Environment International 96 (2016) 65–74

Contents lists available at ScienceDirect

Environment International

j ourna l homepage: www.e lsev ie r .com/ locate /env int

Measurement and health risk assessment of PM2.5, flame retardants,carbonyls and black carbon in indoor and outdoor air in kindergartens inHong Kong

Wen-Jing Deng a,⁎, Hai-Long Zheng b, Anita K.Y. Tsui c, Xun-Wen Chen d

a Department of Science and Environmental Studies, The Education University of Hong Kong, Tai Po, N.T., Hong Kong, Chinab Department of Geography and Resource Management, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong, Chinac Department of Early Childhood Education, The Hong Kong Institute of Education, Tai Po, N.T., Hong Kong, Chinad Department of Civil and Environmental Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China

⁎ Corresponding author.E-mail address: wdeng@eduhk.hk (W.-J. Deng).

http://dx.doi.org/10.1016/j.envint.2016.08.0130160-4120/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 16 May 2016Received in revised form 19 July 2016Accepted 19 August 2016Available online xxxx

Indoor air pollution is closely related to children's health. Polybrominated diphenyl ethers (PBDEs) anddechlorane plus (DP) transmitted through indoor PM2.5 and dust, alongwith carbonyl compounds and black car-bon (BC) aerosol were analysed in five Hong Kong kindergartens. The results showed that 60% of the medianPM2.5 levels (1.3 × 101 to 2.9 × 101 μg/m3 for indoor; 9.5 to 8.8 × 101 μg/m3 for outdoor) in the five kindergartenswere higher than the guidelines set by theWorld Health Organization (2.5 × 101 μg/m3). Indoor PM2.5 mass con-centrations were correlated with outdoor PM2.5 in four of the kindergartens. The PBDEs (0.10–0.64 ng/m3 inPM2.5; 0.30–2.0 × 102 ng/g in dust) and DP (0.05–0.10 ng/m3 in PM2.5; 1.3–8.7 ng/g in dust) were detected in100% of the PM2.5 and dust samples. Fire retardant levels in the air were not correlated with the levels of dustin this study. The median BC concentrations varied by N7-fold from 8.8 × 102 ng/m−3 to 6.7 × 103 ng/m−3

and cooking events might have caused BC concentrations to rise both indoors and outdoors. The total concentra-tions of 16 carbonyls ranged from 4.7 × 101 μg/m3 to 9.3 × 101 μg/m3 indoors and from 1.9 × 101 μg/m3 to4.3 × 101 μg/m3 outdoors, whilst formaldehyde was themost abundant air carbonyl. Indoor carbonyl concentra-tions were correlated with outdoor carbonyls in three kindergartens. The health risk assessment showed thathazard indexes (HIs) HIs of non-cancer risks from PBDEs and DPs were all lower than 0.08, whilst non-cancerHIs of carbonyl compounds ranged from 0.77 to 1.85 indoors and from 0.50 to 0.97 outdoors. The human intakeof PBDEs and DP through inhalation of PM2.5 accounted for 78% to 92% of the total intake. The cancer hazard quo-tients (HQs) of formaldehyde ranged from 4.5E−05 to 2.1E−04 indoors and from 1.9E−05 to 6.2E−05 out-doors. In general, the indoor air pollution in the five Hong Kong kindergartens might present adverse effects tochildren, although different schools showed distinct pollution levels, so indoor air quality might be improvedthrough artificial measures. The data will be useful to developing a feasible management protocol for indoorenvironments.

© 2016 Elsevier Ltd. All rights reserved.

Keywords:Indoor air pollutionPBDEsDechlorane plusParticulate matterChildren health

1. Introduction

Indoor air pollution (IAP) is closely related to human health becausemost urban citizens spend a large amount of their time in indoor envi-ronments (Chithra and Shiva Nagendra, 2012; WHO, 2006). Studieshave increasingly found that exposure to IAP may lead to asthma, car-diopulmonary pathologies, neurological impairments, dysfunctionalimmune systems, reproductive disorders, and even childhood leukae-mia (Sohn et al., 2012; Yang et al., 2009). The World Health Organiza-tion (WHO) has revealed that IAP is responsible for 2.7% of the globalburden of disease (WHO, 2010).

The variety of pollutants detected in indoor air, such as halogenatedcompounds and volatile organic compounds (VOCs) (Harrad et al.,2006; Hwang et al., 2008; Takigami et al., 2009), may be 2–5 times,and occasionally N100 times, higher than the levels of pollutants foundoutdoors (Kohler et al., 2005). Polybrominated diphenyl ethers(PBDEs), which belong to halogenated compounds, can be emittedfrom the polyurethane foam in furniture, mattresses, carpet pads, andelectronic equipment (ATSDR, 2004). They are believed to interferewith thyroid hormone balance and contributes to neurodevelopmentaldeficiencies (Dingemans et al., 2011; Eskenazi et al., 2013; Gascon etal., 2011; Lee et al., 2010). DP has a high production volume but wasonly recently found in the environment. Hoh et al. (2006) first reportedon the environmental occurrence of DPwithin theNorth American GreatLakes Basin, where it is manufactured by OxyChem in Niagara Falls, New

66 W.-J. Deng et al. / Environment International 96 (2016) 65–74

York. Since then, many research studies have been carried out in differ-ent countries involving different environmental matrixes including soil(Wang et al., 2010), air (Venier and Hites, 2008) and indoor dust (Zhuet al., 2007; Ren et al., 2010), which had led to the prospect that DP is aworldwide contaminant. In addition, DP has the typical characteristicsof a persistent organic pollutant: high lipophilicity, resistance to photo-and biodegradation, and accumulation in fish. DP was found to be re-leased from electronic products in the indoor environment (Ren et al.,2009). DP is persistent, bioaccumulative, and subject to long-range trans-port (Qiu andHites, 2008;Möller et al., 2010). However,more research isrequired to better quantify the emissions, exposures, and toxicologicaleffects of DP and its analogs in the environment. In particular, there is aneed to obtain more monitoring, and toxicity information, particularlyto kindergarten children. VOCs, which are emitted from furniture, deco-rations and cleaning products (Dodson et al., 2007; Massolo et al., 2010;Wu et al., 2011), are related to allergic effects, lung cancer and childhoodleukaemia (Loh et al., 2007; Sax et al., 2006). Among the various air pol-lutants, the undesirable health effects of particulate matter (PM) haveraised considerable concern over indoor air quality (Lin et al., 2002;Pope andDockery, 2006) because the presence of indoor PM2.5 (particleswith an aerodynamic diameter b2.5 μm) has been strongly associatedwith cardiopulmonary disease and lung cancer (Gemenetzis et al., 2006).

IAP in schools may be even more serious than in other areas, becausechildren spend up to eight or more hours per day in school (Branco et al.,2014b). Further, children are more sensitive to the effects of air pollutionthan adults,mainly due to their underdeveloped respiratory, immune andendocrine systems, comparatively higher amount of air inhalation relativeto their body size (Salvi, 2007; Schwartz, 2004; Sousa et al., 2012). Thus,they are more likely to suffer from the risk of exposure to IAP.

