Bound PAHs in Indoor and Outdoor of Hotels in Urban and Suburban of Jinan , China : Concentrations , Sources , and Health Risk Impacts

The relationships between polycyclic aromatic hydrocarbons (PAHs) in PM2.5 in outdoor and indoor environments of hotels were examined in Jinan, China from January 6, 2016 to January 29, 2016. The mean concentrations of ∑PAHs for all sampling sites showed the following ascending order: suburban indoor (SUI, 39.58 ng m), first urban indoor near a busy traffic road (URI; 3 m, 63.26 ng m), suburban outdoor (SUO, 67.96 ng m), urban outdoor (URO, 105.30 ng m), and second urban indoor far away from the traffic roads (URI > 320 m, 115.63 ng m). The indoor/outdoor (I/O) ratios of URI and SUI were all less than 1, indicating that the PAHs were mainly infiltrated from the outdoor environment. At URI, 2-ring and some 3and 4-ring PAHs were mainly produced indoors due to cooking, whereas the 5–7-ring PAHs were mainly infiltrated from the outdoor environment. The diagnostic ratios and principal component analysis indicated that emissions from combustion of coal, biomass, diesel fuel and gasoline were the main sources of PAHs in the study area. The impacts of health risk assessment of PAHs suggested that the health risks in the outdoor environment were more severe than those in the indoor environment and the health risks in urban area were significantly higher than those in the suburban area in Jinan.


INTRODUCTION
Currently, considerable efforts have been made worldwide to clarify the effects of indoor air quality on human health (Yang et al., 2009;Hasheminassab et al., 2014;Zhu et al., 2015).People spend most of their times indoors (Lai et al., 2004;Tiwari et al., 2015); thus, indoor air pollutants, even at low concentrations, may have a considerable effect on human health due to the long exposure periods (Simoni et al., 2003).In addition, the concentration of some contaminants indoors exceeds that outdoors (Brooks et al., 1991).Many indoor air pollutants are associated with PM 2.5 , among which polycyclic aromatic hydrocarbons (PAHs) exhibit a more than 90% correlation with PM 2.5 (Albinet et al., 2008;Ringuet et al., 2012).PAHs are persistent pollutants in the atmosphere, and 16 PAHs have been classified as priority pollutants by the United States Environmental Protection Agency (USEPA) (Musa Bandowe et al., 2010).They are generated as byproducts of incomplete combustion or organic matter decomposition (Li et al., 1999;Albinet et al., 2007).PAHs are produced by natural sources (biological processes, forest fires, and volcano eruption) and human activities (emissions from motor vehicles and industries) (Albinet et al., 2007;Romagnoli et al., 2014).In addition, studies have found that indoor PAHs concentrations were affected by kitchen cooking, tobacco smoking, incense and candle burning, and domestic heating (Zhu and Wang, 2003;Gustafson et al., 2008;Castro et al., 2011;Orecchio, 2011).Furthermore, contaminants generated from outdoors can infiltrate to indoors through cracks or convect to indoors through open windows or the air conditioning system (Barraza et al., 2014).
The emissions from motor vehicles, one of the major PAHs sources, accounted for 12.8% of the annual global emissions of the 16 PAHs in 2007 (Shen et al., 2013).The PAHs emissions from vehicle exhaust may be transported to indoors and may become a significant outdoor source.The number of motor vehicles in China has rapidly increased in recent years.In 2015, China became one of the largest motor vehicle ownership countries in the world as the number of civilian vehicles exceeded 162 million (NBSC, 2016).Therefore, the effects of vehicle exhaust on indoor environment quality must be investigated.In addition, as of 2014, China had a floating population of approximately two million, which accounted for 18.6% of the total population (NBSC, 2015).Generally, the majority of them reside in hotels.However, previous studies of PAHs in indoor air have mainly been conducted in office, residential, and school buildings (Sangiorgi et al., 2013;Romagnoli et al., 2014;Hassanvand et al., 2015;Zhu et al., 2015).And few studies have been conducted in hotels.Jinan has the most congested evening traffic in China, even worse than Beijing (CATS, 2016).To more thoroughly understand the effect of motor vehicle exhaust on indoor PAHs, this study selected two hotels in an urban area that have a moderate occupancy of the hotel (≥ 60 rooms each day).The hotels were selected according to their distances from the heavy-traffic roads in the urban area.Moreover, one hotel in a suburban area was selected as far away from the urban area as possible with a small number of motor vehicles.
The main objectives of this study were to: (1) characterize and compare the indoor and outdoor PAHs concentrations at urban and suburban sites in Jinan, China; (2) examine the relationships between indoor and outdoor PAHs concentrations according to the I/O ratios; and (3) analyze the sources and health risks of PAHs in the indoor and outdoor environments of hotels.

