the target hazard quotients (THQ): 1.0 and 0.1 [12]. This agency paid special attention to soil because soil represents a non-renewable (or very slowly renewable) natural resource [13], which acts as a main sink for emitted PAHs [14]. More precisely, lipophilic molecules of PAHs are naturally firmly attached to soil particles, i.e., to soil organic matter (OM), such as humins, humic acids, fulvic acids, etc., where they can persist for a very long period of time [15]. Similarly to other organic molecules, strong soil sorption of PAHs can be explained by their specific characteristics, such as low solubility in water (especially in the case of highly hydrophobic HMW PAHs,with their high values of the partition coefficient: n-octanol/water, Kow), low vapor pressure and low Henry’s law constants (which indicate the chemical’s potential to volatilize), and with generally high organic carbon partitioning coefficients (Koc – an indicator of a compound solubility in lipids, which increases with PAH molecular mass) [6]. The described characteristics make PAH molecules markedly predisposed to the processes of sequestration into soil micro-pores that may lead to a PAH moving into deeper layers of soil, which is known as “aging”; accordingly, PAHs are named “recalcitrant” substances [15]. Importantly, this relocation from the shallow, easily accessible layers of soil to the deeper and less reachable ones causes, in the end, a lower availability of PAHs for soil organisms, which is not desirable for processes aimed at PAHs’ bioremediation, which is an environmentally friendly method for soil remediation. Namely, for PAHs’ decomposition or their total elimination, bioremediation utilizes the abilities of different soil organisms such as various microbes (having the most effective potential for PAH degradation) [4] and then a range of plant species [6,16,17].

In general, PAHs are toxic for all organisms living in soils, causing very serious damage in them initiated by so-called oxidation stress, and in the case of plants, PAHs may affect their metabolism in many ways—at the cellular and at the morphological levels. For instance, PAHs may cause various effects in the photosynthetic system in the metabolism of carbon, numerous amino acids, and nitrogen; to some extent, they may also influence secondary plant metabolites. Finally, on a macro-plane, in sensitive plant species, the produced disorders can decrease the root and shoot growth and commonly yield; flowering may be late; in trichomes, some malformation can be observed; in leaves, chlorosis may occur; and also, white leaf spots may expand into necrotic lesions [6,16]. However, during the long time of evolutionary development, in adaptive efforts to overcome problems with aggressive soil environments, some plants successfully evolved different and very specific mechanisms of resistance to PAHs, and they are known as tolerant species. Practically, except PAH destruction or immobilization in the soil or on root surfaces (usually using different root exudates such as organic acids, enzymes, etc., and often in cooperation with the associated soil microbes), plants may also uptake and accumulate significant PAH quantities into the root cells (inside their specific organelles—vacuoles). In addition, PAHs may be translocated into the aerial parts through the transpiration pathways, depending on the nature of plant species and the pollutant features (its molecular structure, lipo-solubility, and Henry’s law constant) [11,18]. However, the translocation of PAHs is primarily reserved for PAHs with the simplest structures, whereas robust constitutions of complex PAH molecules are less susceptible to these processes [6,19]. Further, inside the plant cells, PAHs may undergo complicated processes of decomposition and transformation into the less poisonous metabolites (the processes of detoxification) which are very comparable with the processes in mammals (the so-called “green liver” concept) [6,16].

Based on these facts, different phytoremediation methods, as part of a bioremediation approach, are developed and, today, they represent relatively new and effective technologies, especially for large contaminated soil surfaces. In contrast to other classical remediation methods (chemical or physical), phytoremediation has no negative or destructive effects regarding the soil structure and quality, and additionally it is even able to improve many soil characteristics [7,9]. Phytoremediation consists of several different procedures for PAH removal or their partial degradation [17,20,21].

For instance, a phytoremediation method, known as phytostabilization, refers to the mentioned excretion of root exudates that may cause PAH decomposition but also their immobilization, producing PAH-reactive radicals in soil. These radicals can be precipitated, polymerized, or covalently attached to the soil OM, which practically stabilizes PAHs and restricts their spreading in the soil matrix [7]. Most importantly, the excreted compounds can also regulate the geochemical environment in the plant rhizosphere, in a manner that offers maximally optimal conditions for the expansion of colonies of useful soil microbiota that promote plant growth; in that case, plants support a practical cooperation with the associated soil microbes which may result in considerable PAH degradation in the rooting zone. The described synergistic effects between plants and rhizosphere microbes helped in the creation of the most successful method for PAH phytoremediation, namely, rhizosphere biodegradation [6,16].

Phytodegradation/phytotransformation is a method which is related to the effects of PAH biotransformation performed inside the plant cells (the “green liver” concept), where plants are capable only for a partial modifications of PAHs, or their accumulation inside cells (most often in vacuoles). For instance, the classical phytoextraction/phytoaccumulation method, which refers to the extraction of soil pollutants by plant roots, and latter translocation/accumulation inside the above-ground plant tissues (that further can be harvested and removed from the contaminated site) [7,11], cannot be recommended as one of useful phytoremediation methods in the case of pollutants such as PAHs. Namely, it was found that the accumulation rates of PAHs in the aerial parts, expressed through bioconcentration factors (BCFs, the ratio between the concentration of a PAH compound in a plant part and its concentration in the soil), are generally very low and range from 0.001 to 0.4, and only sometimes to 2 [6,22]. At the same time, some recent studies showed that, in the case of roots, BCFs for individual PAHs may reach enormously high values. For example, the examined roots of wild blackberry from the Bor region in Eastern Serbia showed an extremely high potential for the extraction and accumulation of many PAHs [23,24]. Namely, blackberry roots were able to extract the complete content of some PAHs from the corresponding soils which resulted in the impossibility of calculating their (enormously high) BCF values. Nevertheless, the utilization of roots in a classical phytoextraction/phytoaccumulation procedure is not so convenient due to the almost unfeasible conditions for their harvesting. At the same time, a significant accumulation of PAHs in plant roots may help in phytostabilization, i.e., in the restriction of PAH dispersion in soil [23,24].

Many plants (but most oftenlegumes and grasses) have beeninvestigated in terms of their application in some of phytoremediation methods [1,9,16,20,21,22,23,24,25,26,27,28], as well as in various biomonitoring procedures regarding PAHs [23,24,29,30]. In this work, two plant species from the Bor region (Eastern Serbia), common lilac (Syringa vulgaris L., SV) and common fig (Ficus carica L., FC) cultivar crna Petrovača, were selected as testing plants for the examination of their natural phytomonitoring and phytostabilization capacities (concerning 16 priority PAHs). More precisely, given that the utilization of aerial parts (in a classically designed phytoextraction method) cannot represent an effective solution for structurally massive PAH molecules, the main focus of this study was the extraction/bioaccumulation capacity of SV and FC roots, as well as the possibility of their application in soil phytomonitoring and phytostabilization as two environmentally friendly methods for soil control, protection, and conservation. According to the available literature data, neither selected plants has been not investigated earlier with this regard and the results obtained in this study may be of significant assistance for all researchers interested in the mentioned environmental issues. Another useful contribution of this work, especially for all environmentally oriented analysts, represents the introduction of several modifications in a relatively new method for PAH isolation, known as the QuEChERS (Quick, Effective, Cheap, Easy, Rugged, Safe) method. These modifications are described in the experimental part and they were applied to plant material in terms of obtaining maximal precision in its chemical analysis. In terms of the identification of eventually increased PAH concentrations in soils of the investigated Bor region, the results obtained for the sampled soils of both plants were compared with the earlier published results in Alagić et al. [23,24] for soils (of wild blackberry) from the related locations in the same region. Finally, soil results were compared with the related USEPA RSLs for each investigated compound [12], with the aim of discovering the exceeded LVs.