Paying attention to the effect of IAP on the health of school childrenis one of the key focuses of the Healthy Environments for Children Alli-ance (WHO, 2003). In addition, the significance of IAP in schools hasbeen highlighted by many studies worldwide (Godwin andBatterman, 2007; Hodgson et al., 2004; Yang et al., 2009; Zhang et al.,2006). For example, PBDE compounds have been detected in 100% ofthe dust samples in California's early childhood education environ-ments (Bradman et al., 2014), where a correlation between pollutantconcentrations and the onset of health problems in schoolchildren hasbeen found (Bono et al., 2015; Cartieaux et al., 2011). Further, a highprevalence of asthma has been reported in children attending schoolin classrooms with PM2.5 levels higher than 17.5 μg/m3 (Annesi-Maesano et al., 2012).

Most studies have focused on one specific group of pollutants or onthermal conditions (Branco et al., 2014a; Branco et al., 2015;Carreiro-Martins et al., 2014; Fonseca et al., 2014; Rivas et al., 2014).Therefore, measuring awide range of health-relevant pollutants has be-come very important for assessing the indoor air pollution levels inschools.

The size distribution of particle-bound fire retardants (FRs) is a crit-ical factor governing their fate in the air. It has also been associatedwithgreater health risk to children because smaller particles can penetratedeeper into the lungs. Yet, few studies have analysed the health riskfor kindergarten children exposed to dechlorane plus (DP), which is ahalogenated FR usedworldwide, and PBDEs transmitted through indoorPM2.5 and dust. The present study evaluates the concentrations of thesepollutants and their relationships, and additionally explores the healthrisks to children from FRs (DP, PBDEs) through indoor PM2.5 and dust,along with carbonyl compounds and black carbon aerosol, found inour kindergartens and children's homes.

2. Methods

2.1. Description of sampling site

This study was carried out in Hong Kong, which is one of the mostdeveloped and urbanised cities in China. Between May 22 and July 23,

2014, indoor and outdoor air samples were obtained from five kinder-gartens (named: K1, K2, K3, K4, K5) with wide geographical coveragerepresentative of the Hong Kong urban area. Fig. 1 shows the locationof these five kindergartens.

Two classrooms from each of the five kindergartens were selectedfor this study, all of which hadwooden flooring, plastic toys, wooden ta-bles and chairs, watercolour pigments, water-based paints and oil paintcovering the walls and ceilings. The rooms were ventilated throughdoors, windows, fans and air conditioning. Data on the characteristicsof the classrooms in each school were collected, and they are presentedin Table 1. The condition of the school buildings, the furnishing mate-rials and table setting were little different among classrooms in thesame kindergarten. As shown in Table 1, the size of the classroomsranged from 3.3 × 101 to 4.0 × 101 m2 (median: 3.8 × 101 m2) andtheir volume ranged between 1.3 × 102 and 1.9 × 102 m3 (median:146.0m3). The classroomswere furnishedwith a computer, a television,an air conditioning system, a fan, small wooden tables, plastic chairs andmany plastic toys.

2.2. Sampling and analysis

The indoor air pollutantsmeasured consisted of fineparticulatemat-ter (PM2.5), black carbon (BC) aerosol, carbonyl chemicals, PBDEs, andDP. The detailed procedures regarding sample preparation, extraction,and instrument analysis are described in the support information, S1.

The sampling equipmentwas placed 1.0m above thefloor level. Onesample of the above referenced pollutants was taken from each class-room. In each site, samples were collected on three non-consecutivedays inside the kindergarten and one day outside of the kindergarten.

The real-time level of PM2.5wasmeasured by aDust-Trak (TSIModel8532), which is a real-time laser photometer instrument used for deter-mining aerosol mass concentrations at the sampling rate of 1.7 L/min. Inaddition, PM2.5 samples were collected on a quartz filter (47 mm,Whatman Inc., USA), using MiniVol portable air samplers (Airmetrics,USA), as a Dust-Trak calibration system. Membrane filters were condi-tioned in dry air for at least 48 h and weighed to a precision of 10 mg.Real-time BC concentrations in the indoor air were measured using amicroAeth Model AE-51 (Magee Scientific, Berkeley, CA) by drawingambient air using a quartz filter based strip which then measured BCusing a single 880 nm LED.

The carbonyl concentrations were quantified using external calibra-tion curves constructed from the standard solutions of a TO11 carbonyl–DNPH (2,4-Dinitrophenylhydrazine) Mix. The sampling periods for car-bonyl compound levels were between five to eight hours depending onthe occupation of the classrooms by the students. In-laboratory experi-ments demonstrated that the collection efficiencies were N93 ± 5% forall target carbonyls under the same flow rate, temperature and RH (rel-ative humidity).

For the analysis of PBDEs and DP in PM2.5 and dust, GC-NCI/MSmethods were used. Deionised water was generated in-house using aSuper-Q water generation system. Dust and filters with sodium sul-phate were weighed in a centrifuge tube. The spiked samples were ex-tracted by a different ratio of solvents. The extracts were analysed byan Agilent gas chromatograph (GC)-mass spectrometer (MS) system(7890B-5977A) (Agilent Technologies, Palo Alto, CA) fitted with a DB-5MS column (15m long × 0.25mm i.d. × 0.10 μmor 0.25 μm film thick-ness), with helium at a rate of 1 ml/min as the carrier gas. Injection of a2 μL sample was performed with an automatic sampler using thesplitless injection mode. Quality assurance/quality control (QA/QC)was conducted by performing field and laboratory blanks, standardspiked recoveries and GC/MS detection limits. Daily multi-level calibra-tions were also conducted. The positive identification of PBDEs and DPpeaks was assured by both their retention times and their mass ratiosreferenced in the standards. All laboratory glassware, gloves and poly-ethylene bags used in sample preparation were checked to ensurethey were free of POPs. One laboratory control sample and one

Fig. 1. The map of selected sampling sites in this work.

67W.-J. Deng et al. / Environment International 96 (2016) 65–74

laboratory blank were included in each analytical batch of six samples.One duplicate sample analysis was performed with every secondbatch. The limits of detection (LOD) for targeted compounds were de-fined as three times the signal to noise (S/N) ratio, and ranged from2.1 to 1.3 × 101 ng/kg. The r2 was 0.9994–0.9999. The absolute recoveryefficiencies were 7.3 × 101 to 1.1 × 102% and the RSD ranged from 3.9 to1.4 × 101%.

2.3. Data analysis

All of the statistical tests were performed with SPSS 16.0 software(SPSS, Inc.). The normality of the data was checked through the Sha-piro-Wilk test. The probability distribution was tested through the Kol-mogorov-Smirnov test and all of the data were normal. The 50thpercentiles were calculated for measured indoor concentrations. The50th percentiles of different indoor and outdoor concentrations at thesame kindergarten were compared using a one-way ANOVA test. Pear-son regressions were used to test the correlations between the contam-inants. The probability value of p b 0.05 was considered to bestatistically significant.

Table 1Characteristics of kindergarten classroom.

School Floor level Area(m2)

Height(m)

Materials and age N

K1 Ground 3.9 × 101 4.9 Floor-PVC, 18 yearWall-paint, 18 year

ST

K2 Ground 4.0 × 101 3.7 Floor-PVC, 1 yearWall-paint, 1 year

ST

K3 Platform(forth)

3.5 × 101 3.7 Floor-PVCWall-paint

ST

K4 Ground 3.5 × 101 3.7 Floor-PVCWall-paint

ST

K5 Ground 3.3 × 101 4.0 Floor-PVC, 18 year Wall-paint, 18year

ST

2.4. Children's health risk assessment

To assess the health risk of children exposed to PBDEs and DP, threedaily exposure models (indoor inhalation, dust intake and dust dermalcontact) were applied to calculate the total daily intake from threenon-dietary sources. For carbonyl compounds, only indoor inhalationwas measured. The health risk to children's exposure pathways wasevaluated using the hazard quotient (HQ), which referred to the expo-sure factors articulated by the United States Environmental ProtectionAgency (2009), which were shown in Table 2. Children's total timethat they spend in schools was adjusted to 8 h, and the HQ values at50th percentile (median) were calculated.

