Determination of Polycyclic Aromatic Hydrocarbons from Atmospheric Deposition in Malva sylvestris Leaves Using Gas Chromatography with Mass Spectrometry (GC-MS)

Abstract

Plant leaves can be used to determine the atmospheric deposition of organic contaminants, including polycyclic aromatic hydrocarbons (PAHs), to assess the contamination status of an area. The purpose of this study was to develop an analytical method for the determination of PAHs deriving from atmospheric deposition using Malva sylvestris leaves. Analytes were recovered from the leaves of the plant using cyclohexane as an organic solvent and subsequent sonication. The percentage recoveries (R%) were good (from 65.8 ± 3.2 to 104.2 ± 16.9), together with the instrumental analytical parameters, including correlation coefficients (r) ≥ 0.995 for all PAHs. The instrumental analysis was carried out using GC-MS in total ion current and single ion monitoring at the same time. Real samples taken from urban environments have shown that they are not always the most contaminated. At the Palermo site, leaves were observed to have high amounts of PAHs due to the deposition of dust generated by combustion processes that occurred near the sampling site. Further studies are recommended to compare the use of plants and classical sampling systems for monitoring the atmospheric deposition of key contaminants toxic to human health.

1. Introduction

Polycyclic aromatic hydrocarbons are a class of compounds formed by two or more aromatic (benzene) rings condensed together [1]. They are generated by incomplete combustion processes such as forest fires, combustion in agriculture, automobile exhaust, home heating, and cigarette smoke [2]. In the large group of PAHs, the International Agency for Research on Cancer (IARC) classifies 16 molecules as certain, possible, and probable carcinogens (group 1, 2A, and 2B, respectively). Of these, benzo[a]pyrene (BaP) represents the most widely used molecule as an indicator of the PAHs class, and is the only one classified by IARC as a certain human carcinogen [3]. At the European level, on the other hand, under Regulation 1272/2008/EU, BaP has been classified as a category 1B carcinogen: substances for which carcinogenic effects to humans are presumed [4]. Along BaP, seven other molecules have been placed in Group 1B:chrysene (CHR), benzo[e]pyrene (BeP), benzo[b]fluoranthene (BbFA), benzo[j]fluoranthene (BjFA), benzo[k]fluoranthene (BkFA), benzo[a]anthracene (BaA), and dibenzo[a,h]anthracene (DBahA). In terms of regulation, at the international level, there are no limit values in the air for PAHs, but the World Health Organization (WHO) recommends a guideline value for BaP in ambient and indoor air of 0.012 ng m⁻³ (UnitRisk/lifetime) 10⁻⁶ [5]. In Europe, limit values have been established for BaP and for the quantitative sum of BaP, BaA, BbFA, and CHR in various foods ranging from 1 to 50 μg kg⁻¹ [6]. In fact, the European Food Safety Authority (EFSA) specifies that regulated PAHs are currently the only possible indicators of the carcinogenic potency of PAHs in food [7]. PAHs, when emitted from the source are released directly into the atmosphere where they can be in gaseous form or be bound to atmospheric particulate matter (PM) or settleable particulate matter (SPM). SPM subsequently reaches surfaces (soils, plants, vegetation, buildings, water bodies) through atmospheric deposition (AD). In fact, one of the main pathways of PAH contamination of plants, vegetables, and soils is AD [8,9,10]. Plant leaves therefore can be an excellent natural “sampler” of the PAHs present in sedimentable dust. Obviously, this phenomenon is influenced by the surface area of the leaf (the larger the surface area, the greater the SPM collected), the location of the plant (rural, urban, or industrial areas), and the season (hot or cold) [11]. In this perspective, the use of plants and vegetables can be useful in monitoring the PAH contamination status of an area and assessing the possible presence of neighboring emission sources. There are many papers in the scientific literature in which the use of plants, better known as bioaccumulators and consequently bioindicators (lichens, mosses) are used to monitor the contamination status of an area and the levels present of major persistent organic pollutants (POPs), such as PAHs, polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), polychlorinated biphenyls (PCBs), and heavy metals [12,13,14,15,16]. Lichens and mosses are the most effective bioaccumulating organisms because they are permanent in time, have no roots and consequently absorb nutrients and pollutants exclusively from the air. Other studies, however, have shown the suitability of other wild plant species in environmental biomonitoring. For example, Quercus ilex L., Pinus, and Sambucus nigra leaves have been used for monitoring PAHs in urban areas [17,18,19,20]. The use of bioaccumulating organisms and plants may be advantageous over classical environmental monitoring methods, such as high-volume samplers for PM. This is because some plant species possess accumulative leaf structures, are permanent over time, provide current photography of the contamination status of an area, and have low operating costs [11]. Malva sylvestris, better known as common Mallow, is a wild edible plant widespread in Europe, Africa, and Asia and is commonly used as a medicinal plant and consumed as a vegetable in human nutrition. Numerous studies demonstrate the high amounts of bioactive compounds (polyphenols, flavonoids, coumarins, and vitamins) and health benefits of its consumption in the diet [21,22,23]. However, of its use as an environmental bioindicator in POP monitoring, only a few studies have been published. In 2013 and 2016, Terzi et al., used M. sylvestris (leaves, roots, and stems) as a bioindicator to assess the levels of PAHs and PCBs in the city of Kocaeli in Turkey [24,25]. In a 2015 paper, the amounts of some heavy metals in M. sylvestris sold in supermarkets were determined, indirectly assessing the contamination status of the area [26]. In this context, the aim of this paper was to develop and optimize a rapid and simple analytical method for the qualification and quantification of PAHs deposited through AD on M. sylvestris leaves. Thus, M. sylvestris will be used to assess the degree of PAH contamination on the territory in two Italian urban areas, trying to establish through PAH profiles the emission sources. Among the PAHs considered, those recommended by EFSA and the European Community used in health assessments and contaminant legislation were determined.