2. Materials and Methods

2.1. The Sampling Area

Bor is a town in Eastern Serbia, surrounded by mountains rich in diverse vegetation. Its geographic coordinates are 44º25ʹN and 22º06ʹE. The dominant winds in the region are northwest (NW), west–northwest (WNW), and west (W). The main parts of the regional economy are copper mining and processing. The possible major sources of PAHs in this region are the copper smelting plant and the neighboring city heating plant, both placed in an industrial complex at the northeastern periphery of the city of Bor. In this study, the investigated locations were selected at various distances from the industrial complex, practically from the new “flash” smelter and the heating plant because both plants spread their waste gases jointly. Over the years, including the year of sampling for this study, the heating plant employed different kinds of coal. In rural settlements, the domestic heating is present, mainly based on wood, coal, and sometimes—fuel combustion. Also, in rural fields, controlled stubble burning is not so unusual. Finally, during the long, hot summers, an inappropriate disposal of various waste materials at the city landfill represents a considerable source of eliminated spontaneous, massive ignitions, fuming a significant part of the investigated territory.

The contents of 16 PAHs were determined in soils and roots of SV and FC that were grown in both zones (Z) of interest: rural (RZ), and urban/industrial (UIZ), at positions differently distant from the industrial/metallurgical complex (Figure 1).

The urban/industrial zone enclosed several parts of the town of Bor, and they are listed here, together with their assigned labels and distances from the complex: (1) Flotacijsko jalovište (FJ), a flotation tailings pond, out of use—0.7 km; (2) Bolničko naselje (BN), a central part in Bor, near the city hospital—1.3 km; (3) Slatinsko naselje (SN), very close to this place, in the local pathway to the rural settlement Oštrelj (O), the city landfill is placed)—3.2 km;(4) Naselje Sunce (NS), a residential part at the periphery of Bor—3.6 km. The rural zone included 4 rural sites and 1 touristic site: (5) Oštrelj (O)—4.5 km; (6) Slatina (S)—6.5 km; (7) Borsko jezero (BJ), Bor Lake, a touristic site—7 km; (8) Krivelj (K)—8 km; (9) Gornjane (G)—19 km (Figure 1).

2.2. Sample Collection and Pretreatment

The sampling of roots and the corresponding soils of the selected plant species, SV and FC, was conducted in individual backyards in UIZ and RZ, where the shrubs were planted, except in the case of FJ, where the plants of both species were found to be spontaneously growing. Basic information on the chosen plant species is given in the Supplementary Materials, in List S1.

Root and soil samples of SV and FC (about 1 kg of each material) were collected at the chosen sites during September 2020, from the related rooting zones (depth: 0–30 cm), using means of stainless steel. At each location, the samples of soils and roots of SV and FC were taken from 3 (site FJ) to 5 individual plants. This initially sampled root and soil material was later joined to obtain a representative sample for roots or soils of each investigated plant species, for each selected location. The collected samples were placed into bags made of aluminum foil and delivered to the laboratory for further treatment. First, root samples were divided into fine roots (which were discarded) and coarse roots, which were subjected to careful cleaning. When soil particles were removed from the coarse roots, they were additionally cleaned with fine sandpaper; finally, they were washed with tap and distilled water. All soil and root samples were air-dried to a constant weight in an isolated room and kept safe from any unwanted contamination. After drying, plant samples were homogenized in a vibratory disc mill, while soil samples were riddled through a stainless steel sieve (2mm). Before the subsequent steps of extraction, cleaning up, and gas-chromatography/mass-spectrometry (GC/MS) analysis, all examined samples were placed in sealed/closed bags and kept at 4 °C.

2.3. Extraction of PAHs from Soil and Plant Samples and Chemical Analyses

2.3.1. Used Reagents and Measured Soil Parameters

The list of used reagents, together with the procedures for the preparation of PAH standard solutions, is given in the Supplementary Materials, in List S2. All measured soil parameters (pH, OM, and electrical conductivity, EC), are given in

Table 1. Measured soil parameters.

	Soil Parameters for SV			Soil Parameters for FC
Location	pH	EC (μS/cm)	OM (%)	pH	EC (μS/cm)	OM (%)
FJ	7.02	566	7.04	7.12	824	8.00
BN	7.37	850	8.27	7.53	588	7.07
SN	7.45	386	7.94	7.27	504	8.30
NS	7.45	549	8.28	7.39	196	9.29
O	7.08	351	7.00	7.77	362	5.98
S	6.46	267	8.73	7.45	982	14.28
BJ	7.37	508	9.45	7.06	284	10.56
K	7.39	518	5.78	7.67	378	7.57
G	6.55	240	5.42	7.05	392	6.87

.

The pH values and EC (μS/cm) of soil samples (in solid to distilled water ratio—1:2.5 w/v) were measured using a CyberScan pH 510 Eutech (NL) pH meter and 4510 Jenway (UK) conductometer. The content of soil OM (%) was determined by the loss-on-ignition method at 550 °C [31].

2.3.2. Extraction and Instrumental Analysis of PAHs with QA/QC

The extraction of PAHs from the collected soil and root samples was conducted using the QuEChERS technique—a relatively new and very precise method in PAH analysis, which has been confirmed in many environmental studies [14,32] including studies on wild blackberry from the municipality of Bor [23,24]. In the present study, some modifications of the latest version given in Alagić et al. [23,24] are related to different quantities of used reagents, and most importantly, instead of primary–secondary amine (as a sorbent in the second, purification step), diatomaceous earth was used, as a low cost sorbent, with a wide surface area, which positively affects the precision as well as the costs of the analysis; basically, this represents the very first utilization of diatomaceous earth in the refining of plant material during PAH analysis. Practically, in this work, each sample of soil and plant material (10 g of each) was weighed into a 50 mL QuEChERS tube and 300 μL of surrogate standard was added. After that, 30 mL of water/acetonitrile mixture (1:2, v/v) was also added. The tubes with samples were shaken for 1 min and then extracted in an ultrasonic bath for 30 min. Then, 2 g of NaCl and 8 g of MgSO₄ were also added into each tube and the obtained mixtures were shaken for about 1 min and centrifuged at 4000 rpm for 10 min. In the subsequent cleaning step, for soil samples, an aliquot of the supernatant (1.5 mL) was transferred into 2 mL tubes containing 50 mg diatomaceous earth and 150 mg MgSO₄; for plant samples, this step was performed using 100 mg diatomaceous earth and 500 mg MgSO₄. After 5 min of shaking, the tubes were centrifuged at 8000 rpm for 10 min. The supernatants of 1 mL volume were transferred to vials.

In order to perform the method validation, soil and plant samples from an unpolluted area were spiked using a PAH standard solution mixture. The spiking concentrations of the 16 priority PAHs were 5.0, 0.5, and 0.05 µg/g for both soil and plant samples. Spiked samples were prepared in the same way, described in the previous paragraph. Method validation parameters are presented in

Table 2. Retention times, corresponding ions, and method validation parameters for analyzed PAHs.