2.4.1. Step 1 exposure assessment of PM inhalation

DDEair ¼Cair ng=m3

� �� IR m3=day� �� IEF

BW kgð Þ ð1Þ

where DDEair means dose of daily exposure of DP or PBDEs through

r people Ventilation Ambience

tudents-30eacher-2

Air conditioning-2; fan-2; door-open;window-close

TrafficwayResidentialbuildings

tudents-30eacher-2

Air conditioning-2; fan-2; door-open;window-close

Parking, marketResidentialbuildings

tudents-25eacher-2

Air conditioning-2; fan-0; door-open;window-close

Parking, marketsResidentialbuildings

tudents-30eacher-2

Air conditioning-2; fan-3; door-open;window-close

Auto repair shopsConstruction siteTrafficway

tudents-30eacher-2

Air conditioning-2; fan-2; door-open;window-close

Trafficway, viaductResidentialbuildings

Table 2The exposure assessment factors of different ages (USEPA, 2009).

Impact factors Infants Children Juveniles Teenagers Adults

Age (year) ≤1 1–5 6–11 12–19 ≥20Body weight (BW) (m3/kg) 5 16 29 52 65Ingestion rate (IR) of air (m3/day) 4.5 7.6 1.1 × 101 14 1.3 × 101

Indoor exposure frequency (IEF) (day/day) 0.88 0.79 0.79 0.88 0.88Dust inhalation rate (DIR) (g/day) 0.02 0.10 0.05 0.05 0.05Body surface area (BSA) (cm2/day) 8.0 × 102 2.6 × 103 3.1 × 103 3.7 × 103 4.6 × 103

Dust adhesion of surface (DAS) (mg/cm2) 0.096 0.096 0.096 0.096 0.096Absorb frequency (AF) 0.03 0.03 0.03 0.03 0.03

68 W.-J. Deng et al. / Environment International 96 (2016) 65–74

inhalation (ng/kg-bw·day), Cair = concentration of DP or PBDEs in theair (ng/m3). IR = inhalation rate, i.e., daily intake amount of air (m3/day), IEF=Indoor exposure factor (day/day), i.e. percentage of childrenactivities in indoors account for whole day. BW=body weight (kg). IR,IEF, BW for different age groups are listed in Table 2 (USEPA, 2009).

2.4.2. Step 2 exposure assessment of dust intake through dust non-dietaryingestion

DDEdust ¼Cdust ng=gð Þ � DIR g=dayð Þ � IEF

BW kgð Þ ð2Þ

where DDEdust means dose of daily exposure of DP or PBDEs throughdust non-dietary ingestion (ng/kg-bw ∙day); Cdust = concentration ofDP or PBDEs in the dust (ng/g). DIR = daily intake rate (g/day).

2.4.3. Step 3 exposure assessment of dust intake through dermal contact

DDEdust ¼Cdust ng=gð Þ � BSA cm2

� �� DAS mg=cm2� �

� AF � IEF

BW kgð Þ � 1000ð3Þ

where DDEdust means dose of daily exposure of DP or PBDEs throughdust dermal contact (ng/kg-bw·day); Cdust = concentration of DP orPBDEs in the dust (ng/g). BSA = body surface area where contactswith dust (cm2/day), AF= absorb factor of DP by children's skin.

In the end, total daily exposure (DDE) of DP or PBDEs via dust intakeand PM inhalation are calculated: DDE=DDEair+DDEdust+DDEdust2.

2.4.4. Step 4 health risk characterization

2.4.4.1. Non-cancer risk. The health risk children exposure pathways willbe evaluated by using the HQ, the ratio of dose of daily exposure (DDE)to reference dose (RfD) (mg/kg-bw·day) or reference concentration(RfC) (mg/M3); by using equation: HQi = DDEi / RfDi, or HQi =BW × DDEi / IR / RfCi, where i represents the various exposure routs,RfD=NOAEL / LOAEL (experimental dose) / (UF×MF), NOAEL is noob-served adverse effect level, LOAEL is the lowest observed adverse effectlevel, UF is the standard uncertainty factor and MF is the modifying fac-tor. The total children health risk for various exposure pathways will beassessed by the hazard index (HI) by using equation: HI = ∑HQi. IfHI N 1, adverse health effects for children are likely to occur, and viceversa.

2.4.4.2. Cancer risk. The health risk children by using equation: HQi =DDEi × SFi, SF is the cancer slope factor. The total children health riskfor various exposure pathways will be assessed by the hazard index(HI) by using equation: HI=∑HQi. If HI N 10−6, adverse health effectsfor children are likely to occur.

3. Results and discussion

3.1. Indoor and outdoor PM2.5 mass

The concentrations of the indoor and outdoor PM2.5measured in thisstudy's total sample are shown in Table 3. The 50th percentile (median)indoor and outdoor PM2.5 concentrations in the five kindergartensranged from 1.3 × 101 to 2.9 × 101 μg/m3 indoors and from 9.5 to8.8 × 101 μg/m3 outdoors, respectively.

Every indoor micro-environment has unique characteristics that aredetermined by the outdoor air quality, specific building characteristicsand indoor activities (Stranger et al., 2007). The WHO requires thePM2.5 24-hour average concentrations in populated areas to beb25 μg/m3. The mean PM2.5 levels of the five kindergartens were allhigher than the guidelines set by the WHO (25 μg/m3) except for K5.Compared with the WHO limitations, the PM2.5 concentration was se-verely above the standard. Continuous particle mass concentrationdata and time-activity diaries were used to identify indoor sources ofPM. PM is usually generated from a variety of outdoor sources, such asautomobile exhaust, industrial production processes, road traffic dustand secondary gaseous pollutants (Gartrell and Friedlander, 1975) Ad-ditionally, PM is generated from indoor sources, such as smoking,cooking, cleaning (Vette et al., 2001) and other activities (Chan, 2002).Outdoor particles penetrate the indoor environment through the aper-tures of doors and windows, and through the fresh air coming fromair-conditioning systems (Yu et al., 2009). The three typical distributionpatterns (significant correlation: K1, K2; no correlation: K3;weak corre-lation: K4, K5) for indoor/outdoor PM2.5 are illustrated in Fig. 2, whichshows the time-series plots of the 10-second average indoor and out-door PM2.5 data for the five schools. The curves illustrate the PM2.5 var-iations, suggesting that the particle events tended to be brief andsusceptible. This is consistent with observations made in other studies(Cao et al., 2005). Both K1 and K2 had many windows and were nearthe road and downtown. The K1 outdoor PM2.5 mass concentrationwas higher than the indoor PM2.5 (p b 0.001) and a significant correla-tion was shown to exist between them (r = 0.517, p b 0.001). Thehigher concentration of outdoor PMpenetrated the indoor environmentthrough the apertures of doors, windows, and air-conditioning systems.As seen in Fig. 2a, the indoor and outdoor PM2.5 concentrations closelytracked each other, showing a similar trend. Similar to K1, the outdoorPM2.5 of K2 was higher than it was indoors (p b 0.001) and there wasa significant correlation between them (r = 0.553, p b 0.001). Similarvariations of outdoor and indoor PM2.5 concentrations were observedin K1 and K2, indicating that outdoor traffic was the main factorinfluencing both of them.