2. Materials and Methods

2.1. Chemicals

Standard solutions of naphthalene (NA), acenaphthylene (ACL), acenaphthene (AC), fluorene (FL), phenanthrene (PHE), anthracene (AN), fluoranthene (FA), pyrene (PY), indeno[1,2,3-cd]pyrene (IP), BaA, CHR, BbFA, BjFA, BkFA, and BaP dissolved in toluene were used. These standard solutions were purchased from CPAChem Ltd., Bogomilovo, Bulgaria. As an internal standard (I.S.), a solution of two deuterated PAHs in toluene at a concentration of 12 µg mL⁻¹ was used. Specifically, deuterated perylene D₁₂, 98% (PE D₁₂) and deuterated chrysene D₁₂, 98% (CHR D₁₂) were purchased from Cambridge Isotope Laboratories (CIL), Tewksbury, MA, USA. The organic solvents used in various extraction tests were toluene and cyclohexane, both of SPS^TM grade (Super Pure Solvents) to minimize instrumental background noise. These solvents were purchased from Romil Pure Chemistry. Additionally, n-pentane, n-hexane, and dichloromethane were purchased from Sigma-Aldrich (Milan, Italy), as well as the anhydrous sodium sulfate.

2.2. Sample Collection

Leaves of M. sylvestris were sampled in a rural area in the Molise Region, Rotello (Italy; 300 km SW from Rome), two urban areas (Rome and Palermo), and two background sites used as references for the respective cities of Rome and Palermo. Being classified as urban areas, the main emission sources of PAHs in the cities of Rome and Palermo are car emissions (exhaust gases) and recurring fires during the summer season. Moreover, Palermo, being a maritime city, is characterized by strong shipping transport, which contributes significantly to emissions of pollutant gases (nitrogen oxides (NO_x), carbon monoxide (CO)), and POPs (including PAHs). On the other hand, the rural site of Rotello, with a population of around 1000 people, does not have a high level of car traffic and no industrial settlement. Sampling was conducted during the summer, from June to September 2023, as the hot season is characterized by high-ground deposition (and thus higher contaminant input) and dry weather conditions [27]. Assuming a low contamination level, the leaves collected at the Rotello site were used for the development and optimization of the extraction and analysis procedure for PAHs in GC-MS. In Figure 1, the sampling points in the sites considered in this study are indicated.

For each sampling site, multiple leaves were collected at ten adjacent points to ensure representative sampling [11]. The leaves, once collected, were stored in sterile dark glass containers and kept at a low temperature with ice throughout the transport to avoid potential degradation of PAHs due to light exposure. Upon arrival at the laboratory, all samples were stored at −20 °C till analysis.