Compound	Rt (min)	Quantifier Ion	LOD (µg/kg)	LOQ (µg/kg)	Calibration Curve Equation	Correlation Coefficient	Recoverys (%)	Recoveryp (%)	RSDs (%)	RSDp (%)
Nap	11.792	128.0	0.78	2.60	y = 0.0343x	0.99	113	102	10.8	5.6
Acy	17.721	152.0	0.29	0.96	y = 0.0056x	0.99	100	99	6.37	4.17
Ace	18.460	153.0	0.19	0.65	y = 0.0627x	0.99	94	110	2.74	5.37
Flr	20.380	166.0	0.48	1.60	y = 0.004x	0.99	95	88	7.38	6.74
Phe	24.000	178.0	1.50	5.00	y = 0.298x	0.99	96	95	9.11	7.44
Ant	24.177	178.0	0.39	1.30	y = 0.109x	0.98	100	111	4.79	9.54
Flt	28.575	202.0	0.51	1.70	y = 0.069x	0.98	96	93	6.34	7.45
Pyr	29.383	202.0	0.60	2.00	y = 0.104x	0.99	96	89	4.76	6.54
CHR	34.072	228.0	1.02	3.40	y = 0.053x	0.98	116	106	4.97	5.42
BaA	34.238	228.0	0.50	1.67	y = 0.029x	0.99	112	95	7.98	8.42
BbF+BkF	38.103	253.0	0.05	0.18	y = 1.495x	0.99	91	98	3.87	2.97
BaP	39.039	253.0	0.60	2.00	y = 0.0225x	0.99	92	98	6.54	5.56
IcP	42.760	277.0	0.63	2.10	y = 0.076x	0.99	99	84	4.79	5.45
DhA	42.878	279.0	1.17	3.90	y = 0.0037x	0.99	90	107	7.86	5.87
BgP	43.637	276.0	0.90	2.00	y = 0.626x	0.99	96	100	8.93	6.87

Rt—retention time; LOD—limit of detection; LOQ—limit of quantification; Recovery_s—recovery values for soil samples; Recovery_p—recovery values for plant samples; RSD_s—relative standard deviation for soil samples; RSD_p—relative standard deviation for plant samples.

.

The calibration curves of 16 analyzed components were constructed by using a series of standard PAH solutions in the concentration range of 0.017–16.667 µg/mL; they were constructed on the basis of eight different PAH concentrations. A linear relationship between ratios of peak areas and PAH concentrations was noticeable. In all cases, the correlation coefficient was 0.99, except for anthracene, fluoranthene, and chrysene (0.98). A solution of the internal standards containing perylene d₁₂ and acenaphthene d₁₀ was added to every vial prior to GC/MS analysis to calculate each PAH concentration by comparison of the peak areas of the internal standard and each PAH compound. A surrogate standard solution (2-fluorobiphenyl, p-terphenyl-d₁₄, and 2,4,6-tribromophenol) was also added to each sample before extraction to monitor the efficiency of the extraction process. In method optimization experiments, soil samples were obtained from the unpolluted area and were used for blank and spiked sample preparation. Before spiking, the absence of PAHs in question was verified by analyzing the soil sample applying the standard EAHO [33] procedure.

To prevent the matrix effect on peak positions in the chromatogram, all spiked and blank samples were analyzed under the same conditions [34]. The gas-chromatography/mass-spectrometry analysis of soil and plant samples from the unpolluted area (blank samples), pretreated by the above-described QuEChERS procedure and followed by the GC/MS analysis, showed the absence of any potential interferences with the peaks of target analytes and surrogate and internal standards in the related chromatogram. LOD was calculated to be 3-times higher than the level of noise, while LOQ equaled10 times the noise level. Precision was evaluated using the repeatability (the so-calledintra-day) method. Repeatability was investigated on the same day, with the same instrument, by the same operator, using three spiked samples, and the results were expressed as repeatability relative standard deviation (RSDr) for both soil and plant samples. The highest RSD value for soil samples was obtained for naphthalene (10.8%), whereas, for plant samples, the value was 9.54% for anthracene. Accuracy was evaluated at 3 different spiking levels, for both soil and plant samples and expressed as recovery values (Table 2). The recoveries obtained in this study are in accordance with the acceptable recovery values for PAHs, i.e., 70–120% [34]. After the addition of 200 µL of internal standards (perylene d₁₂, and acenaphthene d₁₀, each in concentrations of 80 µg/mL) in all vials, GC/MS analysis was performed. All samples of soil and root for both plant species were prepared in triplicate and the obtained results were expressed in μg/kg dry weight (DW) ± SD (standard deviation).

The prepared extracts were analyzed on a 7890/7000B GC-MS-MS triple quadrupole device equipped with a Combi PAL auto sampler (Agilent Technologies, Santa Clara, CA, USA), and a HP-5MS capillary column (5% Phenyl Methyl Siloxane, dimensions: 30 m × 250 µm, film thickness 0.25 µm). First, the recording was performed in SCAN mode, with the aim of determining the retention times and characteristic ions that were used for the analysis of the related compounds.

After determining the retention times of elution of the analyzed compounds and the identification of ions that can be used for the compounds’ quantification, the recording was continued in SIM mode. This mode was applied because the yield of fragment ions in the multiple reaction monitoring (MRM) mode is low due to the stability of the conjugated system of the analyzed PAHs. The temperature regime was as follows: isothermal at 75 °C for the first 3 min, then a linear increase up to 300 °C (gradient 6 °C/min), and finally, isothermal at 300 °C for the last 10 min. The total analysis time was 50.5 min. The injected volume was 2.5 μL, in splitless mode. A carrier gas was helium with a flow rate of 1.0 mL/min. The ionization potential was −70 eV. Quantitative and qualitative analyses were performed based on the corresponding quantifier ions and the retention times given in Table 2 using the Mass Hunter QQQ Quantitative Analysis software B.05.02/Build 5.2.365.0 (Agilent Technologies, USA). The applied instrumental conditions were selected in terms of enabling the best resolution of all examined compounds; only in the case of BkF and BbF, the separation was not at the adequate level, and the results are given as a sum: BkF+BbF.

2.4. Data Processing

In order to study the existing relations between the investigated variables such as the detected contents of 16 PAHs in SV and FC roots and soils (also including the measured soil parameters pH, EC, and OM), the statistical method of Pearson’s correlation study (PCS) was performed. The same method was applied to the PAH contents and the distances of the selected locations from the industrial/metallurgical complex, in terms of identification of this complex as a possible main source of pollution. Hierarchical cluster analysis (HCA) was conducted for biomonitoring purposes, i.e., to illustrate the classification of the investigated polluted locations according to their similarities [35]. For all statistical methods, IBM SPSS Statistics 20 software version 20 (USA) was used.

The accumulation rates of PAHs in the roots of SV and FC were expressed through the so-called bioconcentration factors, BCFs, that are generally used to illustrate a concentration proportion of a plant part to the surrounding soil [22,23,24]. In this work, they were simply calculated as BCF = C_root/C_soil (C_root—the concentration of a compound in the root; C_soil—its concentration in the soil). A good accumulation capacity of roots of the investigated plant for an individual PAH compound is characterized by a BCF value higher than 1 which principally implies a good potential of the plant for its application in the phytoremediation method—phytostabilization [23,24].

3. Results and Discussion

The measured concentrations of individual PAH compounds in soil and plant samples of SV and FC from the selected sites are shown in

Table 3. The concentrations of PAHs (μg/kg, DW) in soil samples of SV and FC.

