Analysis of PAHs Associated with Particulate Matter PM 2 . 5 in Two Places at the City of Cuernavaca , Morelos , México

This study was carried out between January and February 2013, at two sites in the city of Cuernavaca, México, using low-volume equipment. Fifteen Polycyclic aromatic hydrocarbons (PAHs), were identified by gas chromatography coupled with mass spectrometry. The total average concentration observed for PAHs was 24.0 ng·m−3, with the high molecular weight compounds being the most abundant. The estimated equivalent concentration for Benzo (a) P (BaPE) was 4.05 ng·m−3. Diagnostic ratios together with the principal components analysis (PCA) allowed for establishing coal burning and vehicle OPEN ACCESS Atmosphere 2015, 6 1260 emissions as being the main sources of these compounds in the area. The PAHs used to calculate this index account for 51% of the 15 PAHs identified, which probably involves a risk to the exposed population.


Introduction
Polycyclic aromatic hydrocarbons (PAHs) are one of the most studied families of organic compounds present in the atmosphere, given their proven negative effects on health and persistence in the environment.Different PAHs, specifically those with high molecular weight, which predominate in the particulate phase of atmospheric aerosols, have shown to be carcinogenic [1].
It is also known that PAHs accumulated in particles are mainly associated with fine particulate matter (PM2.5).These fine particles are better able to penetrate the respiratory system, which increases their potential health effects [2,3].
PAHs are produced by incomplete combustion and pyrolysis of fossil fuels such as oil and coal, and from other organic materials from natural and anthropogenic sources [4][5][6], including vehicle emissions, wood burning, waste incineration, coke production, and metal production [7][8][9].
Their concentrations in ambient air are lower in rural than in urban and suburban areas.Moreover, they can have different profiles depending on the geographical location, the type of sources and the dominant atmospheric characteristics [10].To identify possible sources of these compounds, diagnostic ratios were used.However, these should be used with caution, because sometimes the discrimination between these compounds is difficult, taking into account the reactivity of some of the PAHs with species such as NOx, O3, etc. [11].
For the city of Cuernavaca, there is little information on the chemical characterization of PM2.5, specifically in determining PAHs.Therefore, the present study has as its main objective to determine ambient levels of these compounds and their possible sources.The results will serve as a basis for mitigation programs of air pollution and/or generation of regulatory standards for the environmental authorities responsible for monitoring of air quality.

Wind Trajectories
During the study period, the winds came mainly from the southeast of the state (66%), which probably suggests that part of air pollutants identified in the Autonomous University of State of Morelos (in the north) had its origin in the industrial zone of the city (located to the south) and the downtown area of the city.However, other possible sources could be the emissions from the Popocatepetl volcano, located to the south east of the State (Figure 1).

Concentration of PAHs
PAHs with 2 and 3 aromatic rings (Acen, Ace, Flu, Phe and Ant) showed the lowest levels of concentration at both sites, whereas those with 4-6 rings (Flt, Pyr, BaA, Chr, BbF, BkF, BaP, IP, DBahA, BghiP) were the most abundant.This suggests that the combustion of gasoline constitutes one of the main sources of these compounds in the study area, which is consistent with that observed by Zielinska et al. 2004 [12].
At the downtown site the average total PAHs concentration was 22.57 ng•m −3 (Σ15 PAHs), while that at the CIQ site was 25.43 ng•m −3 (Σ15 PAHs), indicating that there were no statistically significant differences in the concentration of this family of compounds between sites (p < 0.05).Such behavior suggests relatively stable atmospheric conditions or similar emission sources in the study area (Table 1).Also the proximity of the University Campus to the México-Cuernavaca highway, probably is affecting the concentration of PAHs in this area, as well as transportation from the southern part of the city, such as was seen in the wind field.