2.3. Extraction of PAHs from Dust Deposited on Malva sylvestris Leaves

The authors chose to extract and determine PAHs in the dust deposited on the surface of Malva sylvestris leaves because the leaves act as a “passive” sampler for sedimentable particulate matter. The leaves were defrosted at a low temperature, removing those affected by cuts, necrosis, and yellowing. In the samples collected in Rotello, it was verified that there were no detectable amounts of PAHs to use as “blank” samples. Since there is no certified reference material, recovery tests of analytes were conducted on the blank sample (2 g of leaves [20,24]) after adding a known volume of standard solution in toluene (500 µL at 0.5 µg mL⁻¹). Before proceeding with the extraction, toluene was allowed to evaporate and the PAHs were left to equilibrate with the leaf matrix. The final concentration of PAHs in the spiked samples was 125 ng g⁻¹. The extraction was carried out by adding 30 mL of SPS^TM-grade cyclohexane to the beaker containing the leaves, covering it with aluminum foil, and placing it in an ultrasonic bath (Transsonic 310 by Elma^®) at a power of 400 W for 10 min. The extracts were collected in a dark glass 250 mL flask to avoid photodegradation. The extraction was repeated twice, adding a total of 60 mL of cyclohexane. All extracts were collected in the same flasks and evaporated in a Büchi rotary evaporator at a maximum bath temperature of 30 ± 0.5 °C and a pressure of 235 mbar [28]. The extract was concentrated to a final volume of 2 mL and then passed through a glass column containing anhydrous sodium sulfate (dried in an oven for 12 h at 120 °C), previously conditioned with cyclohexane. The flasks containing the extracts were rinsed twice with 1 mL of fresh cyclohexane and also passed through a Na₂SO₄ column. This allowed a quantitative recovery of analytes and column washing. The collected solutions were then transferred to tared conical-bottom vials and concentrated under a gentle stream of N₂ with purity >98% to a final volume of 100 µL. No clean-up process was applied to the samples because the matrix interferents did not overlap with the peaks of the analytes of interest. Before instrumental analysis, 20 µL of I.S. solution was added to each sample to a final concentration similar to the amount of added PAHs. The main function of the I.S. is to normalize all data, preventing errors due to small variations in solution volumes or injected volumes. One µL of the samples was injected into the GC-MS.

2.4. Instrumental Analysis of PAHs by GC-MS

Instrumental analysis was conducted using an Agilent Gas Chromatograph (GC-6890) coupled with a single quadrupole mass spectrometer (MSD-5973). For the separation of PAHs, a DB-EUPAH column (60 m × 250 µm × 0.25 µm i.d.) purchased from Agilent (Milan, Italy) was used. Ultrapure helium was used as the mobile phase at a constant flow rate of 1.2 mL min⁻¹. A split–splitless injector in the splitless mode was used. Injector temperature and pressure were set at 250 °C and 21.7 psi, respectively, and the splitter valve was opened after 60 s. To optimize the separation, the following temperature program was set in the GC oven: the initial temperature was 70 °C, held for 60 s, then at 300 °C (5 °C min⁻¹) for 15 min and 325 °C (10 °C min⁻¹) for 10 min with a total run time of 74.5 min. The transfer line (TL) temperature between the end of the column and the ionization chamber was set at 300 °C. Ionization of the molecules was done through electronic impact (EI), with ionization energy at 70 eV. As for the MSD, the temperature of the source and quadrupole were set at 240 and 180 °C, respectively. The acquisition of the instrumental signal was carried out simultaneously in full scan mode (detection of all fragments) and single ion monitoring (SIM). Therefore, two chromatograms were acquired for each sample.

Table 1. Retention time, number of benzene rings, molecular ion (M+), and ion confirmation (Q_x) for each PAH.

Compound	Retention Time (min)	Benzene Rings	M+ (m/z)	Q1 (m/z)	Q2 (m/z)
NA	17.5	2	128	127	102
ACL	25.7	3	152	151	76
AC	26.4	3	153	154	76
FL	28.8	3	166	165	82
PHE	34.2	3	178	176	152
AN	34.4	3	178	176	152
FA	40.4	4	202	200	101
PY	41.9	4	202	200	101
BaA	47.9	4	228	226	113
CHR-D12	48.2	4	240	120	-
CHR	48.3	4	228	226	113
BbFA	55.3	5	252	253	126
BjFA	55.5	5	252	253	126
BkFA	55.8	5	252	253	126
BaP	59.3	5	252	253	126
PE-D12	60.3	5	264	132	-
IP	69.2	6	276	138	-

shows the retention times and individual fragments acquired in SIM mode for all the considered PAHs.