Study Area
The sampling sites (Fig. 1) in the urban area were selected: the Green Tree Inn was selected as the first urban indoor (URI 1 ) site, which is located near a busy traffic road (Shanda Road, approximately 20000 vehicles pass by each day, approximately 3 m).Shandong University Hotel was selected as the second urban indoor (URI 2 ) site, which is located in the central campus of Shandong University campus.This hotel is far from the traffic roads (Shanda South Road, approximately 13000 vehicles pass by per day, > 320 m), compared with URI 1 , it has a commercial kitchen that can serve 150 people per meal.All the indoor samplers were placed on the fourth-floor room of the those hotels (approximately 12 m above the ground) and in the middle of the rooms at approximately 2 m away from the doors, windows, walls, and ventilation inlets.The windows were opened to allow ventilation before sampling and were kept closed during the sampling process.The urban outdoor sampling (URO) site is located at the west gate of Shandong University Center Campus (approximately 1.5 m above the ground).The sampling site is surrounded by commercial areas and many restaurants are also near Shanda Road (approximately 3 m) on which approximately 20000 vehicles pass by each day.
The sampling sites (Fig. 1) in suburban area were selected at the Seven Star Hotel, which is located in the Seven Star Scenic Area and nearly 50 km from the Jinan downtown area.This is a vacation spot with a total area of 20 km 2 and a forest coverage rate of 95%.The suburban indoor (SUI) sampling was conducted on the fourth-floor room of the hotel, approximately 12 m above the ground.Similarly, the windows of SUI were also opened before sampling and were kept closed during the sampling process.The suburban outdoor (SUO) sampling was conducted on the balcony (approximately 12 m above the ground) of the corresponding indoor room, which is far from the traffic road and approximately 150 m from the kitchen.

Sample Collection
Indoor and outdoor samples were simultaneously collected for 11.5 h (from 8:00 to 19:30 and from 20:00 to 7:30 the next day) from January 6, 2016 to January 29, 2016.PM 2.5 samples were collected in prebaked (600°C for 6 h) quartz fiber filters (88 mm) by using TH-16A Intelligent PM 2.5 samplers (Wuhan Tianhong Corporation, China, 100 L min -1 ) at URI 1 , URO, SUO, and SUI.The samples at URI 2 were collected on polytetrafluoroethylene (PTFE) filters (47mm) by using MiniVol Tactical Air Sampler (Airmetrics, USA, 5 L min -1 ).Before the sampling process was started, the airflow rates of the samplers were calibrated.When the sampling was completed, the filters were packed with prebaked (600°C for 6 h) aluminum foil sheets and stored in a freezer at -20°C until analysis.