Location/Plant		LMW PAHs
		Nap	Acy	Ace	Flr	Ant	Phe
FJ	SV	106 ± 14	12 ± 1	3.6 ± 0.6	11.9 ± 0.3	3.4 ± 0.7	49 ± 4
	FC	149 ± 3	17.8 ± 0.5	4.3 ± 0.5	12.0 ± 0.1	2.3 ± 0.3	69 ± 4
BN	SV	178 ± 4	52 ± 4	38 ± 5	36 ± 3	166 ± 6	395 ± 13
	FC	68 ± 1	16.8 ± 0.8	5.8 ± 0.3	10.0 ± 0.8	14 ± 1	65 ± 8
SN	SV	165 ± 3	31.0 ± 0.2	15.8 ± 0.7	24 ± 3	9.2 ± 0.6	154 ± 13
	FC	158 ± 13	35 ± 3	16.4 ± 0.7	28 ± 2	8.0 ± 0.6	119 ± 15
NS	SV	64 ± 2	32 ± 1	4.6 ± 0.1	10.1 ± 0.1	17 ± 2	96 ± 2
	FC	165 ± 10	36 ± 2	10.2 ± 0.6	20 ± 2	11.5 ± 0.4	116 ± 8
O	SV	187 ± 26	29 ± 4	16.1 ± 0.8	33 ± 1	4.7 ± 0.3	109 ± 17
	FC	85 ± 6	14.6 ± 0.9	9.0 ± 0.2	26 ± 3	nd	102 ± 9
S	SV	147 ± 8	18.4 ± 0.4	8.2 ± 0.6	33 ± 1	nd	189 ± 11
	FC	111 ± 11	46 ± 1	24.3 ± 0.4	37 ± 2	53 ± 4	160 ± 4
BJ	SV	106 ± 5	21 ± 1	37 ± 2	27 ± 3	nd	255 ± 18
	FC	190 ± 17	35 ± 5	13 ± 1	20 ± 3	nd	86 ± 8
K	SV	66 ± 5	6.2 ± 0.7	13.6 ± 0.4	11 ± 2	nd	221 ± 13
	FC	86 ± 4	9 ± 1	7.5 ± 0.3	14 ± 2	nd	71 ± 3
G	SV	108 ± 7	15 ± 1	9.4 ± 0.1	19.3 ± 0.9	nd	73 ± 5
	FC	143 ± 8	22 ± 2	8.7 ± 0.3	15.5 ± 0.9	5.6 ± 0.9	96 ± 2
USEPA RSL a (mg/kg)
Residential		2.0 b,c	ne b,c	3600 b/360 c	2400 b/240 c	18,000 b/1800 c	ne b,c
Industrial		8.6 b,c	ne b,c	45,000 b/4500 c	30,000 b/3000 c	230,000 b/23,000 c	ne b,c
Location/Plant		HMW PAHs
		Flt	Pyr	BaA	CHR	BaP	BbF+BkF	DhA
FJ	SV	67 ± 4	35.5 ± 0.3	83 ± 9	17 ± 2	8 ± 2	12 ± 1	30 ± 2
	FC	99 ± 4	52 ± 4	71 ± 5	14.6 ± 0.8	7.7 ± 0.9	10.0 ± 0.1	24.7 ± 0.7
BN	SV	1504 ± 70	831 ± 39	560 ± 41	270 ± 25	6.5 ± 0.5	5.0 ± 0.6	70 ± 7
	FC	199 ± 10	119 ± 13	107 ± 5	39 ± 4	5.8 ± 0.6	nd	nd
SN	SV	232 ± 6	131 ± 2	165 ± 20	28 ± 5	9.4 ± 0.3	11.1 ± 0.3	nd
	FC	147 ± 13	81 ± 4	61 ± 6	18 ± 1	9.5 ± 0.3	7.4 ± 0.5	nd
NS	SV	256 ± 3	194 ± 3	108 ± 6	34 ± 2	7.0 ± 0.8	nd	nd
	FC	307 ± 37	260 ± 20	248 ± 13	75 ± 3	9.7 ± 0.1	50 ± 4	128 ± 6
O	SV	68 ± 8	30 ± 2	59 ± 7	nd	9.3 ± 0.2	10.1 ± 0.3	nd
	FC	58 ± 5	23 ± 3	63 ± 5	6.3 ± 0.5	9.8 ± 0.4	13 ± 2	nd
S	SV	271 ± 12	132 ± 9	99 ± 9	42 ± 4	9.3 ± 0.7	8.2 ± 0.3	69 ± 5
	FC	401 ± 15	230 ± 6	309 ± 33	44 ± 3	9.4 ± 0.8	nd	109 ± 6
BJ	SV	38.5 ± 0.6	25 ± 3	42 ± 5	10.0 ± 0.5	10.7 ± 0.5	nd	nd
	FC	102 ± 14	65 ± 7	155 ± 8	14 ± 1	8.7 ± 0.6	8.7 ± 0.4	nd
K	SV	21 ± 2	12.8 ± 0.4	nd	nd	8.2 ± 0.7	nd	nd
	FC	19.5 ± 0.6	11.0 ± 0.6	nd	nd	9.3 ± 0.3	nd	nd
G	SV	24 ± 3	13.2 ± 0.4	38 ± 5	nd	9.0 ± 0.6	nd	nd
	FC	177 ± 1	117 ± 2	68 ± 17	28.6 ± 0.4	9.5 ± 0.6	nd	nd
USEPA RSL a (mg/kg)							BbF/BkF
Residential		2400 b/240 c	1800 b/180 c	1.1 b,c	110 b,c	0.11 b,c	1.1 b,c/11 b,c	0.11 b,c
Industrial		30,000 b/3000 c	23,000 b/2300 c	21.0 b,c	2100 b,c	2.1 b,c	21 b,c/210 b,c	2.1 b,c

SV—Syringa vulgaris; FC—Ficus carica cv. crna Petrovača; nd—not detected; ne—criterion not established; ^a—USEPA regional screening levels (RSL) from generic tables defining maximally allowed concentrations of PAHs in residential and industrial soil [12]; ^b—USEPA summary table based on criteria TR = 1 × 10⁻⁶ and THQ = 1.0 [12]; ^c—USEPA summary table based on criteria TR = 1 × 10⁻⁶ and THQ = 0.1 [12].

and

Table 4. The concentrations of PAHs (μg/kg, DW) in root samples of SV and FC.