K3 was located on a fourth floor platform about 10 m above theground. Accordingly, the outdoor particles were less affected by auto-mobile exhaust, road dust and other outdoor PM sources and showedno correlation with the indoors (r = 0.027, p = 0.154 N 0.05). As seenin Fig. 2c, the K3 outdoor PM2.5 mass concentration was significantlylower than the indoor PM2.5 (p b 0.001). K4 and K5 were away fromthe street or near the coast. The K4 and K5 outdoor PM2.5 mass concen-trations were lower than the indoor PM2.5 (p b 0.001) and there was aweak correlation between them (p b 0.001). In K3, K4 and K5, as seen

Table 3Concentrations of PM2.5 in sampling sites (μg/m3).

Site Indoor Outdoor

Min Max 50th percentile Min Max 50th percentile

K1 1.6 × 101 4.1 × 101 2.1 × 101 6.4 × 101 1.1 × 102 8.8 × 101

K2 1.6 × 101 3.8 × 101 2.3 × 101 1.1 × 101 8.3 × 101 3.6 × 101

K3 2.5 × 101 3.6 × 101 2.9 × 101 6.5 2.8 × 101 1.0 × 101

K4 1.5 × 101 3.6 × 101 1.8 × 101 9.0 2.9 × 101 1.5 × 101

K5 1.1 × 101 1.8 × 101 1.3 × 101 7.0 2.0 × 101 9.5

69W.-J. Deng et al. / Environment International 96 (2016) 65–74

in Fig. 2c, d and e, the indoor PM was primarily generated from indooractivities, such as cooking, cleaning and students' activities (smokingwas prohibited in school). These particles can penetrate the lungsmore deeply than coarse particles can through inhalation (Ohura etal., 2005). An increasing amount of data has shown that higher levelsof PM2.5 and PM10 should result in the greater prevalence of acute and

Fig. 2. Three typical distribution patterns

chronic health effects, including asthma among children (Daisey et al.,2003; Mendell and Heath, 2005).

More effort should therefore be made to control fine particle pollu-tion in schools. Currently there is no special PM2.5 standard for kinder-gartens, so more effort should also be made toward establishing aPM2.5 concentration standard in China. Different schools in this study

for indoor and outdoor PM2.5 data.

70 W.-J. Deng et al. / Environment International 96 (2016) 65–74

exhibited different pollution levels, indicating that indoor air quality canbe improved through artificial measures.

3.2. PBDEs and DP

Table S2 in the supporting information summarises the FR concen-trations (PBDEs and DP) of the indoor air, and indoor dust in the fivekindergartens (except for the indoor dust in K5) and three children'shomes. Twelve indoor air or dust samples were analysed for thirteenPBDE congeners and two DPs (anti-DP, syn-DP). We detected PBDEsand DP in 100% of the PM2.5 and dust samples collected. The∑13PBDE levels ranged from 0.10 ng/m3 to 0.64 ng/m3 in PM2.5, andwere similar to the previous observations made in Hong Kong(9.2 × 101–4.2 × 102 pg/m3) by Deng et al. (2007). These also reachedthe same levels as other schools, such as those in the US early childhoodeducation environments in California (0.01–2.0 × 101 pg/m3) (Bradmanet al., 2014) and the primary schools in Korea (n.d.–3.7 × 101 pg/m3)(Lim et al., 2014). In general, FR levels in the air were low, often belowdetection, or b1 ng/m3.

The levels of ∑13PBDE in the dust of the present sample (0.29–2.0 × 102 ng/g) were somewhat lower than the levels reported inother studies, such as those conducted in the primary schools in Austra-lia (1.1 × 101–2.2 × 103 ng/g) (Toms et al., 2015) and Korea (2.0 × 102–4.5 × 104 ng/g) (Lim et al., 2014).

The median concentration of anti-DP and syn-DP ranged from3.6 × 101 pg/m3 to 8.5 × 101 pg/m3 and from 1.0 × 101 pg/m3 to1.5 × 101 pg/m3 in PM2.5 in the indoor air, which was a little higherthan in comparable northeast Asian cities (0.87–6.7 pg/m3) (Kakimotoet al., 2015). The median concentrations of anti-DP and syn-DP in thedust ranged from 0.94 ng/g to 6.6 ng/g and from 0.32 ng/g to 2.2 ng/gin the individual kindergartens, whilst they ranged from 2.9 ng/g to5.4 ng/g and from 0.70 ng/g to 1.7 ng/g in the dust found in students'homes. These concentrations were about one order lower than in thehouse dust reported from Ottawa, Ontario, Canada (mean,2.9 × 101 ng/g) (Zhu et al., 2007). The concentrations of anti-DP weremostly higher than the syn-DP.

In this study, FR levels in the air were not correlated with levels inthe dust, which coincided with the findings of Vorkamp et al. (2011)and Imm et al. (2009).

In this study, BDE47 and BDE99 comprised the bulk of the targetedFR mass measured in the PM2.5 and dust samples. Although BDE47and BDE99 are no longer in use, furniture and other long-lasting prod-ucts containing these materials can still be found in many buildings.BDE-47, BDE-99, BDE-100, BDE-153, and BDE-154 congeners are mostprevalent in human tissue and compose approximately 90% of thetotal body burden (Frederiksen et al., 2009).

3.3. Black carbon

The median of BC concentrations during the sampling period variedby N7-fold from 8.8 × 102 ng/m−3 to 6.7 × 103 ng/m−3. The average

Table 4Comparison of BC concentrations with other studies (ng/m3).

Study City Environment

Jansen et al. (2005) Seattle Indoor (homeOutdoor (amb

Cao et al. (2009) Xi'an AmbientRichmond-Bryant et al. (2009) New York Outdoor (Scho

Outdoor (SchoReche et al. (2011) London Urban traffic

Urban backgroRehman et al. (2011) Surya Indoor (kitche

Outdoor (ambThis study Hong Kong Indoor (classr

Outdoor (amb

indoor BC concentrations in K1, K2, K3, K4 and K5 were 1.6 × 103 ng/m−3, 1.6 × 103 ng/m−3, 3.3 × 103 ng/m−3, 4.2 × 103 ng/m−3, and2.0 × 103 ng/m−3, respectively. The outdoor BC concentrations were3.8 × 103 ng/m−3, 3.4 × 103 ng/m−3, 1.4 × 103 ng/m−3, 1.6 × 103 ng/m−3 and 8.8 × 102 ng/m−3 in K1, K2, K3, K4 and K5, respectively. Theblack carbon concentrations in the five kindergartens were different re-gardless ofwhether theywere indoors or outdoors. Differences in theO/I (outdoor/indoor) ratios were substantial, varying from 0.37 to 2.4.Black carbon (BC) is one component of airborne particulate matter(PM) and it is generated only by the incomplete combustion of fuelscontaining carbon (Hansen et al., 1988). The highest O/I ratios were2.4, 2.2, 0.42, 0.37 and 0.45 for K1, K2, K3, K4 and K5, respectively. Thedifferencesmay have been attributed to diverse sources such as cookingand auto emissions (Rehman et al., 2011). Table 4 shows the resultsfrom similar studies in other cities.