Sample Extraction, Clean-up, and Analysis
The methods we used to extract and analyze PAHs were similar to those described by Li et al. (2014a) with minor modifications.All the PTFE filters and 14.73 cm 2 of the quartz fiber filters were used for experimental treatment.The samples were Soxhlet extracted with 150 mL of dichloromethane (DCM, J.T. Baker, USA) for 8 h.A rotary evaporator (RE201, Shanghai Bingyue Electronic Instrument Co., Ltd., China) with a water bath at 30 ± 1°C was used to condense each extract to approximately 1-2 mL.Then, the concentrates were purified using a silica gel and alumina column.Subsequently, the column was eluted with 20 mL of n-hexane (J.T. Baker, USA), which was discarded, followed by 70 mL of n-hexane/DCM (1:1, v/v, collected).Subsequently, the extract was concentrated to approximately 1-2 mL, with 10 mL of n-hexane added to replace the solvent, and concentrated to approximately 1-2 mL by using the rotary evaporator again.Next, the concentrate was condensed to 1 mL under nitrogen stream (high purity, 99.999%) with a water bath at 30 ± 1°C and then spiked with 100 ng of the internal standards (naphthalene-d 8 , anthracene-d 10 , pyrened 10 , and perylene-d 12 from AccuStandard, USA).
An Agilent 7890A gas chromatograph (GC) coupled with an Agilent 7001B mass selective detector (MS/MS) was used to quantify PAHs.An HP-5MS (30 m × 0.25 mm × 0.25 µm, Agilent, USA) capillary column was used to separate PAHs.The oven temperature of the GC was set at 60°C for 1 min, then increased to 150°C at a slope of 40 °C min -1 and maintained for 5 min, and increased to 300°C at a rate of 4 °C min -1 and maintained for 15 min.High-purity helium and nitrogen were used as the carrier gas and reagent gas, respectively.Samples were automatically injected at 1 µL in pulsed splitless mode.A total of 19 PAHs were quantified with the MS/MS in selected ion monitoring (SIM) mode through electron impact ionization (Albinet et al., 2006;Li et al., 2014a, b).

Quality Control
Pesticide-grade solvents of the analytical reagent nhexane and DCM were used for the experimental pretreatment process.All pieces of glassware used during pretreatment were rinsed with chromic acid and baked at 100°C and oven dried.For each batch of 10 samples, a calibration solution of PAHs was analyzed to ensure the stability of instrument performance.The contents of PAHs in the reagent and procedural blank filter samples were negligible.Internal calibration method was employed to quantify the data, which based on five-point calibration curves and the regression coefficient (R 2 ) value for each individual PAHs was higher than 0.99.To obtain the surrogate recoveries, one-third of the samples were randomly selected and spiked with 100 ng of surrogates (mixture of acenaphthene-d 10 and chrysene-d 12 , AccuStandard, USA) before Soxhlet extraction.The surrogate recoveries were 85% ± 20% for acenaphthene-d 10 and 79% ± 17% for chrysene-d 12 .