The qualitative analysis was performed by comparing both the retention times and the mass spectra of each individual PAH with the reference standard mixture under the same instrumental conditions. Quantitative analysis was conducted by constructing the calibration curve (linear regression with the least squares method) for each PAH, with the ratio of the analyte area to the I.S. area (A_a A_I.S.⁻¹) plotted against concentration in µg mL⁻¹. The identification of PAHs in the samples was carried out by referencing retention times and mass spectra in SIM mode. All tests were conducted in triplicate.

3. Results and Discussion

To optimize the extraction process of PAHs from M. sylvestris leaves, the authors evaluated the extraction capacity of different organic solvents and the influence of ultrasonication times during sonication.

3.1. Optimization of PAHs Extraction Conditions

To evaluate the extraction efficiency, five organic solvents of different polarity were tested: dichloromethane (DCM), n-pentane, hexane, cyclohexane, and toluene. Recoveries (R%) were calculated from the ratio of the PAH concentration determined in the fortified sample to the known PAH amount added in the fortified sample, all multiplied by 100. Three extractions were conducted for each solvent, and the average percentage recoveries (R%) and respective standard deviations are shown in

Table 2. PAH recoveries (R%) related to different extraction organic solvents with relative s.d.

Compound	DCM	n-Pentane	Hexane	Cyclohexane	Toluene
NA	42.8 ± 2.4	110.0 ± 16.2	25.2 ± 6.1	69.6 ± 8.7	5.1 ± 0.9
ACL	41.0 ± 6.1	47.5 ± 7.3	24.1 ± 6.2	66.7 ± 2.9	4.2 ± 1.2
AC	11.4 ± 4.2	13.4 ± 1.9	7.3 ± 1.8	65.8 ± 3.2	7.3 ± 0.8
FL	29.4 ± 6.6	36.1 ± 7.2	14.7 ± 5.0	71.2 ± 10.3	7.1 ± 1.1
PHE	44.8 ± 15.8	63.9 ± 12.3	25.3 ± 7.0	83.7 ± 6.7	49.0 ± 2.6
AN	47.1 ± 16.5	60.9 ± 11.6	27.4 ± 7.4	104.2 ± 16.9	40.4 ± 1.8
FA	39.8 ± 14.5	57.7 ± 12.3	23.0 ± 6.9	75.3 ± 5.9	71.9 ± 7.8
PY	38.3 ± 14.6	55.6 ± 12.0	22.3 ± 6.9	73.5 ± 9.7	57.0 ± 13.5
BaA	33.5 ± 9.6	48.9 ± 11.8	22.2 ± 7.4	93.5 ± 12.9	12.5 ± 1.5
CHR	26.9 ± 2.3	34.9 ± 7.9	13.7 ± 8.8	102.1 ± 7.6	86.8 ± 16.0
BbFA	35.9 ± 3.5	56.7 ± 14.5	20.2 ± 6.9	81.5 ± 3.3	-
BjFA	35.9 ± 4.0	59.5 ± 15.0	21.9 ± 7.2	83.4 ± 2.4	-
BkFA	36.1 ± 5.5	57.5 ± 14.8	21.3 ± 6.9	79.6 ± 3.5	-
BaP	33.5 ± 6.1	46.5 ± 12.6	19.3 ± 7.2	78.3 ± 1.6	-
IP	21.8 ± 2.8	23.8 ± 6.4	15.1 ± 5.6	62.3 ± 4.4	-

.

Cyclohexane, compared to all other solvents, has the highest recoveries for all PAHs except naphthalene (69.6 ± 8.7), where the highest recovery occurred using n-pentane (110.0 ± 16.2). Using cyclohexane, R% ranged from 62.3 ± 4.4 to 104.2 ± 16.9, which is much higher in comparison with DCM (11.4 ± 4.2 to 47.1 ± 16.5), n-pentane (23.8 ± 6.4 to 110.0 ± 16.2), and hexane (7.3 ± 1.8 to 27.4 ± 7.4). Toluene, on the other hand, showed the worst recoveries, especially for low molecular weight PAHs with 2–3 benzene rings. This may be justified by the fact that toluene, having a higher boiling temperature (110.6 °C at 1 atm) than the other solvents, requires a prolonged time during vacuum evaporation, resulting in the risk of loss due to volatilization of low molecular weight PAHs, especially naphthalene, acenaphthylene, acenaphthene, and fluorene. In addition, the standard deviation values for PAH recoveries with cyclohexane were shown to be low compared to the other solvents, indicating good repeatability of the extraction. In other studies, however, other organic solvents have yielded good R% for PAH extraction. For example, Orecchio et al. [20] showed that using DCM to extract PAHs from Pinus leaves yields an average R% of all compounds of 82%. In another work, a DCM-acetone 1:1 = v:v mixture was used to extract PAHs from Quercus ilex L. leaves, achieving good recoveries [18]. The ISTISAN Report 03/22 [9] also suggests using acetone or a mixture of acetone and DCM for the extraction of PAHs from plants. In this study, the recoveries obtained using cyclohexane as the extraction solvent for PAHs from Malva sylvestris leaves turned out to be the best compared to the others. This suggests that depending on the type of plant, some organic solvents perform better than others.