Location/Plant		LMW PAHs
		Nap	Acy	Ace	Flr	Ant	Phe
FJ	SV	32 ± 2	35 ± 6	60 ± 7	68 ± 8	21 ± 2	301 ± 34
	FC	31 ± 1	25.7 ± 0.7	38 ± 3	143 ± 9	14 ± 1	269 ± 16
BN	SV	23.4 ± 0.9	3.9 ± 0.1	94 ± 8	127 ± 6	nd	335 ± 12
	FC	129 ± 3	185 ± 9	10.4 ± 0.4	51 ± 3	nd	138.4 ± 0.7
SN	SV	25 ± 3	4.39 ± 0.05	10 ± 1	403 ± 24	19.68 ± 0.05	157 ± 8
	FC	59 ± 2	30 ± 2	15.4 ± 0.1	131 ± 5	nd	133 ± 1
NS	SV	27 ± 2	nd	10 ± 1	222 ± 4	nd	426 ± 19
	FC	99 ± 5	nd	16 ± 2	85 ± 5	nd	145 ± 5
O	SV	24 ± 3	183 ± 17	65 ± 1	331 ± 54	nd	52 ± 2
	FC	29.6 ± 0.7	2.91 ± 0.05	21.2 ± 0.8	81.4 ± 0.8	nd	108.2 ± 0.4
S	SV	52 ± 5	nd	29 ± 1	279 ± 40	nd	nd
	FC	114 ± 7	nd	15 ± 2	300 ± 28	nd	101 ± 3
BJ	SV	20 ± 3	nd	15.0 ± 0.7	627 ± 33	nd	16 ± 2
	FC	89 ± 4	3.1 ± 0.2	5.2 ± 0.6	64 ± 5	nd	nd
K	SV	18 ± 2	nd	111 ± 8	5592 ± 90	nd	nd
	FC	82 ± 8	nd	9.4 ± 0.8	98 ± 8	nd	141 ± 4
G	SV	15.0 ± 0.9	nd	55 ± 3	4990 ± 156	nd	79 ± 9
	FC	79 ± 4	2.3 ± 0.3	7.7 ± 0.8	189.7 ± 0.5	nd	149 ± 8
Location/Plant		HMW PAHs
		Flt	Pyr	BaA	CHR	BaP	BbF+BkF	DhA
FJ	SV	36 ± 5	14 ± 1	nd	nd	nd	2.8 ± 0.5	nd
	FC	124 ± 4	12.5 ± 0.6	77 ± 3	nd	nd	4.5 ± 0.3	90 ± 2
BN	SV	14.2 ± 0.1	5.6 ± 0.4	nd	nd	nd	7.4 ± 0.4	58 ± 5
	FC	nd	nd	192 ± 4	nd	nd	20 ± 1	146 ± 33
SN	SV	385 ± 52	78 ± 13	50 ± 2	nd	nd	nd	nd
	FC	nd	nd	8 ± 1	nd	nd	nd	97 ± 6
NS	SV	36 ± 4	7.2 ± 0.2	nd	nd	nd	8.3 ± 0.2	901 ± 16
	FC	nd	nd	816 ± 52	nd	nd	nd	nd
O	SV	2.5 ± 0.3	2.6 ± 0.1	nd	nd	nd	4.7 ± 0.5	285 ± 26
	FC	nd	nd	nd	nd	nd	nd	172.8 ± 0.3
S	SV	7.1 ± 0.7	6.4 ± 0.8	nd	nd	nd	nd	220 ± 2
	FC	nd	nd	148 ± 27	nd	nd	9.7 ± 0.5	129 ± 3
BJ	SV	231 ± 4	39.5 ± 0.5	nd	nd	nd	8.3 ± 0.3	329 ± 42
	FC	nd	nd	47 ± 6	nd	nd	7.4 ± 0.4	29 ± 4
K	SV	20 ± 3	6.7 ± 0.5	nd	nd	nd	3.7 ± 0.2	191 ± 5
	FC	nd	nd	nd	nd	nd	nd	48.7 ± 0.9
G	SV	17 ± 4	4.3 ± 0.2	nd	nd	nd	2.6 ± 0.3	278 ± 27
	FC	nd	nd	nd	nd	nd	7.3 ± 0.4	55 ± 6

SV—Syringa vulgaris; FC—Ficus carica cv. crna Petrovača; nd—not detected.

, respectively.

It is obvious that PAHs such as BgP and IcP were not detected in both kinds of investigated samples (Table 3 and Table 4), which suggests that these dangerous compounds are not of concern for the local environment and human population. In addition, BaP and CHR were not detected in any of the root samples (Table 4). At the same time, a great majority of the examined soil PAHs were present at almost all locations. Generally speaking, the most abundant soil compounds were Nap, Phe, and Flt for both investigated plants (Table 3). In the case of plant roots, only Nap, Ace, and Flr were present at all locations. At the same time, these compounds were the most abundant in the plant roots (plus Phe, in some cases), which is not unexpected due to their relatively simple molecular structures and maybe due to their optimal values of log Kow, log Koc, etc. (for an easy phytoextraction). In addition, the role of the excreted root exudates cannot be excluded in these but also in all other cases (of the investigated PAHs), as well as the influence of the measured soil parameters. However, it should be pointed here that, before consideration of the calculated BCFs and the computed Pearson’s correlations between the possible influencing factors, some final and accurate conclusions regarding the extraction/accumulation of the investigated compounds cannot be evolved. More precisely, root concentrations as such cannot express exactly the natural bioaccumulation potentials of plants and, generally, the results obtained from the conducted chemical analyses must be further processed using the above-mentioned methods. Compounds such as Ant, Flt, and Pyr in the root of FC were detected only at site FJ, which is the closest to the industrial/metallurgical complex containing the city heating plant (Table 4). The highest concentration in soil samples was found for Flt—1504 μg/kg in SV at site BN, while in the plant roots, it was for Flr—5592 μg/kg in SV at site K, and 4990 μg/kg at site G (Table 3 and Table 4, respectively).

The concentrations of the investigated PAHs in the corresponding soils of SV and FC from the chosen sites (Table 3) ranged from nd for Ant (in SV, sites S, BJ, K, and G, and in FC, sites O, BJ, and K), BaA (in SV and FC, site K), CHR (in SV and FC, site K, and also in SV, sites O and G), BbF+BkF (in SV, sites NS, BJ, K, and G, and in FC, sites BN, S, K, and G), and DhA (in SV and FC, sites SN, O, BJ, K, and G, and also in FC, site BN) to 1503.69 μg/kg (in SV, site BN) for Flt.

Interestingly, all detected soil concentrations were below the related LVs defined in the USEPA RSLs [12], except in the case of DhA in FC at 128 μg/kg (given in bold in Table 3) at the suburb site NS, considering both criteria: (1) TR = 1 × 10⁻⁶ and THQ = 1.0, and (2) TR=1 × 10⁻⁶ and THQ = 0.1 for residential soil. These findings confirm that only regarding DhA (a probable carcinogen) in FC soil at the UI site NS should some rehabilitation measures be applied. However, at the same time, the comparison of the detected soil concentrations with the matching concentrations from an earlier period, i.e., in wild blackberry soils from several identical locations in the region of Bor [23,24], showed that in most cases, there was a rise in individual PAH content, except in the case of BaP, where the soil content noticeably decreased (Table 3). This increase in soil PAH concentrations suggests that environmental concerns and protective measures should also be raised considering the complete local ecosystem. For instance, the filters in the stacks of the city heating plant, as well as in the smelter plant, should be modernized in terms of reducing the emissions of carcinogenic and all other PAHs. Then, the types of vehicles in regional traffic should favor of novel solutions, such as electric or hybrid cars; also, many innovations in the organization of the city landfill must be realized for its proper and adequate work, etc.

Finally, it can be said that, in most cases, the concentrations of individual LMW PAHs were higher in SV soils, whereas the concentrations of individual HMW PAHs were higher in FC soils (Table 3). There was no possibility for the definition of similar rules for the analyzed root samples (Table 4). Additionally, there were no rules regarding the presence of individual PAHs in UIZ and RZ considering both kinds of investigated materials (Table 3 and Table 4), which was the first sign of different generators of PAHs in the investigated region.

The concentrations of individual PAHs in roots of both investigated plants from different sites ranged as follows (Table 4): from nd for Acy (in SV and FC, sites NS, S, and K, as well as in SV, sites BJ and G), Ant (in SV and FC, sites BN, NS, O, S, BJ, K, and G, as well as in FC, site SN), Phe (in SV, sites S and K, and in FC, site BJ), Flt (in FC, sites BN, SN, NS, O, S, BJ, K, and G), Pyr (in FC at sites BN, SN, NS, O, S, BJ, K, and G), BaA (in SV and FC at sites O, K, and G, as well as in SV at sites FJ, BN, NS, S, and BJ), BbF+BkF (in both plants, sites K and G, as well as in SV, sites SN and S, and in FC, sites SN, NS, O, and K), and DhA (in SV, sites FJ and SN, and in FC, site NS) to enormous 5592.09 μg/kg (in SV, site K) for Flr.

By comparing these concentrations with the matching concentrations in blackberry roots from the same selected locations [23,24], it was noticeable that the contents of LMW PAHs in SV and FC roots were generally higher than in blackberry roots, except in the case of Ant. For HMW PAHs, it was not possible to describe any rule, except in the case of CHR, which was not detected in blackberry roots either. It is worth mentioning that BaP was significantly absorbed by blackberry roots, much higher than in the case of SV and FC roots.