Our mean indoor and outdoor BC concentrations of 2.6 × 103 ng/m3

and 2.4 × 103 ng/m3 were close to the values reported in the studies ofRichmond-Bryant et al. (2009) for outdoor ambient BC concentrationsin New York schools, and were somewhat below the values reportedfor the traffic environment of urban areas and the cooking environmentin households. Rehman et al. (2011) noted that the BC concentrations inkitchens increased to 6.2 × 104 ng/m3 during evening cooking, whichwas about 2.4 times more than the outdoor BC (2.6 × 104 ng/m3).

Fig. 3 shows the daily diurnal variations of ambient and indoor BCconcentrations at the five kindergartens. Fig. 3a shows the daily diurnalvariation of the ambient BC concentration at K1. This variation had asimilar trend to the indoor BCmass concentrationswith the peaks close-ly following the lunchtime cooking cycle, suggesting the strong influ-ence of indoor cooking on outdoor BC concentrations (r = 0.474,p b 0.001). Further, the outdoor variation shows a 15 minute delay. InFig. 3b and c, no correlations were found between indoor and outdoorBC concentrations. Unlike Fig. 3a, Fig. 3d and e shows that the daily di-urnal variations of indoor BC concentrations were higher than theywere outdoors (p b 0.001), but again no correlations were found(p N 0.05). During lunchtime cooking hours, the outdoor BC concentra-tion of K1varied from 3.5 × 103 to 4.6 × 103 ng/m3with amean value of4.0 × 103 ng/m3 and the indoor BC concentration varied from 816 to1905 ng/m3 with a mean value of 1682 ng/m3, respectively. Similarly,BC concentrations caused by cooking events were found in other sam-pling sites.

3.4. Carbonyl compounds

Concentrations of the 16 carbonyls found in the indoor and outdoorair of the five Hong Kong kindergartens were determined as shown inTable 5.

Large variations in the 16 carbonyl compounds were observed be-tween the five sampling sites. The total concentrations of 16 carbonylsranged from 4.7 × 101 μg/m3 to 9.3 × 101 μg/m3 indoors and from1.9 × 101 μg/m3 to 4.3 × 101 μg/m3 outdoors. Formaldehyde was themost abundant air aldehyde, followed by acetaldehyde, acetone, 2,5-

Mean Range

) 1.3 × 104 –ient) 2.0 × 104 –

1.5 × 104 –ol 1) 1.8 × 103 8.0 × 102–2.5 × 103

ol 2) 1.9 × 103 9.0 × 102–3.4 × 103

3.2 × 103 1.6 × 103–5.1 × 103

und 1.9 × 103 8.0 × 102–3.4 × 103

n-evening cooking) 6.2 × 104 3.0 × 103–1.1 × 106

ient-evening cooking) 2.6 × 104 3.0 × 103–1.8 × 105

oom) 2.6 × 103 9.5 × 102–6.7 × 103

ient) 2.4 × 103 8.8 × 102–3.8 × 103

Fig. 3. Daily diurnal variation of BC concentration for the sampling period in each site.

71W.-J. Deng et al. / Environment International 96 (2016) 65–74

dimethylbenzaldehyde and benzaldehyde. The median concentrationsof formaldehyde ranged from 1.0 × 101–4.8 × 101 μg/m3 (indoors) to4.3–1.3 × 101 μg/m3 (outdoors) in five kindergartens.

The same levels of formaldehyde concentrations were found in pri-mary schools in Turkey (indoor 6.7 × 101–1.1 × 102 μg/m3; outdoor0.60–8.8 μg/m3) (Sofuoglu et al., 2011), elementary schools in Portugal

Table 5Concentrations (50th percentile) of the sixteen carbonyl compounds (μg/m3).

Carbonyls K1 K2

IDa ODb ID OD

Formaldehyde 2.5 × 101 1.4 × 101 1.2 × 101 1.3 × 101

Acetaldehyde 5.2 5.2 4.4 2.3Acetone 9.9 c

Propanal 5.5 2.2 × 101

Methyl ethyl ketone 0.68Iso- + n-butanal 0.17 0.19Benzaldehyde 0.20 0.44 0.68Isopentanal 0.21 0.16 0.27n-Pentanal 0.43 1.1 1.7 0.29o-Tolualdehyde 0.82 0.66 1.15m-Tolualdehyde 1.7 1.2 0.05 0.52p-Tolualdehyde 0.16 1.3 4.6Hexanal 0.16 9.732,5-Dimethylbenzaldehyde 3.1 0.63GlyoxalMethylglyoxalp-Value 0.01R 0.75

a ID: indoor.b OD: outdoor.c Not detected.

(indoor 1.48–4.2 × 101 μg/m3) (Pegas et al., 2011), in nine schools inAustria (indoor 6.5–1.4 × 102 μg/m3) (Wallner et al., 2012), and primaryschools in the Netherlands (indoor 1.3 × 101–2.6 × 101 μg/m3)(Scheepers et al., 2015).

Levels of formaldehyde and acetaldehyde were higher indoors thanoutdoors (mean indoor to outdoor [I/O] ratio = 2.4 and 1.9,

K3 K4 K5

ID OD ID OD ID OD

1.5 × 101 1.4 × 101 1.0 × 101 4.3 4.8 × 101 8.36.6 1.8 1.1 4.3 5.2 5.3

0.124.3 × 101 8.5 × 101 3.7 2.4 × 101 1.6

0.23 0.39 0.94 0.450.15 0.220.23 0.230.37 0.57 0.52 0.120.67 0.22 0.72 0.37 0.290.05 1.6 3.7 1.10.38 0.20 0.31 0.63 0.37 0.16

1.2 0.07 0.552.5 × 101 0.19 0.39 0.31 0.78

4.47 0.05 2.38 2.45 1.1

0.05 0.010.56 0.83

Table 6Non-cancer and cancer quotient (HQ) and index (HI) levels.

Site Non-cancer risk Cancer risk (fromformaldehyde)

PBDEs (HQ) DPs (HQ) Indoor carbonylsd Indoor total Outdoor carbonylsd Indoord Outdoord

K1 PM2.5a 8.0E−02 6.0E−03 1.0 1.1 0.56 1.1E−04 4.5E−05

DNIb 5.5E−03 1.6E−09DDCc 4.1E−04 1.2E−10

K2 PM2.5 4.4E−02 7.4E−03 0.60 0.65 0.70 5.3E−05 5.7E−05DNI 4.9E−03 5.3E−10DDC 3.6E−04 3.9E−11

K3 PM2.5 2.5E−02 6.4E−03 0.77 0.80 0.97 6.6E−05 6.2E−05DNI 5.6E−04 4.1E−09DDC 4.2E-05 3.0E−10

K4 PM2.5 1.5E−02 6.2E−03 0.77 0.80 0.31 4.5E−05 1.9E−05DNI 1.5E−05 7.1E-10DDC 1.1E−06 5.3E−11

K5 PM2.5 1.2E−02 5.8E−03 1.9 1.9 0.50 2.1E−04 3.6E−05

a PM2.5 inhalation.b Dust non-dietary ingestion.c Dust dermal contact.d Inhalation.