The influence of sample residence time in the ultrasonic bath on the extraction of PAHs was also evaluated. Specifically, the R% of all analytes was determined in triplicate by subjecting M. sylvestris leaves to three different sonication times: 2, 10, and 20 min. Three times were chosen of which one can be considered short (2 min), intermediate (10 min), and long (20 min), according to other studies using ultrasonic extraction of PAHs and other POPs (sonication time ranging from 10 to 20 min).

Table 3. PAHs recoveries (R%) related to different sonication times with relative s.d.

Compound	2 min	10 min	20 min
NA	97.5 ± 1.0	87.4 ± 2.3	93.3 ± 12.4
ACL	77.4 ± 3.2	77.7 ± 11.9	74.5 ± 3.0
AC	61.3 ± 1.0	61.4 ± 3.2	60.8 ± 0.5
FL	59.9 ± 3.6	61.9 ± 9.5	55.6 ± 2.2
PHE	91.3 ± 9.2	97.4 ± 5.6	83.8 ± 3.6
AN	108.1 ± 5.5	112.0 ± 10.0	100.0 ± 6.4
FA	82.9 ± 6.3	88.0 ± 6.7	76.7 ± 1.8
PY	80.4 ± 5.7	84.6 ± 6.3	75.1 ± 2.3
BaA	65.5 ± 3.6	67.2 ± 3.2	67.9 ± 1.4
CHR	95.5 ± 4.4	96.2 ± 5.2	99.3 ± 7.4
BbFA	83.4 ± 3.3	85.8 ± 3.6	79.8 ± 8.9
BjFA	85.9 ± 2.8	88.9 ± 4.3	82.9 ± 8.8
BkFA	83.5 ± 2.9	86.4 ± 4.5	80.5 ± 8.5
BaP	68.3 ± 3.5	69.7 ± 3.8	70.1 ± 2.5
IP	60.7 ± 4.0	64.4 ± 2.1	62.0 ± 6.0

shows the recoveries of PAHs at the three different sonication times with their respective s.d.

The recoveries for all the three times tested show little difference. Comparing the tests at 2 and 10 min, it can be seen that the R% are always higher in the10-min test for all compounds except naphthalene (97.5 ± 1.0 at 2 min and 87.4 ± 2.3 at 10 min). The % recoveries range from 60.7 ± 4.0 to 108.1 ± 5.5 and 61.4 ± 3.2 to 112.0 ± 10.0 for the 2 and 10 min tests, respectively. Comparing the 10- and 20-min tests, however, it is evident that eleven PAHs (including ACL, AC, FL, PHE, AN, FA, PY, BbFA, BjFA, BkFA, and IP) out of fifteen show slightly higher recoveries in the 10 min tests. Thus, for most compounds, the greatest recoveries were found using a sonication time of 10 min. In general, among the methods for extracting PAH from plant and food matrices, the use of ultrasound is satisfactory and widely applied. Indeed, in other publications, sonication is effectively used to aid the extraction of PAH from the plant matrix. For example in [19], the R% of PAHs from Sambucus nigra leaves obtained using ultrasound ranged from 64.8 ± 1.3 to 75.1 ± 1.9 for the lowest molecular weight PAHs (NA, ACL, AC, FL) and from 70.0 ± 7.6 to 91.2 ± 16.9 for the highest molecular weight PAHs (BaA, BaP, BbFA, BjFA, IP). Terzi et al. [24] also conducted the extraction of PAHs using ultrasound at 25 °C and obtained high recoveries for all the compounds analyzed. Based on the overall results, a sonication time of 10 min and cyclohexane as the extraction solvent were chosen as the best extraction conditions. The authors, after defining the best-extracting solvent and verifying the best ultrasonic sample residence time, conducted recovery tests at two different concentrations (400 and 10 ng g⁻¹, identifiable as high and low concentration, respectively) of which the one at 10 ng g⁻¹ was similar to the expected PAHs concentrations in real samples. In

Table 4. PAH recoveries (R%) in M. sylvestris leaves related to different spiked concentrations; 400 and 10 ng g⁻¹ with relative s.d.