In terms of a better understanding of the relations between the investigated PAH concentrations (Table 3 and Table 4) and the distance of the selected locations from the industrial/metallurgical complex (as a possible main source of pollution), Pearson’s correlation study was applied (Supplementary Materials, Tables S1 and S2).

It was shown that all detected SV soil concentrations had negative correlations with the distance, except for BaP; namely, BaP concentrations had a (weak) positive correlation, with the calculated Pearson’s coefficient of p = 0.280 (Table S1). SV root concentrations had mostly negative correlation coefficients with distance but in several cases, positive: Ace, p = 0.040, Flr, p = 0.762*, and DhA, p = 0.135 (the coefficient for Flr was at the statistically significant level, which is labeled with an asterisk in superscript) (Table S1). In general, positive correlations mean that with distance increasing, the concentrations also increase, whereas negative correlations denote the concentrations decreasing [34]. Numerous negative correlation coefficients, calculated for SV, further indicated that the metallurgical plant and the city heating plant represent the emitters of most investigated PAHs (because, in most cases, the detected concentrations decreased with distance increasing). However, given that the calculated correlation coefficients were mostly at the low-to-medium levels (without correlations at the statistically significant levels), it can be concluded that the impact of the mentioned point sources was not of key importance; this further confirmed the presence of some other sources of PAHs that may include various point and diffusion sources in the investigated area. These sources may include the mentioned traffic based on fuel combustion (in constant development, in all zones), the domestic heating in all zones (based on coal, wood, and sometimes fuel), controlled fires in agricultural fields (every year, during autumn) in RZ, some rare accidental fires in forests across RZ, and also, spontaneous ignitions at the city landfill for various waste disposal (in UIZ).

In the case of FC (Table S2), the computed correlation coefficients, although with different values in comparison with SV, also directed us to a very similar conclusion evolved to that for SV. Additionally, in this case, the number of positive correlations (considering bothplant and soil correlations, Table S2) was slightly higher than in the case of SV (Table S1).

As explained previously, the results of PCS (Tables S1 and S2) also helped us to better understand the root accumulation levels in both plants. Namely, given that the accumulation rates of the investigated PAHs in each plant, articulated through the calculated BCFs, showed the different accumulation capacities of the investigated plants (

Table 5. Bioconcentration factors (BCFs) for PAHs in SV and FC.

Location/Plant		LMW PAHs
		Nap	Acy	Ace	Flr	Ant	Phe
FJ	SV	0.30	2.77	16.88	5.71	6.10	6.12
	FC	0.21	1.44	8.91	11.92	6.30	3.90
BN	SV	0.13	0.08	2.45	3.48	nc	0.85
	FC	1.90	11.02	1.81	5.12	nc	2.14
SN	SV	0.15	0.14	0.66	16.60	2.14	1.01
	FC	0.37	0.88	0.94	4.74	nc	1.12
NS	SV	0.42	nc	2.09	21.94	nc	4.42
	FC	0.60	nc	1.55	4.18	nc	1.24
O	SV	0.13	6.26	4.05	10.07	nc	0.48
	FC	0.35	0.20	2.34	3.20	nc	1.06
S	SV	0.35	nc	3.54	8.52	nc	nc
	FC	1.03	nc	0.62	8.01	nc	0.63
BJ	SV	0.19	nc	0.41	23.14	nc	0.06
	FC	0.47	0.09	0.40	3.23	nc	nc
K	SV	0.27	nc	8.15	499.55	nc	nc
	FC	0.47	0.09	0.40	3.23	nc	nc
G	SV	0.14	nc	5.85	258.51	nc	1.08
	FC	0.55	0.10	0.89	12.27	nc	1.55
Location/Plant		HMW PAHs
		Flt	Pyr	BaA	CHR	BaP	BbF+BkF	DhA
FJ	SV	0.54	0.39	nc	nc	nc	0.24	nc
	FC	1.26	0.24	1.09	nc	nc	0.45	3.63
BN	SV	0.01	0.01	nc	nc	nc	1.47	0.84
	FC	nc	nc	1.79	nc	nc	nc	nc
SN	SV	1.65	0.59	0.31	nc	nc	nc	nc
	FC	nc	nc	0.13	nc	nc	nc	nc
NS	SV	0.14	0.04	nc	nc	nc	nc	nc
	FC	nc	nc	3.29	nc	nc	nc	nc
O	SV	0.04	0.09	nc	nc	nc	0.46	nc
	FC	nc	nc	nc	nc	nc	nc	nc
S	SV	0.03	0.05	nc	nc	nc	nc	3.19
	FC	nc	nc	0.48	nc	nc	nc	1.18
BJ	SV	5.98	1.59	nc	nc	nc	nc	nc
	FC	nc	nc	0.30	nc	nc	0.85	nc
K	SV	0.95	0.52	nc	nc	nc	nc	nc
	FC	nc	nc	nc	nc	nc	nc	nc
G	SV	0.71	0.32	nc	nc	nc	nc	nc
	FC	nc	nc	nc	nc	nc	nc	nc

SV—Syringa vulgaris; FC—Ficus carica cv. crna Petrovača; nc—not calculable.

), the additional information from PCS was very useful to elucidate the role and possible impact of the measured soil parameters on PAH uptake/accumulation by roots.

Precisely speaking, very high BCF values were observed for the compound from the group of LMW PAHs, namely Flr, in the roots of SV at sites K (499.55) and G (281.51). BCF values for this compound were greater than 1 in the case of both plants at all tested locations and ranged from 3.20 to 499.55. On the basis of these facts, it can be supposed that both plants can be successfully used in Flr phytoextraction by roots, i.e., in Flr phytostabilization, because dispersal of the pollutant is significantly restricted when stored in plant roots [6,11,18]. Values of BCF > 1 were also noticed for Ace at several locations: FJ, BN, NS, and O for both plants, and additionally, for SV at sites S, K, and G (Table 5).

It should be explained here that enormously high BCF values were also present in cases where the soil concentrations were at the level nd, whereas the corresponding root concentrations had some (more or less) high detected values. This means that the extraction/accumulation of some PAH compounds such as(1) BbF+BkF (sites G, K, BJ, and NS) in SV, and (2) BbF+BkF (sites G, S, and BN), as well as DhA (sites G, K, BJ, O, and SN) in FC were at extremely high levels, and practically, all of the corresponding soil concentrations were extracted by plant roots. Here, the related BCFs were not possible tocalculate. In these cases, extraction/accumulation by the root may also be suggested for application in PAH phytostabilization. It was also impossible to calculate the accumulation rates of PAHs for cases where the root concentrations were at the levelnd (a sign of the absence of PAH uptake/accumulation, especially in the case of BaP and CHR) (Table 5).

Comparing the values of the calculated BCFs for each compound in each plant species from each investigated location, it was noticed that, generally, SV was more successful in Ace and Flr uptake and accumulation, whereas FC was more successful in Nap and DhA uptake/accumulation. For Acy, Ant, Phe, Flt, Pyr, BaA, and BbF+BkF, there were no rules that could be used for the safe recognition of more successful plant actions. However, it could generally be said that there was no rule or confident information (in BCF values), regarding the possible influence of the compounds’ chemical structure or physico-chemical properties during the mentioned processes. These findings are more in favor of the preposition that some soil factors might be counted as significantly influencing and, also, that each plant species developed its own, specific ability for individual PAH uptake/accumulation in the root. However, because both plants retained significant quantities of different PAH compounds in their roots (which is obvious from the numerous very high BCFs), both of them can be recommended for the limitation of PAH spreading in soil, i.e., in their phytostabilization.