Diagnostic Ratios
Given the variability in the concentration of PAHs emitted into the atmosphere, often diagnostic ratios between them are used in order to estimate the possible sources [20].However, they should be used with caution, if one considers, for example, reactivity with some of the PAH species such as NOx, O3, etc. [9].
Diagnostic ratios used in this study, were calculated and reported by various authors, which established ranges or values indicating possible sources of origin of polycyclic aromatic hydrocarbons (Table 2).The ratio Flt/(Flt + Pyr) was between 0.40 and 0.50, suggesting that the study area is influenced by emissions from gasoline and diesel [21,22].Meanwhile the IP/(IP + BghiP) ratio obtained in this study was 0.5.Some previous studies that were done in other places have reported values between 0.18 and 0.40 corresponding to vehicle emissions, 0.56 corresponding to coal burning and 0.62 to wood combustion [21][22][23][24][25], indicating that in the study area coal burning is an important source.The Phe/(Ant + Phe) ratio was also estimated, giving a value of 0.52.This value is associated with vehicle emissions, specifically gasoline [23,26].Meanwhile the IP/BghiP (1.05) ratio indicated coal burning as a major source in this zone [27,28].These results are congruent with the dimensions, anthropogenic activities and population density of the Cuernavaca city.Recently, Ortíz-Hernández et al. [29] performed an emissions inventory, this study suggests vehicle emissions, coal burning and forest activities as the principal sources of greenhouse gases in this region.

Principal Component Analysis
A principal component analysis (PCA) with varimax rotation was performed to determine the possible sources of these compounds in the city of Cuernavaca.The principal components (PC) were selected with eigenvalues greater than 1.0.In total 3 PC were extracted.The PC1 explained 75.6% of the total variance.In this component the compounds loads were similar.Although the BaA, Chr and BaP stand out slightly, these compounds are considered as markers of pyrogenic sources (combustion of oil and coal) [30].While the PC2 represented 8.5% of the remaining variance, the compounds with the greatest burden in this component were BkF, IP and BghiP, which are associated with vehicle emissions [30,31].Meanwhile, the PC3 explained 7.0%, and the greatest burden was provided by Ace and Flu, considered as indicators of coke production (Table 3).
The results show that the study area is being specifically affected by emissions from gasoline and diesel.
The combined results of the PCA and diagnostic ratios suggests that vehicle emissions and coal combustion are probably the main sources of PAHs associated with the PM2.5 in the study area.

Evaluation of Health Risk
The value obtained for the BaPE in this study was 4.05 ng•m −3 .This value is lower than that reported for Nanging, China, (7.1 ng•m −3 ) [6] and Zonguldak, Turkey (14.1 ng•m −3 ) [32].However, it was higher than that found in Hamilton, Canada (0.84 ng•m −3 ) [33] and in Florence, Italy (0.92 ng•m −3 ) [34].The compounds used to calculate BaPE (BaA, BaP, BbF, BkF, IP, DBah) represent 51% of the PAHs identified in Cuernavaca, which constitutes a risk factor for the exposed population, taking into account the high degree of penetration in the respiratory system that have the PM2.5 particles.

Sampling Sites
The study was conducted at two sites in the city of Cuernavaca.The first one is located in the downtown of the city (Government Building, GB), which is characterized by high commercial and vehicular activity.The second is located north of the city, at the Autonomous University of State of Morelos Campus ((Center of Chemical Research, CIQ), for its acronym in Spanish).This location is characterized as being surrounded by mountains, having a great variety of vegetation, and being influenced by the Mexico-Cuernavaca highway.The city has a total area of 207.5 km 2 and a population of 338,620 inhabitants.During the study period, the average temperature was 23.1 °C (minimum 9.6 ºC, maximum 26.7 ºC), the relative humidity was 36.2% and the average wind speed was 5.3 m•s −1 (Figure 2).The particles were collected on quartz filters of 47 mm diameter and in a pore size of 2.0 µm, baked to 180 ºC for at least 24 h to remove adsorbed organics, after which they were transferred to a chamber with relative humidity of <40% at 20-23 °C for another 24 h for conditioning.Afterward, the filters with particles were equilibrated in the chamber for an additional 24 h.Low volume equipment (Mini-Vol) were used at a constant flow of 5.0 L•min −1 , equipped with impactors for PM10 and PM2.5.Sampling was carried out in periods of 24 h (12:00-12:00), from January 25 to 28 February 2013, every two days.In total, 11 samples were collected for each site.After sampling, the filters were refrigerated (4 °C) until extraction.