72 W.-J. Deng et al. / Environment International 96 (2016) 65–74

respectively). In general, as would be expected, indoor air levels werehigher than outdoor levels for several carbonyls. The concentration ofspecific carbonyls in the indoor air was higher than the outdoor air, in-dicating the release of carbonyls from indoor sources. The indoor andoutdoor carbonyl concentrations from K1, K4, and K5 show significantcorrelations between them (r = 0.75, p b 0.01 for K1; r = 0.56,p b 0.05 for K4; r = 0.83, p b 0.01 for K5).

Carbonyl compounds are awell-known group of indoor and outdoorchemical pollutants belonging to VOCs and have been studied for theiradverse effects on human health. Large quantities of carbonyl com-pounds are used worldwide as raw materials and intermediates in thechemical, wood and plastics manufacturing industries (Lee et al.,2001; Lee et al., 2002; Li et al., 2001).

3.5. Risk assessment

In this study, as shown in Table 6, the non-cancer HQs from PBDEsand DP were both lower than 0.08 (HQ b1) in the kindergartens andchildren's homes using the oral reference doses (RfDs) suggested bythe US Environmental Protection Agency's Risk Information System.Non-cancer HQs of PBDEs through air inhalationwas at least two ordershigher than it was through dust exposure, non-dietary ingestion anddermal contact. The same thing also happened with DP, where the

Fig. 4. Relative proportion of each congener's mass to total target fire retardant mass.

non-cancer HQs from air inhalation was at least seven orders higherthan it was through dust.

Research by Jones-Otazo et al. (2005) found that the human intakeof PBDEs through exposure to indoor dust accounted for 82% of thetotal intake amount. In this study, the human intake of PBDEs and DPfrom inhalation of PM2.5 accounted for 78.3% to 92.3% of the total intake(Fig. 4).

Smaller sized particles aremore resident in the atmosphere and theymore easily penetrate the respiratory system. FRs are inclined to adsorbsmall size particles because of their strong hydrophobicity, and theypose a higher adverse effects to children. Thus, the combined exposureto indoor particles and FRs may present further adverse effects in addi-tion to those calculated in this study.

Non-cancer HIs of indoor carbonyl compounds ranged from 0.77 to1.9 (HI N 1), and accounted for N92%of the total HIs indoors. Non-cancerHIs of outdoor carbonyl compounds ranged from 0.50 to 0.97, whichwere close to 1. When below 1, adverse effects are unlikely and whenabove 1, adverse effects are possible.

As shown in Table 6, the cancer HQs of formaldehyde ranged from4.5E−05 to2.1E−04 indoors and from1.9E−05 to 6.2E−05 outdoors,which were both more than 1E−06. Thus, adverse effects of formalde-hyde to the children in the kindergartens are possible.

4. Conclusion

This study detected PBDEs, DP, carbonyl compounds and black car-bon aerosol in five Hong Kong kindergartens. The PM2.5 levels of thefive kindergartens were higher, and several of them exceeded theWHO standard. Indoor PM2.5 mass concentrations were correlatedwith outdoor PM2.5 in four of the kindergartens,whereas indoor carbon-yl concentrations were correlated with outdoor carbonyls in threekindergartens. Formaldehyde was the most abundant carbonyl com-pound. Cooking events might cause BC concentration levels to rise,both indoors and outdoors. Carbonyl compounds might or might notpose adverse effect to children,whilst PBDEs andDPdonot pose any ad-verse effect at all.

Acknowledgements

The authors acknowledge the support of the Early Career Start/Gen-eral Research Fund of HongKong (Code no. ECS/GRF 845212), the FLASSDean's Research Grants (04018 and 04176) and Internal Research Grant(R3679) of The Education University of Hong Kong.

73W.-J. Deng et al. / Environment International 96 (2016) 65–74

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.envint.2016.08.013.

References

Annesi-Maesano, I., Hulin, M., Lavaud, F., Raherison, C., Kopferschmitt, C., de Blay, F., et al.,2012. Poor air quality in classrooms related to asthma and rhinitis in primaryschoolchildren of the French 6 Cities Study. Thorax 67, 682–688.

ATSDR, U., 2004. Toxicological Profile for Polybrominated Biphenyls and PolybrominatedDiphenyl Ethers. US Dept. Division of Toxicology/Toxicology Information Branch, At-lanta, Georgia, pp. 11–13.

Bono, R., Tassinari, R., Bellisario, V., Gilli, G., Pazzi, M., Pirro, V., et al., 2015. Urban air andtobacco smoke as conditions that increase the risk of oxidative stress and respiratoryresponse in youth. Environ. Res. 137, 141–146.

Bradman, A., Castorina, R., Gaspar, F., Nishioka, M., Colón, M., Weathers, W., et al., 2014.Flame retardant exposures in California early childhood education environments.Chemosphere 116, 61–66.

Branco, P.T.B.S., Alvim-Ferraz, M.C.M., Martins, F.G., Sousa, S.I.V., 2014a. Indoor air qualityin urban nurseries at Porto city: particulate matter assessment. Atmos. Environ. 84,133–143.

Branco, P.T.B.S., Alvim-Ferraz, M.C.M., Martins, F.G., Sousa, S.I.V., 2014b. The microenvi-ronmental modelling approach to assess children's exposure to air pollution – a re-view. Environ. Res. 135, 317–332.

Branco, P.T.B.S., Alvim-Ferraz, M.C.M., Martins, F.G., Sousa, S.I.V., 2015. Children's exposureto indoor air in urban nurseries-part I: CO2 and comfort assessment. Environ. Res.140, 1–9.

Cao, J.J., Lee, S.C., Chow, J.C., Cheng, Y., Ho, K.F., Fung, K., et al., 2005. Indoor/outdoor rela-tionships for PM2.5 and associated carbonaceous pollutants at residential homes inHong Kong - case study. Indoor Air 15, 197–204.

Cao, J.J., Zhu, C.-S., Chow, J.C., Watson, J.G., Han, Y.-M., Wang, G.-h., et al., 2009. Black carbonrelationships with emissions and meteorology in Xi'an, China. Atmos. Res. 94, 194–202.

Carreiro-Martins, P., Viegas, J., Papoila, A., Aelenei, D., Caires, I., Araújo-Martins, J., et al.,2014. CO2 concentration in day care centres is related to wheezing in attending chil-dren. Eur. J. Pediatr. 173, 1041–1049.

Cartieaux, E., Rzepka, M.A., Cuny, D., 2011. Indoor air quality in schools. Arch. Pediatr. 18,789–796.

Chan, A.T., 2002. Indoor–outdoor relationships of particulate matter and nitrogen oxidesunder different outdoor meteorological conditions. Atmos. Environ. 36, 1543–1551.

Chithra, V.S., Shiva Nagendra, S.M., 2012. Indoor air quality investigations in a naturallyventilated school building located close to an urban roadway in Chennai, India.Build. Environ. 54, 159–167.

Daisey, J.M., Angell, W.J., Apte, M.G., 2003. Indoor air quality, ventilation and health symp-toms in schools: an analysis of existing information. Indoor Air 13, 53–64.

Deng, W.J., Zheng, J.S., Bi, X.H., Fu, J.M., Wong, M.H., 2007. Distribution of PBDEs in air par-ticles from an electronic waste recycling site compared with Guangzhou and HongKong, South China. Environ. Int. 33, 1063–1069.