Given that there were some clues that several soil factors might be counted as influencing regarding the occurrence of very dissimilar BCF values and different (root) concentration levels as well, the impact of soil PAH concentrations, soil pH, EC, and OM on the level of individual PAH uptake/accumulation in SV and FC roots was finally considered using PCS (Tables S1 and S2).

In the case of SV, the results of PCS (Table S1) showed that soil parameters such as pH, EC, and OM had low and sometimes medium correlations with the root concentrations (having positive as well as negative character). In addition, there was a total lack of domination of some of them, i.e., some individual PAH compounds had positive, whereas the others had negative, correlation coefficients with the soil pH, EC, or OM. The only correlation with a statistically significant value was noticed in the case of the correlation between soil OM and root Flr concentration, OMsoil-Flrroot: p = −0.788*, which means that the locations with the high soil OM had very low concentrations of root Flr, and vice versa. This also means that soil OM can be counted as the only controlling (soil) factor during Flr uptake/accumulation processes in SV. In addition, this Pearson’s coefficient confirmed that the highest root concentrations were found exactly in the case of Flr in SV at sites K and G (Table 4), where the soil OM content was at the lowest level (Table 1). All of the other correlations suggest that there was a different kind of influence of the measured soil parameters but not at some strong or controlling level. In the case of FC (Table S2), the situation was similar—numerous positive and negative correlations were present, withonly one statistically significant coefficient for the relation: ECsoil-Flrroot: p = 0.694*; namely, the locations with the high soil EC (Table 1) also had very high root Flr concentrations (Table 4).

When the relations between soil and root PAH concentrations in both plants were investigated using PCS (Tables S1 and S2), it was evident that all kinds of correlation coefficients were present: low, medium, and high, having positive or negative character, and a number of them were even at the statistically significant level, excluding the cases of pairs of the corresponding root and soil PAHs.

For example, the correlations between the concentrations of the investigated individual PAHs in roots and the related, i.e., corresponding, soil concentrations were as follows (Tables S1 and S2): (1) for Nap in SV: p = 0.215, and in FC: p =−0.157; (2) for Acy in SV: p = 0.104, and in FC: p = −0.270; (3) for Ace in SV: p = 0.152, and in FC: p = −0.267; (4) for Flr in SV: p = −0.419, and in FC: p = 0.608; (5) for Ant in SV: p = −0.168, and in FC: p = −0.183; (6) for Phe in SV: p = −0.043, and in FC: p = −0.279; (7) for Flt in SV: p = −0.143, and in FC: p = −0.213; (8) for Pyr in SV: p = −0.130, and in FC: p = −0.235; (9) for BaA in SV: p = −0.082, and in FC: p = 0.609; (10) for BbF+BkF in SV: p = −0.560, and in FC: p = −0.442, and (11) for DhA in SV: p = −0.346, and in FC: p = −0.258.

Clearly, most of these correlations were at a very low level and with negative character (except in the cases of BbF+BkF in SV, and Flr and BaA in FC), which symbolizes the decrease in PAH uptake with the soil concentration increasing (although to a low extent). At first sight, this is a little unusual because, generally, root/plant concentrations have positive correlations with the corresponding concentrations in the soil. However, this situation may also be a new indication that, in general, each plant species has a specific mechanism for the uptake/accumulation of individual PAHs (as well-known toxic compounds), including, probably, different roles of different and diverse lipid contents in the root membranes and maybe the excretion of different kinds of root exudates [6]. In that sense, it seems that for some PAHs both tested plant species developed a strategy of avoiding compound assimilation. It can be supposed that some of the detected soil quantities of PAHs were toxic for SV and FC (in the cases of many locations for Acy, Ant, Flt, Pyr, BaA, and especially CHR and BaP) (Tables S1 and S2). Unlike this situation, Alagić et al. [23] showed that blackberry roots from the same locations were very tolerant of BaP, because they were able to accumulate its significant quantities in the roots, often at a BCF level a little higher than 1.

On the other hand, some statistically significant coefficients were found in the cases of relations between several individual root PAHs and various PAHs in soils: (1) for SV: Antroot-BbF+BkFsoil, p = 0.678*; Pheroot–BaPsoil, p = −0.740* (Table S1), and (2) for FC: Acyroot–BaPsoil, p = −0.899**; Flrroot–Antsoil, p = 0.772*; BaAroot–Pyrsoil, p = 0.766*; BaAroot–CHRsoil, p = 0.884**; BaAroot–BbF+BkFsoil, p = 0.879**; BaAroot–DhAsoil, p = 0.787*; BbF+BkFroot-BaPsoil, p = −0.795*; DhAroot–Napsoil, p = −0.686*(Table S2).

This situation may be a sign that (in the case of both plants), the locations with high soil concentrations of some PAHs also had very high concentrations of some other individual root PAHs; practically, it seems that some soil PAHs enhanced the uptake of some other PAHs. Simultaneously, some other soil PAHs made the uptake of different individual PAHs more difficult (which is denoted with negative coefficients). All of these facts may further represent an indication of some strong competitive but also some synergistic effects during the uptake of different PAHs at the investigated locations. All other root–soil correlations (both positive and negative) were mostly atvery low levels, and only sometimes atmedium levels, which means that they cannot be counted as safe indicators of influencing parameters.

The statistically significant correlations were also found between the concentrations of some individual root PAHs, such as(1) for SV: Fltroot–Pyrroot, p = 0.992**, Fltroot–BaAroot, p = 0.847**, Pyrroot-BaAroot, p = 0.893** (Table S1), and (2) for FC: Acyroot-BbF+BkFroot, p = 0.782*, Aceroot–Antroot, p = 0.868**, Aceroot-Pheroot, p = 0.757*, Aceroot-Fltroot, p = 0.868**, Aceroot-Pyrroot, p = 0.868**, Antroot-Pheroot, p = 0.745*, Antroot-Flrroot, p =1.000**, Antroot-Pyrroot, p =1.000**, Pheroot-Fltroot, p =1.000**, Pheroot-Pyrroot, p = 0.745*, and Fltroot-Pyrroot, p =1.000** (Table S2).

Obviously, the investigated locations had various pairs of root PAHs with positively correlated concentrations, and similarly to the case with the root–soil correlations, the rest of the root–root correlations (positive or negative) were atvery low levels, and very rarely, atmedium levels.