Wind Trajectories
To determine the behavior of winds in the study area, HYSPLIT4 model trajectories obtained from the NOAA (National Oceanic and Atmospheric Administration) were used.The trajectories were made for the entire study period [35].

Extraction of Organic Matter
Sampled filters were extracted using 10 mL of dichloromethane (Burdick & Jackson, HPLC grade) in an ultrasound bath (Branson 3210) for 30 min.This procedure was performed twice.The Erlenmeyer flasks in which the filters were introduced, were fitted with cooled condensers with water at 10 °C.The extracts were filtered through a polytetrafluoroethylene (PTFE) membrane (0.22 µm, Millipore), and then concentrated on a rotary evaporator (Buchi R-3) to about 1.0 mL and stored under refrigeration until analysis.

Quality Control of the Analytical Method
Field blanks (glass fiber filters) were performed once a week.Field blanks were stored in Petri dishes and returned to the laboratory until analysis.Further, blanks of laboratory material were used to discard the presence of analytes of interest during each one of the stages.The efficiency of the extraction methodology was evaluated by the enrichment of sampled filters with airborne particles with a solution containing 16 PAHs: Acenaphthylene (Acen), Acenaphthene (Ace), Fluorene (Flu), Phenanthrene (Phe), Anthracene (Ant), Fluoranthene (Flt), Pyrene (Pyr), Benzo[a]Anthracene (BaA), Chrysene (Chr), Benzo[b]Fluoranthene (BbF), Benzo[k]Fluoranthene (BkF), Benzo[a]Pyrene (BaP), Indeno[1,2,3-cd]Pyrene (IP), Dibenzo[a,h]Anthracene (DBahA) and Benzo[g,h,i]Perylene (BghiP), at a concentration of 250 parts per billion (ppb).The amount recovered for each compound was compared to the amount added to calculate the recoveries.Once the percent recovery was determined on the enriched filters, the repeatability was calculated by estimating the coefficient of variation (% CV).The recoveries ranged from 60% to 90%, for Fluorene and IP, respectively.The repeatability in terms of % CV was less than 10%, which is consistent with the variation (30%) set by the United States Environmental Protection Agency [36], for these samples.
Instrumental detection limits were performed based on a weight regression, proposed by Miller and Miller 2002 [37]; these ranged between 16.71 and 61.70 ng•mL −1 for BghiP and BaP, respectively (Table 4).The concentrations of the compounds in ambient air were corrected using the recovery percentages.

Instrumental Analysis
The extracts obtained were concentrated under a gentle stream of nitrogen to near dryness and subsequently resuspended with 150 µL of dichloromethane in an insert injection.To this volume, 50 µL of a solution containing 6 PAH-d (internal standard) were added to a final volume of 200 µL (naphthalene-d8, acenaphthene-d10, phenanthrene-d10, pyrene-d10, chrysene-d12 and perylene-d12) (Chemservice West Chester, PA), at a final concentration of 2500 ppb of internal standard.
The chromatographic analysis was performed on an Agilent Technologies, Model 6890 gas chromatograph (GC) coupled to a mass spectrometer (MS) 5973N with quadrupole mass filter.The separation of the compounds was carried out on a capillary column (J & W Scientific, USA) with an internal diameter of 0.25 mm with 5% phenyl stationary phase and 95% dimethyl polysiloxane and a film thickness of 0.25 µm.The oven temperature program was as follows: 60 °C for 10 min at a gradient of 5 °C per minute to 300 °C for 10 min.The injector temperature was 300 °C in the splitless mode (purge time 30 sec) and the injection volume of 2.0 µL.Helium was used as carrier gas at a flow of 1.0 mL•min −1 .The MS was operated in the electron impact (70 eV) mode and the temperature of the ion source and the quadrupole filter were 230 °C and 150 °C, respectively.

Figure 2 .
Figure 2. Sampling sites at the city of Cuernavaca, Morelos.

Table 1 .
Average concentration of PAHs associated with particulate matter PM2.5 in two places of the city of Cuernavaca, Morelos during the winter season of 2013 (ng•m −3 ).

Table 3 .
Principal Component Analysis for Cuernavaca.
LOD: limit of detection.