Dingemans, M.M.L., van den Berg, M., Westerink, R.H.S., 2011. Neurotoxicity of brominat-ed flame retardants: (in)direct effects of parent and hydroxylated polybrominateddiphenyl ethers on the (developing) nervous system. Environ. Health Perspect. 119,900–907.

Dodson, R.E., Houseman, E.A., Levy, J.I., Spengler, J.D., Shine, J.P., Bennett, D.H., 2007. Mea-sured andmodeled personal exposures to and risks from volatile organic compounds.Environ. Sci. Technol. 41, 8498–8505.

Eskenazi, B., Chevrier, J., Rauch, S.A., Kogut, K., Harley, K.G., Johnson, C., et al., 2013. Inutero and childhood polybrominated diphenyl ether (PBDE) exposures andneurodevelopment in the CHAMACOS study. Environ. Health Perspect. 121, 257–262.

Fonseca, J., Slezakova, K., Morais, S., Pereira, M.C., 2014. Assessment of ultrafine particlesin Portuguese preschools: levels and exposure doses. Indoor Air 24, 618–628.

Frederiksen, M., Vorkamp, K., Thomsen, M., Knudsen, L.E., 2009. Human internal and ex-ternal exposure to PBDEs – a review of levels and sources. Int. J. Hyg. Environ. Health212, 109–134.

Gartrell, G., Friedlander, S.K., 1975. Relating particulate pollution to sources: the 1972 Cal-ifornia aerosol characterization study. Atmos. Environ. (1967) 9, 279–299.

Gascon,M., Vrijheid, M., Martínez, D., Forns, J., Grimalt, J.O., Torrent, M., et al., 2011. Effectsof pre and postnatal exposure to low levels of polybromodiphenyl ethers onneurodevelopment and thyroid hormone levels at 4 years of age. Environ. Int. 37,605–611.

Gemenetzis, P., Moussas, P., Arditsoglou, A., Samara, C., 2006. Mass concentration and el-emental composition of indoor PM2.5 and PM10 in University rooms in Thessaloniki,northern Greece. Atmos. Environ. 40, 3195–3206.

Godwin, C., Batterman, S., 2007. Indoor air quality in Michigan schools. Indoor Air 17,109–121.

Hansen, A.D.A., Bodhaine, B.A., Dutton, E.G., Schnell, R.C., 1988. Aerosol black carbonmea-surements at the South Pole: initial results, 1986–1987. Geophys. Res. Lett. 15,1193–1196.

Harrad, S., Hazrati, S., Ibarra, C., 2006. Concentrations of polychlorinated biphenyls in in-door air and polybrominated diphenyl ethers in indoor air and dust in Birmingham,United Kingdom: implications for Human Exposure. Environ. Sci. Technol. 40,4633–4638.

Hodgson, A.T., Shendell, D.G., Fisk,W.J., Apte, M.G., 2004. Comparison of predicted and de-rived measures of volatile organic compounds inside four new relocatable class-rooms. Indoor Air 14, 135–144.

Hoh, E., Zhu, L., Hites, R.A., 2006. Dechlorane Plus, a chlorinated flame retardant, in theGreat Lakes. Environ. Sci. Technol. 40, 1184–1189.

Hwang, H.-M., Park, E.-K., Young, T.M., Hammock, B.D., 2008. Occurrence of endocrine-disrupting chemicals in indoor dust. Sci. Total Environ. 404, 26–35.

Imm, P., Knobeloch, L., Buelow, C., Anderson, H.A., 2009. Household exposures topolybrominated diphenyl ethers (PBDEs) in a Wisconsin Cohort. Environ. HealthPerspect. 117, 1890–1895.

Jansen, K.L., Larson, T.V., Koenig, J.Q., Mar, T.F., Fields, C., Stewart, J., et al., 2005. Associa-tions between health effects and particulate matter and black carbon in subjectswith respiratory disease. Environ. Health Perspect. 113, 1741–1746.

Jones-Otazo, H.A., Clarke, J.P., Diamond, M.L., Archbold, J.A., Ferguson, G., Harner, T., et al.,2005. Is house dust the missing exposure pathway for PBDEs? An analysis of theurban fate and human exposure to PBDEs. Environ. Sci. Technol. 39, 5121–5130.

Kakimoto, K., Nagayoshi, H., Akutsu, K., Konishi, Y., Kajimura, K., Hayakawa, K., et al., 2015.Dechlorane plus and decabromodiphenyl ether in atmospheric particles of northeastAsian cities. Environ. Sci. Pollut. Res. 22, 14600–14605.

Kohler, M., Tremp, J., Zennegg, M., Seiler, C., Minder-Kohler, S., Beck, M., et al., 2005. Jointsealants: an overlooked diffuse source of polychlorinated biphenyls in buildings. En-viron. Sci. Technol. 39, 1967–1973.

Lee, S.C., Ho, K.F., Chan, L.Y., Zielinska, B., Chow, J.C., 2001. Polycyclic aromatic hydrocar-bons (PAHs) and carbonyl compounds in urban atmosphere of Hong Kong. Atmos.Environ. 35, 5949–5960.

Lee, S.C., Li, W.-M., Ao, C.-H., 2002. Investigation of indoor air quality at residential homesin Hong Kong—case study. Atmos. Environ. 36, 225–237.

Lee, E., Kim, T.H., Choi, J.S., Nabanata, P., Kim, N.Y., Ahn, M.Y., et al., 2010. Evaluation ofliver and thyroid toxicity in Sprague-Dawley rats after exposure to polybrominateddiphenyl ether BDE-209. J. Toxicol. Sci. 35, 535–545.

Li,W.M., Lee, S.C., Chan, L.Y., 2001. Indoor air quality at nine shoppingmalls in Hong Kong.Sci. Total Environ. 273, 27–40.

Lim, Y.-W., Kim, H.-H., Lee, C.-S., Shin, D.-C., Chang, Y.-S., Yang, J.-Y., 2014. Exposure as-sessment and health risk of poly-brominated diphenyl ether (PBDE) flame retardantsin the indoor environment of elementary school students in Korea. Sci. Total Environ.470–471, 1376–1389.

Lin, M., Chen, Y., Burnett, R.T., Villeneuve, P.J., Krewski, D., 2002. The influence of ambientcoarse particulate matter on asthma hospitalization in children: case-crossover andtime-series analyses. Environ. Health Perspect. 110, 575–581.

Loh, M.M., Levy, J.I., Spengler, J.D., Houseman, E.A., Bennett, D.H., 2007. Ranking cancerrisks of organic hazardous air pollutants in the United States. Environ. HealthPerspect. 115, 1160–1168.

Massolo, L., Rehwagen, M., Porta, A., Ronco, A., Herbarth, O., Mueller, A., 2010. Indoor–out-door distribution and risk assessment of volatile organic compounds in the atmo-sphere of industrial and urban areas. Environ. Toxicol. 25, 339–349.

Mendell, M.J., Heath, G.A., 2005. Do indoor pollutants and thermal conditions in schoolsinfluence student performance? A critical review of the literature. Indoor Air 15,27–52.

Möller, A., Xie, Z., Sturm, R., Ebinghaus, R., 2010. Large-scale distribution and transport ofDechlorane Plus in air and sea water of the Arctic to Antarctic. Environ. Sci. Technol.44, 8977–8982.