Finally, statistically significant correlations were also found between different soil PAHs: (1) for SV: Napsoil-Flrsoil, p = 0.870**, Napsoil-BbF+BkFsoil, p = 0.668*, Acysoil–Antsoil, p = 0.809**, Acysoil–Fltsoil, p = 0.827**, Acysoil–Pyrsoil, p = 0.842**, Acysoil–BaAsoil, p = 0.859**, Acysoil–CHRsoil, p = 0.798*, Acesoil–Phesoil, p = 0.850**, Acesoil–Phesoil, p = 0.850**, Antsoil-Phesoil, p = 0.744*, Antsoil-Fltsoil, p = 0.984**, Antsoil-Pyrsoil, p = 0.983**, Antsoil-BaAsoil, p = 0.973**, Antsoil-CHRsoil, p = 0.987**, Antsoil-BaPsoil, p = −0.677*, Phesoil–Fltsoil, p = 0.750*, Phesoil–Pyrsoil, p = 0.734*, Phesoil–BaAsoil, p = 0.697*, Phesoil–CHRsoil, p = 0.762*, Phesoil–BaAsoil, p = 0.697*, Fltsoil-Pyrsoil, p = 0.997**, Fltsoil-BaAsoil, p = 0.987**, Fltsoil-CHRsoil, p = 0.997**, Fltsoil-DhAsoil, p = 0.694*, Pyrsoil-BaAsoil, p = 0.984**, Pyrsoil-CHRsoil, p = 0.993**, Pyrsoil-BaPsoil, p = −0.697*, CHRsoil-DhAsoil, p = 0.709*, BaAsoil-CHRsoil, p = 0.983** (Table S1), and (2) for FC: Acysoil-Acesoil, p = 0.842**, Acysoil-Flrsoil, p = 0.728*, Acysoil-Phesoil, p = 0.811**, Acysoil-Fltsoil, p = 0.768*, Acysoil-Pyrsoil, p = 0.727**, Acysoil-BaAsoil, p = 0.831**, Acesoil-Flrsoil, p = 0.921**, Acesoil-Antsoil, p = 0.768*, Acesoil-Phesoil, p = 0.900**, Flrsoil-Antsoil, p = 0.671*, Flrsoil-Phesoil, p = 0.941**, Antsoil-Phesoil, p = 0.759*, Antsoil-Fltsoil, p = 0.867**, Antsoil-Pyrsoil, p = 0.706*, Antsoil-BaAsoil, p = 0.788**, Phesoil-Fltsoil, p = 0.737*, Phesoil-BaAsoil, p = 0.713*, Fltsoil-Pyrsoil, p = 0.957**, Fltsoil-BaAsoil, p = 0.900**, Fltsoil-CHRsoil, p = 0.849**, Fltsoil-DhAsoil, p = 0.829**, Pyrsoil-BaAsoil, p = 0.882**, Pyrsoil-DhAsoil, p = 0.953**, BaAsoil-CHRsoil, p = 0.771**, BaAsoil-DhAsoil, p = 0.868**, and CHRsoil-DhAsoil, p = 0.815** (Table S2).

The rest of the correlations for soil–soil PAH concentrations were at the low or medium level, practically, with insignificant effects.

As can be seen from all the presented Pearson’s coefficients, the maximum number of statistically significant correlations was found in the latest example, i.e., in the case of different pairs of soil PAHs (for each plant). Many authors provided evidence that the identified pairs of pollutants with positive correlations may represent a verification of their common origin [36]. As clarified before, albeit prevailingly of anthropogenic origin, some minor quantities of PAHs in soils may also originate from natural processes (occurring in soils and in the atmosphere as well) [5,37]. It can be supposed that the sampled soils from the city of Bor and its surroundings also contained both kinds of PAHs—naturally and anthropogenically formed from pyrogenic, petrogenic, diagenetic, and biogenic sources. Also, due to their recalcitrant nature, PAHs may form an important recalcitrant pool in soil, reflecting in that way a situation mainly resulting from long-term soil exposure [15,38]. The greatest amounts of significant positive correlation coefficients (≈1) between the pairs of different soil compounds (for each plant) suggested that their origin was the most similar (probably anthropogenic), whereas negative correlation coefficients between BaP and some other soil PAHs suggested that this particular compound had the most dissimilar, i.e., natural (most probably diagenetic, or biogenic) origin, and especially in SV soil, where BaPhad negative correlations with all PAHs (and in particular with Ant and Pyr, Table S1).

All of the explained details of the investigated matrices, namely, soils and roots, pointed to very complicated processes during PAH extraction/accumulation and supported some similar details also found in the case of wild blackberry from the Bor region [23,24]. The first similarity is thatplant roots are not capable of reflectingtheactual situation in the soil totally precisely, so they cannot be used in classical biomonitoring procedures as a safe tool for the detection of soil pollution. The second similarity is thatthe level of the current soil pollution can be better estimated by analyzing the sampled soils, which also may provide some useful indications regarding the origin and possible sources of individual PAHs. Finally, the investigated plant species showed different abilities to manageindividual PAHs, because each plant species showed an unique capability, sensitivity, and tolerance regarding most of the investigated compounds; as follows: “only in some cases, and especially regarding BaP and CHR, SV and FC developed a similar avoiding strategy, totally excluding these compounds from the uptake, which is the most important dissimilarity with the wild blackberry.

The described specifics of the two investigated plants are evidently illustrated through different classifications of the polluted locations in the related hierarchical dendrograms for plants’ soils and roots (Figure 2). They were computed based on the soil as well as root PAH concentrations of SV (Figure 2a,b, respectively) and FC (Figure 2c,d, respectively). During the performed HCA, these concentrations were treated as variables, the chosen locations were treated as cases, while the applied method was Ward’s linkage with the squared Euclidean distance as a measurement interval.

Large differences between the SV hierarchical dendrograms based on soil (Figure 2a), and root concentrations (Figure 2b) are obvious. Although they both have two main clusters, the arrangements of the locations are completely different. The dendrogram based on soil concentrations (Figure 2a) separated the UI location BN (as the most polluted) into an isolated (first) main cluster, whereas the first main cluster, based on root concentrations (Figure 2b), consisted of the two R locations K-G. The second main cluster in the dendrogram based on soil concentrations (Figure 2a) was constituted by the two main subclusters containing numerous other subclusters in the grouping as follows, NS-S-SN/O-G-FJ-K-BJ, whereas the second main cluster in the dendrogram based on root concentrations (Figure 2b) was even simpler: NS-BJ-N-S-O-BN-FJ.

The groupings of locations in the analogous hierarchical dendrograms for FC (Figure 2c,d) were a little more complicated but, again, the differences between the dendrograms based on soil (Figure 2c), and root concentrations (Figure 2d) were very clear.

These dendrograms based on soil and root contents (separately) confirmed that plant roots cannot illustrate the authentic situation in the corresponding soil, which further points to the problematic utilization of SV and FC roots in classically planned phytomonitoring procedures. At the same time, the results obtained from the analysis of the corresponding soil samples provided more informative data about the current level of soil PAHs, as well as into the possible sources of PAHs.

4. Conclusions

On the basis of the results of this study, obtained from the applied combination of methods such as QuEChERS, GC/MS, BCF, HCA, and PCS, it can be supposed that the presented dissimilarities between the investigated plant species may be ascribed to their different managing actions and sensitivity regarding the individual PAHs. However, despite the obviously dissimilar results, the final effects of both plants, SV and FC (considering their prospective application in soil monitoring and remediation), were very similar. Namely, a careful analysis of the obtained results suggests that the key controlling role in the complete and very complex processes of PAH uptake/bioaccumulation had each plant species by its self, whereas the other (possible influencing) processes/factors were not of crucial significance, including the influence of a majority of soil factors, different chemical and physical properties of the investigated compounds, processes of competition, synergism, impacts of the existing sources of PAHs, etc. Namely, it is more possible that the roots of SV and FC have some mechanisms and ability to recognize and then respond adequately to the individual compound’s toxicity. These mechanisms may include various roles of probably different phospho-lipid components (and contents, too) in their root cell membranes, or the excretion of different kinds of exudates, which may help in(1) the immobilization or (2) the liberation of PAHs from soil (existing there in different forms), and laterin their extraction and capture. These plant potentials may be suitably applied in PAH phytostabilization but not in soil PAH monitoring. Being very important, the supposed mechanisms will be better explored in one of our future experimental works.

Planting of the investigated species, SV and FC, which may survive in the different and obviously very aggressive circumstances of a mining/metallurgical region can be recommended for phytostabilization purposes in various climatic areas and also in diverse orographic and geochemical conditions.

Finally, based on the closing results of this study, it can be said that, in general, a modern approach in environmental protection and management cannot be organized without the careful planning and application of environmentally friendly technologies—most desirably, in natural environments (i.e., in situ) where the investigated plants show their natural potential in the best way and where the most important factoristhe right choice of prospective plant species—and then a proper combination of highly accurate chemical methods, adequate indices, and a safe and sound statistical analysis.