PAHs Concentrations in Ambient Air
The mean concentrations and standard deviations of individual PAH at all sampling sites are presented in Table 1.The mean concentrations of ∑PAHs at all sampling sites showed the following trend: SUI < URI 1 < SUO < URO < URI 2 .The indoor and outdoor concentrations of 19 PAHs in the urban area in this study were comparable to those in the study by Zhu et al. (2015), who reported ∑PAHs concentrations of 111.07 and 140.34 ng m -3 in indoor and outdoor environments, respectively, in autumn of Jinan.But the concentration was lower than that in Beijing, which were 187.3 ng m -3 and 387.0 ng m -3 in indoor and outdoor environments, respectively (Han et al., 2016).Moreover, the ∑PAHs in this study area was determined to be lower than the levels of 16 PM 2.5 -bound PAHs in Taiyuan and in Beijing with concentrations of 420.8 ng m -3 (Li et al., 2016a) and 407.6 ng m -3 (Wang et al., 2008) in winter, respectively.However, the PAHs concentrations in the present study sites were usually higher than those in the southern region of China.Such as, the total 18 PAHs concentration in fine particles in Nanjing was 50.6 ng m -3 during pre-Spring Festival (Kong et al., 2015); the concentration of PM 2.5associated 15 ∑PAHs was 22.54 ng m -3 during the autumn and winter in Nanchang (Liu et al., 2016); and 16 PAHs concentrations in PM 2.5 was 23.7 ng m -3 at Guangzhou with a heating source (Li et al., 2006).The lower PAHs concentration in South China compared with the present study sites may be attributed to the following reasons: no heating demand in winter, high precipitation frequency, and large precipitation intensity in South China (Li et al., 2014b).In addition, the PAHs concentrations in this study were higher than those in many areas of developed countries (Gigliotti et al., 2005;Motelay-Massei et al., 2005;Albinet et al., 2007;Romagnoli et al., 2014).The 16 USEPA PAHs can be classified on the basis of the number of aromatic rings as follows: low molecular weight [LMW, 2-3-ring PAHs; 2-ring (Nap), 3-ring (Ace, Acy, Flo, Phe, and Ant)]; middle molecular weight [MMW, 4-ring PAHs (Flt, PYR, BaA, and Chr)]; and high molecular weight [HMW, 5-ring (BbF, BkF, BaP, and DBA) and 6ring (IcdP and BghiP)] ( Kim et al., 2013;Cheruyiot et al., 2015;Lu et al., 2015).A higher concentration of LMW PAHs indicates emissions from noncombustible petroleum products; however, HMW PAHs are significantly correlated with pyrolysis products generated from the combustion of fossil fuel (Hassanien and Abdel-Latif, 2008).The ringwise distributions of PAHs at all sites are presented in Fig. 2. The proportions of the LMW and MMW PAHs in the total PAHs were similar at all study sites.LMW PAHs accounted for 21%-49% of ∑PAHs; MMW PAHs accounted for 29%-39% of the total PAHs, whereas HMW species were dominant at each sampling site and accounted for 36%-49% of the total PAHs.This may be due to the fact that HMW PAHs have a relatively large surface area and are more likely to be adsorbed on the surface of the particles (Kong et al., 2010;Hassanvand et al., 2015).Moreover, HMW PAHs are less volatile than LMW PAHs, which mainly exist in the gas phase (Possanzini et al., 2004).The higher proportion of HMW PAHs implies that fuel combustion contributed more to PAHs in the study area.

Daytime and Nighttime Variations and Indoor and Outdoor Concentration Relationships of PAHs
The daytime and nighttime PAHs concentrations in the outdoor and indoor environments of the urban and suburban areas are presented in Fig. 3.At site URO, the concentrations of PAHs during daytime were higher than those during nighttime.In particular, the concentration of 4-6 ring PAHs exhibited obvious differences between daytime and nighttime.This may be attributable to the much heavier traffic during daytime.However, at SUO, the PAHs concentrations during daytime were lower than those during nighttime (Fig. 3A).These results can be explained as follows.In Jinan, the suburban areas have fewer motor vehicles compared with the urban areas, which leads to less PAHs emissions generated by motor vehicles.In addition, the surface inversion layer is stable during nighttime in winter, which is not conducive to the spread and dilution of pollutants.The daytime and nighttime concentration variations at sites URI 1 and SUI were consistent with the corresponding outdoor concentrations, indicating that the outdoor source has a significant effect on the indoor environment.The PAHs concentrations were higher during nighttime at URI 2 (Fig. 3B).This may be related to the indoor kitchen cooking at this site.
The I/O ratios of the 19 PAHs are presented in Fig. 4. The I/O ratios of URI 1 and SUI were all less than 1, indicating that the PAHs were mainly infiltrated from the outdoor ambient air.At URI 2 , the I/O ratios of 2-ring and some 3-4-ring PAHs were greater than 1, whereas those of 5-7-ring PAHs were less than 1, signifying that 2-ring and some 3-4-ring PAHs were mainly produced from the indoor sources, whereas the 5-7-ring PAHs were mainly infiltrated from the outdoor ambient air.The higher 3-ring PAHs concentrations at this site can be attributed to the cooking activities and oil fumes (Zhu and Wang, 2003).In addition, at site URI 2 , naphthalene has a higher concentration than outdoor, mostly deriving from the evaporation of mothball including a great deal of naphthalene, used as an indoor insecticide to prevent clothes against moth.