Compound	400 ng g−1	10 ng g−1
NA	102.3 ± 2.2	104.1 ± 3.3
ACL	78.2 ± 4.1	72.8 ± 1.9
AC	69.8 ± 1.2	63.3 ± 3.2
FL	66.9 ± 5.0	68.5 ± 4.8
PHE	88.3 ± 4.3	81.4 ± 1.7
AN	95.2 ± 3.2	94.3 ± 2.8
FA	88.1 ± 6.0	81.0 ± 5.8
PY	99.3 ± 2.8	96.9 ± 2.4
BaA	93.5 ± 2.9	102.9 ± 4.9
CHR	90.8 ± 2.1	89.4 ± 1.7
BbFA	92.8 ± 6.7	86.4 ± 3.8
BjFA	94.3 ± 6.2	86.0 ± 3.7
BkFA	93.6 ± 6.1	90.4 ± 3.1
BaP	100.5 ± 7.8	79.2 ± 3.2
IP	68.7 ± 2.9	65.0 ± 2.5

, R% is reported. The R% obtained at the two concentrations show good results and range from 66.9 ± 5.0 to 102.3 ± 2.2 and 63.3 ± 3.2 to 104.1 ± 3.3 for the 400 and 10 ng g⁻¹ concentrations, respectively. In conclusion, the proposed method is sensitive to the expected PAH concentrations in real samples.

3.2. Analytical Parameters

For the analytical method validation, the authors determined the following analytical parameters: linearity, the limits of detection and quantification (LODs, LOQs), reproducibility, and precision.

Table 5. Equations of the calibration curves, correlation coefficient (r), limits of detection (LODs) and quantification (LOQs), and repeatability/reproducibility as intra-day and inter-day values (RSD%) for each PAH.

Compound	Calibration Curve Equation	r	LOD (µg kg−1 FW)	LOQ (µg kg−1 FW)	Intra-Day (RSD%)	Inter-Day (RSD%)
NA	y = 0.415x + 0.012	0.995	0.45	1.20	13.3	15.3
ACL	y = 0.510x + 0.022	0.995	0.50	1.25	4.1	6.3
AC	y = 0.622x + 0.013	0.997	0.40	1.10	2.2	4.5
FL	y = 0.530x + 0.002	0.998	0.35	0.95	4.0	8.0
PHE	y = 0.533x + 0.018	0.998	0.40	1.00	4.3	8.2
AN	y = 0.497x + 0.020	0.996	0.35	0.95	6.4	9.4
FA	y = 0.618x + 0.009	0.997	0.30	0.85	2.3	10.8
PY	y = 0.610x + 0.005	0.998	0.35	1.00	3.1	7.0
BaA	y = 0.589x + 0.003	0.998	0.50	1.30	2.1	5.2
CHR	y = 0.376x + 0.002	0.998	0.35	1.20	5.1	7.4
BbFA	y = 0.573x + 0.001	0.999	1.25	2.80	4.2	7.6
BjFA	y = 0.553x + 0.002	0.999	1.20	2.35	4.9	8.0
BkFA	y = 0.527x + 0.002	0.999	1.00	1.90	5.2	7.7
BaP	y = 0.537x + 0.001	0.998	0.50	1.00	5.5	7.3
IP	y = 0.540x + 0.012	0.998	0.55	1.25	6.1	7.0

shows the equations of the calibration curves obtained by linear regression, the respective values of the correlation coefficient (r), values of LODs and LOQs, expressed as µg kg⁻¹ of fresh weight (F.W.), and the intra-day and inter-day precision, calculated by adding known PAH amounts to real samples. The linear dynamic range (LDR) goes from 0.05 to 5 µg mL⁻¹.

It can be seen that the values of the correlation coefficients are all greater than 0.995, which represents the lowest value for naphthalene and acenaphthylene. The calculation of LOD and LOQ was done following the European Commission’s Guidance Document on the Estimation of LOD and LOQ for Measurements in the Field of Contaminants in Feed and Food [29]. The LOD represents the lowest measured concentration from which the presence of the analyte can be inferred with reasonable statistical certainty and was calculated as equal to three times the standard deviation of the mean of the blank determinations (n > 20). The LOQ, on the other hand, represents the lowest analyte concentration that can be measured with reasonable statistical certainty and was calculated as equal to ten times the standard deviation of the mean of the blank determinations (n > 20). The minimum and maximum LOD values ranged from 0.30 for FA to 1.25 µg kg⁻¹ F.W. for BbFA, respectively. LOQ values ranged from 0.85 to 2.80 for FA and BbFA, respectively. The LOD and LOQ values are shown to be in line with other work in the literature using the same analytical instrumentation [18,19,20,21,22,23,24,25,26,27,28,29,30]. To highlight the strengths of this work, a comparison with previous studies [13,30,31,32,33] performed on PAHs in similar plant matrices in terms of LOD and LOQ is given in

Table 6. Comparison of LOD and LOQ data reported in the literature with those found in this study.