Ohura, T., Noda, T., Amagai, T., Fusaya, M., 2005. Prediction of personal exposure to PM2.5

and carcinogenic polycyclic aromatic hydrocarbons by their concentrations in resi-dential microenvironments. Environ. Sci. Technol. 39, 5592–5599.

Pegas, P.N., Alves, C.A., Evtyugina, M.G., Nunes, T., Cerqueira, M., Franchi, M., et al., 2011.Indoor air quality in elementary schools of Lisbon in spring. Environ. Geochem.Health 33, 455–468.

Pope, C.A., Dockery, D.W., 2006. Health effects of fine particulate air pollution: lines thatconnect. J. Air Waste Manage. Assoc. (1995) 56, 709–742.

Qiu, X.H., Hites, R.A., 2008. Dechlorane Plus and other flame retardants in tree bark fromthe northeastern United States. Environ. Sci. Technol. 42, 31–36.

Reche, C., Querol, X., Alastuey, A., Viana, M., Pey, J., Moreno, T., et al., 2011. New consider-ations for PM, Black Carbon and particle number concentration for air quality moni-toring across different European cities. Atmos. Chem. Phys. 11, 6207–6227.

Rehman, I.H., Ahmed, T., Praveen, P.S., A., K., 2011. Black carbon emissions from biomassand fossil fuels in rural India. Atmospheric Chemistry and Physics (ACP) & Discus-sions (ACPD).

Ren, G.F., Yu, Z.Q., Ma, S.T., Li, H.R., Peng, P.G., Sheng, G.Y., Fu, J.M., 2009. Determination ofdechlorane plus in serum from electronics dismantling workers in South China. Envi-ron. Sci. Technol. 43, 9453–9457.

Ren, G.F., Yu, Z.Q., Ma, S.T., Chen, L.G., Luo, X.F., Sheng, G.Y., Fu, J.M., 2010. Dechlorane Plusin in-and outdoor air of an urban city in southern China. Organohalogen Compd. 72,852–855.

Richmond-Bryant, J., Saganich, C., Bukiewicz, L., Kalin, R., 2009. Associations of PM2.5 andblack carbon concentrations with traffic, idling, background pollution, and meteorol-ogy during school dismissals. Sci. Total Environ. 407, 3357–3364.

Rivas, I., Viana, M., Moreno, T., Pandolfi, M., Amato, F., Reche, C., et al., 2014. Child expo-sure to indoor and outdoor air pollutants in schools in Barcelona, Spain. Environ.Int. 69, 200–212.

Salvi, S., 2007. Health effects of ambient air pollution in children. Paediatr. Respir. Rev. 8,275–280.

Sax, S.N., Bennett, D.H., Chillrud, S.N., Ross, J., Kinney, P.L., Spengler, J.D., 2006. A cancerrisk assessment of inner-city teenagers living in New York city and Los Angeles. En-viron. Health Perspect. 114, 1558–1566.

Scheepers, P.T.J., de Hartog, J.J., Reijnaerts, J., Beckmann, G., Anzion, R., Poels, K., et al.,2015. Influence of combined dust reducing carpet and compact air filtration uniton the indoor air quality of a classroom. Environ. Sci.: Processes Impacts 17,316–325.

Schwartz, J., 2004. Air pollution and children's health. Pediatrics 113, 1037–1043.

74 W.-J. Deng et al. / Environment International 96 (2016) 65–74

Sofuoglu, S.C., Aslan, G., Inal, F., Sofuoglu, A., 2011. An assessment of indoor air concentra-tions and health risks of volatile organic compounds in three primary schools. Int.J. Hyg. Environ. Health 214, 36–46.

Sohn, J.R., Moon, H.J., Ryu, S.H., Hwang, T., 2012. Indoor air quality in school classroomswith mechanical ventilation systems. Proceedings of the 7th International Sympo-sium on Sustainable Healthy Buildings, Seoul, Korea, p. 349.

Sousa, S.I.V., Pires, J.C.M., Martins, E.M., Fortes, J.D.N., Alvim-Ferraz, M.C.M., Martins, F.G.,2012. Short-term effects of air pollution on respiratory morbidity at Rio de Janeiro— part II: health assessment. Environ. Int. 43, 1–5.

Stranger, M., Potgieter-Vermaak, S.S., Van Grieken, R., 2007. Comparative overview of in-door air quality in Antwerp, Belgium. Environ. Int. 33, 789–797.

Takigami, H., Suzuki, G., Hirai, Y., S-i, S., 2009. Brominated flame retardants and otherpolyhalogenated compounds in indoor air and dust from two houses in Japan.Chemosphere 76, 270–277.

Toms, L.-M.L., Mazaheri, M., Brommer, S., Clifford, S., Drage, D., Mueller, J.F., et al., 2015.Polybrominated diphenyl ethers (PBDEs) in dust from primary schools in SouthEast Queensland, Australia. Environ. Res. 142, 135–140.

USEPA, 2009. Exposure Factors Handbook: 2009 Update. National Center for Environmen-tal Assessment, Washington D.C.

Venier, M., Hites, R.A., 2008. Flame retardants in the atmosphere near the Great Lakes. En-viron. Sci. Technol. 42 (13), 4745–4751.

Vette, A.F., Rea, A.W., Lawless, P.A., Rodes, C.E., Evans, G., Highsmith, V.R., et al., 2001.Characterization of indoor-outdoor aerosol concentration relationships during theFresno PM exposure studies. Aerosol Sci. Technol. 34, 118–126.

Vorkamp, K., Thomsen, M., Frederiksen, M., Pedersen, M., Knudsen, L.E., 2011.Polybrominated diphenyl ethers (PBDEs) in the indoor environment and associationswith prenatal exposure. Environ. Int. 37, 1–10.

Wallner, P., Kundi, M., Moshammer, H., Piegler, K., Hohenblum, P., Scharf, S., et al., 2012.Indoor air in schools and lung function of Austrian school children. J. Environ.Monit. 14, 1976–1982.

Wang, B., Iino, F., Huang, J., Lu, Y., Yu, G., Morita, M., 2010. Dechlorane Plus pollution andinventory in soil of Huai'an City, China. Chemosphere 80 (11), 1285–1290.

WHO, 2003. Physical School Environment: An Essential Component of a Health-promot-ing School.

WHO, 2006. Air quality guidelines—global update 2005.WHO, 2010. WHO Guidelines for Indoor Air Quality: Selected Pollutants.Wu, X., Apte, M.G., Maddalena, R., Bennett, D.H., 2011. Volatile organic compounds in

small- and medium-sized commercial buildings in California. Environ. Sci. Technol.45, 9075–9083.

Yang, W., Sohn, J., Kim, J., Son, B., Park, J., 2009. Indoor air quality investigation accordingto age of the school buildings in Korea. J. Environ. Manag. 90, 348–354.

Yu, B.F., Hu, Z.B., Liu, M., Yang, H.L., Kong, Q.X., Liu, Y.H., 2009. Review of research on air-conditioning systems and indoor air quality control for human health. Int. J. Refrig.32, 3–20.

Zhang, G., Spickett, J., Rumchev, K., Lee, A.H., Stick, S., 2006. Indoor environmental qualityin a ‘low allergen’ school and three standard primary schools inWestern Australia. In-door Air 16, 74–80.

Zhu, J., Feng, Y.L., Shoeib, M., 2007. Detection of dechlorane plus in residential indoor dustin the city of Ottawa, Canada. Environ. Sci. Technol. 41, 7694–7698.