Source Apportionment of PAHs Diagnostic ratios
The isomer ratio is one of the effective methods to identify the possible origins of PAHs.When isomer pairs are mixed with natural particulate matter, they are diluted to a similar degree.In addition, they could exhibit a similar distribution pattern among other phases because of their comparable thermodynamic partitioning and kinetic mass transfer coefficients (Dickhut et al., 2000;Zhu et al., 2015).The isomer ratios Flt/(Flt + PYR), Ant/(Ant + Phe), BaA/(BaA + Chr) for particulate PAHs were usually employed to determine the possible sources of PAHs (Dickhut et al., 2000;Yunker et al., 2002;Liu et al., 2007;Zhu et al., 2015).The calculated isomer ratios are  presented in Fig. 5. Flt/(Flt + PYR) values greater than 0.5 indicate that the source of PAHs is combustion (coal or biomass burning), and those between 0.4 and 0.5 indicate that the source is petroleum combustion, whereas Flt/(Flt + PYR) values less than 0.4 imply oil sources (unburned petroleum) (Budzinski et al., 1997;Yunker et al., 2002;Liu et al., 2007).In this study, the Flt/(Flt + PYR) ratios were greater than 0.5 at all sites.This result indicated that PAHs mainly originated from coal or biomass combustion during winter in the study area.
The isomer ratio of Ant/(Ant + Phe) is commonly used to distinguish between combustion and petroleum sources (Soclo et al., 2000).A value less than 0.10 usually suggests a petroleum source, whereas a value greater than 0.10 implies a dominance of combustion (Budzinski et al., 1997;Yunker et al., 2002).Fig. 5 shows that the Ant/(Ant + Phe) values were  in the range from 0.06 to 0.13, indicating mixed contributions of combustion and petroleum sources dominated in the region.The BaA/(BaA + Chr) ratios were greater than 0.35 at all sites, indicating a strong contribution from fuel combustion (Soclo et al., 2000).This finding is consistent with the report by Zhu et al. (2015) in Jinan.

Principal Component Analysis
Principal component analysis (PCA) was employed to analyze the independent source tracers (Li et al., 2006;Hong et al., 2007;Li et al., 2016b) for 19 PAHs, and the factors are listed in Table 2. Two factors accounted for 93.5% of the total variance of the data.Factor 1, which constituted 51.1% of the total variation, showed high loading for PAHs of IcdP, BbF, DBA, BghiP, Cor, BaP, BeP, BkF, PYR, Flt, and Chr.Among these compounds, BbF, BaP, PYR, BkF, Flt, and Chr are the tracers of oil and coal combustion (Khalili et al., 1995), which is consistent with the analysis of the characteristic ratio.In addition, IcdP, BbF, BghiP, and BaP were considered as the typical markers of vehicle emission (Miguel and Pereira, 1989;Harrison et al., 1996).Therefore, the results suggested that factor 1 was mostly related to oil and coal combustion and vehicle emissions.Factor 2, which constituted 42.4% of the total variation, showed high loading for PAHs of lower molecular weight (Flo, Ace, Bip, Nap, Phe, Acy, BaA, and Ant).Among those compounds, Ace, Phe, and Ant are mainly originated from coal burning, whereas Acy is mainly originated from diesel and petrol vehicle exhaust emissions.The presence of these compounds in atmospheric particulate matter, as characterized by Factor 2, might be explained by the motor vehicle emissions with unburned diesel fuel and gasoline.These findings were supported by Li et al. (2016b), who reported that oil and coal combustion and diesel and gasoline emissions were the main sources of polycyclic aromatic hydrocarbons in eastern China.