Compounds	LODs (µg kg−1 FW)	LOQs (µg kg−1 FW)	References
NA, ACL, AC, FL, PHE, AN, FA, PY, BaA, CHR, BbFA, BjFA, BkFA, BaP, IP, Retene, Benzo[g,h,i]perylene, Cyclopenta[c,d]pyrene, Picene, Coronene, Benzo[e]pyrene, Dibenz[a,h]anthracene, 3-Methylcholanthrene	0.2–0.7	0.8–2.8	[30]
NA, ACL, AC, FL, PHE, AN, FA, PY, BaA, CHR, BkFA, BaP, IP, Benzo[g,h,i]perylene, Dibenz[a,h]anthracene	0.07–1.29	N/A	[31]
PHE, AN, FL, PY, BaA, CHR, BbFA, BjFA, BkFA, BaP, IP, Benzo[g,h,i]perylene, Dibenz[a,h]anthracene	N/A	10	[13]
NA, ACL, AC, FL, PHE, AN, FA, PY, BaA, CHR, BkFA, BaP, IP, Benzo[g,h,i]perylene	0.08–1.32	N/A	[32]
BbFA, BkFA, BjFA, BaP, IP, CHR, 5-methylchrysene, Benzo[g,h,i]perylene, Dibenz[a,h]anthracene, Dibenz[a,l]pyrene	0.02–1.30	0.05–3.50	[33]
This study	0.30–1.25	0.85–2.80

N/A means not available.

.

The repeatability and reproducibility of the method were determined by conducting extractions on the actual samples at three different concentrations, calculating the inta-day and inter-day error as percent relative standard deviation (RSD%). The repeatability values ranged from 2.3 to 13.3 (RSD%), while the reproducibility values ranged from 4.5 to 15.3 (RSD%), showing good accuracy of the method. Figure 2a,b show the chromatograms in Total Ion Current (TIC) and Selected Ion Monitoring (SIM) of a 2 µg mL⁻¹PAHs standard solution, respectively. Figure 3a,b show chromatograms in TIC and SIM of a sample of M. sylvestris leaves taken at the Rotello site, without the addition of a PAH standard solution. Finally, in Figure 4a,b, the chromatograms of a sample of M. sylvestris leaves, taken in Rotello, with the addition of 500 µL of a 0.5 µg mL⁻¹ PAH standard solution, are shown.

3.3. Analysis of Leave Samples

The concentration of each PAH determined in the leave samples is shown in

Table 7. PAH concentration (µg kg⁻¹ FW) on M. sylvestris leaves in Rotello (rural), Rome, and Palermo (rural and background) with relative s.d. Data below the LOD are reported as below the PAH-specific detection limit value.

Compound	Rotello	Rome		Palermo
	Rural (A)	Urban (B)	Background (C)	Urban (D)	Background (E)
NA	<0.45	<0.45	<0.45	<0.45	<0.45
ACL	<0.50	<0.50	<0.50	<0.50	<0.50
AC	<0.40	<0.40	<0.40	<0.40	<0.40
FL	<0.35	<0.35	<0.35	<0.35	<0.35
PHE	<0.40	7.6 ± 0.1	<0.40	3.4 ± 0.3	15.0 ± 0.3
AN	<0.35	2.5 ± 0.0	<0.35	6.3 ± 0.5	8.0 ± 0.1
FA	<0.30	23.0 ± 0.9	<0.30	19.9 ± 1.8	32.2 ± 1.2
PY	<0.35	18.1 ± 0.7	<0.35	15.3 ± 1.0	20.4 ± 0.8
BaA	<0.50	3.8 ± 0.2	<0.50	2.4 ± 0.1	4.9 ± 0.2
CHR	<0.35	2.3 ± 0.0	<0.35	1.1 ± 0.0	2.0 ± 0.0
B[b+j+k]FA	<3.25	7.8 ± 0.9	<3.25	8.9 ± 0.4	11.6 ± 0.8
BaP	<0.50	1.2 ± 0.1	<0.50	1.0 ± 0.0	1.5 ± 0.0
IP	<0.55	3.9 ± 0.2	<0.55	2.9 ± 0.3	4.4 ± 0.6
Total	-	70.2 ± 3.1	-	61.2 ± 4.4	100.0 ± 4.0

.

The concentrations of each individual PAH determined on M. sylvestris leaves at the rural Rotello site (A) were all below the detection limits (LODs). In all sites, the quantities of NA, ACL, AC, and FL were lower than the LODs. This is due to the fact that these PAHs are characterized by a low molecular weight (LMW) and therefore tend to be volatile and not deposited through AD on the leaf surface. In the urban sites B and D, total PAH quantities of 70.2 ± 3.1 and 61.2 ± 4.4 µg kg⁻¹ FW were found for Rome and Palermo, respectively. The background site for the Rome area (C) shows no appreciable amount of PAHs; in fact, all molecules are below the LODs, confirming the hypothesis that at the urban site (B) there are multiple emission sources (mainly car traffic and domestic heating systems) and consequent contamination of M. sylvestris leaves trough AD. Several studies in the literature show that AD is one of the main mechanisms of PAH contamination for plants and also for other environmental matrices (water and soil) [34,35,36]. Findings were different in the background site for the Palermo area (E), with the highest PAH total amount of 100.0 ± 4.0 µg kg⁻¹ FW. In particular, the same PAHs were found in the M. sylvestris leaves collected in the urban site. The presence of the same PAHs in sites D and E is due to the fact that during any formation process, PAHs are always present as a class and never as individual compounds [9]. Furthermore, this suggests that there was the same source of contamination in the two sampling areas. The authors hypothesize that the high amounts of PAHs found on M. sylvestris leaves at the background site in Palermo are due to the numerous fires that occurred during the summer of 2023 and that involved several background areas outside the city (including the Bellolampo landfill), where the M. sylvestris sampling point is very close. In fact, the Regional Authority for Environmental Protection of Sicily (ARPA Sicilia), together with the National System for Environmental Protection (SNPA), through multiple monitoring of the territory have concluded that the increase in concentrations of specific pollutants (including benzene and suspended particulate matter PM₁₀ and PM_2.5) during the days of the fires are related to combustion products [37]. In other words, the high amounts of PAHs found at the E background site are directly correlated with the increased concentrations of benzene and PM₁₀-PM_2.5 recorded during the days of the fires. The authors also attempted to determine the sources of PAH contamination in the cities of Rome and Palermo using the diagnostic ratio approach [38,39] (

Table 8. Diagnostic ratio formula to estimate possible PAH sources in Rome and Palermo sampling sites.

Ratio formula	Range	Rome		Palermo
		Urban (B)	Background (C)	Urban (D)	Background (E)
AN/(AN + PHE)	>0.1 *	0.24	N/A	0.65	0.35
BaA/(BaA + CHR)	0.2–0.35 ** >0.35 *** <0.2 ****	0.63	N/A	0.68	0.71

* grass, wood and coal combustion, ** coal combustion, *** combustion sources, **** petrogenic, N/A means not available.

). It can be seen that the molecular ratios at all sites studied indicate that the likely source of contamination was a combustion process. In any case, in order to identify whether these sources were from waste combustion (for Palermo site), it is necessary to study the profile and levels of other micropollutants characteristic of such processes, including PCDDs, PCDFs, and PCBs.

4. Conclusions

The main objective of this study was to develop and optimize a rapid, simple, and reproducible analytical method for the qualitative and quantitative determination of PAHs in M. sylvestris leaves. The validated analytical method showed excellent recovery percentages (R%) of PAHs by using cyclohexane as the extraction solvent and ultrasonic extraction for 10 min. Furthermore, the method showed good analytical parameters, including the correlation coefficients (r) of the calibration curves and good LOD and LOQ values, suitable for dosing classes of micropollutants such as PAHs. The method was applied to samples of M. sylvestris leaves collected from different Italian locations, with different anthropic impacts. The results show that urban centers where there are multiple sources of contamination are not always more polluted. Emission sources, such as fires, can increase the polluting load of PAHs on M. sylvestris leaves due to the deposition of dust generated by the combustion process. The application of the proposed method using M. sylvestris plants could be of interest to monitor the contamination status of a territory by combining them with classic monitoring tools.