Risk Assessment of PAHs Pollution
The BaP equivalent concentration (BaPeq) is typically employed to assess the carcinogenic potency of PAHs (Zhu et al., 2015).The BaPeq of each PAHs was calculated by using the toxic equivalent factor (TEF), which is defined as the relative cancer potency of a given species (Nisbet and LaGoy, 1992).The concentrations of carcinogenic PAHs were estimated as follows (Pongpiachan, 2016): where BaPeq i is the BaP equivalent concentration of the ith target PAH; PAH i is the concentration of the ith target PAH, TEF i is the toxic equivalent factor for the ith target PAH (the values of TEF i are listed in Table 3), and TEQ is the toxic equivalent of the target compound.
As presented in Table 3, the TEQ values of the 16 USEPA PAHs were in the range of 415.68-1088.24× 10 -2 ng m -3 at all sites, which are significantly higher than the WHO standard (1 ng m -3 , WHO, 1987).In addition, the highest TEQ value was observed at URO, which exceeded the national standard of China (10 ng m -3 ).Additionally, the ∑TEF -PAHs values were employed to obtain the lifetime lung cancer risk (LLCR) in the indoor and outdoor environments.One unit of risk for a 70-year lifetime of PAHs exposure is 8.70 × 10 -5 (ng m -3 ) -1 (WHO, 2000).The values of the LLCR from the exposure to PAHs at URO (9.47 × 10 -4 ), URI 1 (8.39 × 10 -4 ) and URI 2 (8.18 × 10 -4 ) were all determined to exceed the health guideline (10 -5 ) (Chithra and Shiva Nagendra, 2013).The values of LLCR from the exposure to PAHs at SUO (6.54 × 10 -4 ) and SUI (3.62 × 10 -4 ) were also exceed the health guideline but significantly lower than the values in urban sites.These results indicated that the health risks in urban areas were higher than those in the suburban areas in Jinan.In addition, the LLCR in indoor environment was lower than that in the outdoor environment in both urban and suburban area in Jinan.And the LLCR at URI 2 was lower than URI 1 , which indicated that the distance from the traffic source may have an effect on the health impacts.

SUMMARY
The mean concentrations of ∑PAHs were highest at URI 2 (115.63 ng m -3 ), followed by URO (105.30ng m -3 ), and the concentrations of ∑PAHs were lowest at SUI (39.58 ng m -3 ).At site URO, the PAHs had a higher concentration during daytime.However, the PAHs concentrations during daytime were lower than those during nighttime at SUO.
The I/O at URI 1 and SUI were all less than 1 suggesting those PAHs were mainly produced from outdoor environment.At URI 2 , the I/O values of 2-ring and some 3-4-ring PAHs were larger than 1, suggesting that they mainly generated from the indoor atmosphere, whereas the emission sources of 5-7-ring PAHs were mainly infiltrated from outdoor sources.
The ratio of Flt/(Flt + PYR), Ant/(Ant + Phe), BaA/(BaA + Chr) and PCA results all clearly declared that the mixed sources that coal or biomass combustion and diesel and gasoline emissions were the main sources of PAHs in the study region.
The results of health risk assessment of PAHs pollution has significance for public health.The LLCR in the urban areas were significantly higher than those in the suburban in Jinan.In addition, the LLCR in outdoor environment were more severe than those in the indoor environment.This indicated that urban area and outdoor environment in Jinan have relatively high health risk.

Fig. 1 .
Fig. 1.The locations of sampling sites in Jinan, China.

Fig. 4 .
Fig. 4. The indoor-to-outdoor concentration ratios of PAHs at the urban and suburban sites.

Table 1 .
Mean PAHs concentrations (ng m -3 ) and standard deviations at all sampling sites.

Table 2 .
Factorial weights matrix at all sites of urban and suburban in Jinan obtained from principal component analysis.