Polybrominated Dibenzo-p -dioxins/Furans (PBDD/Fs) and Diphenyl Ethers (PBDEs) in the Indoor and Outdoor of Gymnasiums

It is still unknown whether the organobromine compounds in the indoor environment of the gyms would affect human health. This study tried to investigate PM 2.5 and indoor-dust levels of polybrominated dibenzo-p -dioxins/furans (PBDD/Fs) and diphenyl ethers (PBDEs) in the gyms and then to further assess the health risks of the gym users and employees who were exposed to these organobromines through inhalation and ingestion. The gyms were to be examined for PBDD/DF/DEs on PM 2.5 indoors and outdoors and in indoor dust. The results showed that indoor-dust, indoor PM 2.5 , and outdoor PM 2.5 levels of PBDD/Fs were 37.8 ± 13.7 pg WHO 2005 -TEQ g –1 and 0.00650 ± 0.00340 and 0.00469 ± 0.00101 pg WHO 2005 -TEQ m –3 , respectively, in the gyms, while for PBDEs, they were 2670 ± 2330 ng g –1 and 9.54 ± 4.68, and 8.27 ± 3.46 pg m –3 . The PBDEs levels observed in this study were relatively lower when compared with the current global data of PM 2.5 and indoor dust levels of PBDEs. The observed PM


INTRODUCTION
Due to the high prevalence of hypertension, hyperglycemia, and hyperlipidemia in Taiwan (Hsu et al., 2021), these chronic diseases are related to metabolic syndromes and are also associated with people with insufficiently physical exercise (Booth et al., 2012).For the improvement of the health of the Taiwanese citizens, the Health Promotion Administration, and the Ministry of Health and Welfare, Taiwan encourage the citizens for sufficient activities which in turn accelerated the exponentially increased numbers of gymnasiums or gyms in Taiwan (Chou and Huang, 2021).The gyms were provided for sports and physical activity in the indoor environment.The indoor air quality (IAQ) in the microenvironments of the gyms was related to human health, especially for physical activities.The exponentially increased numbers of indoor sports facilities and workers were observed due to the dramatical increase of sports enthusiasts in the last decade in Taiwan (Chou and Huang, 2021).Thus, the concerns of both public and government (Environmental Protection Administration, Taiwan (TEPA)) for IAQ of the gyms have been raised.It was observed that the concentration levels of carbon dioxide (CO2) and particulate matters (PMs) were significantly affected by the time of elevated human occupation, frequent human activity, and poor ventilation in fitness centres or gyms (Castro et al., 2015;Andrade and Dominski, 2018;Slezakova et al., 2018;Salonen et al., 2020).A gym located in a university was known to contain higher levels of PM2.5, PM10, and carbonyl compounds including acrolein and formaldehyde indoors than that outdoors (Alves et al., 2013;Castro et al., 2015).PM2.5 levels in gyms or fitness centres were also notably associated with human occupation and frequent activities (Castro et al., 2015).This indicated that the short-term physical activity like cardio activities might have significant effects on the daily inhalation dose and it was also demonstrated that the exercising females possibly inhaled the ultrafine particulate (UFP) 1.2 times higher when compared with the exercising males in the fitness centres (Slezakova et al., 2018).
Like the other indoor environments in workplaces, a lot of consumer products or electronic equipment are now available in the gyms also.It is still unknown whether the brominated flame retardants (BFRs), such as polybrominated diphenyl ethers (PBDEs) and the dibrominated by-product of BFRs, including polybrominated dibenzo-p-dioxins/furans (PBDD/Fs), are widespread in the gym environments.However, it has been demonstrated to have a negative impact on human exposure to PBDEs or PBDD/Fs leading to human health effects such as decreased semen quality, alteration of thyroid and growth hormone homeostasis, delay of neurological behaviour and neurodevelopment, and disruption of menstrual cycles and pregnancy periods (Chao et al., 2014a).Our previous studies indicated that PBDEs or PBDD/Fs were found in several indoor environments including residential houses, classrooms in elementary schools, offices, medical and dental clinics, and workplaces in electronic and vehicle dismantler factories (Chao et al., 2014c;Shy et al., 2015;Chao et al., 2016;Chou et al., 2016;Gou et al., 2016aGou et al., , 2016b)).PM2.5-bound PBDEs were also observed in public indoor environments like libraries, rail stations, department stores, offices, hospitals, and supermarkets in Taiwan (Chao et al., 2016).Among PBDEs, BDE-209 was the predominated PBDE congener in the Taiwanese microenvironments (Chao et al., 2014c;Shy et al., 2015;Chao et al., 2016;Chou et al., 2016;Gou et al., 2016aGou et al., , 2016b)).Our previous findings (Shy et al., 2015) showed that for the non-dietary daily intake of PBDEs in the indoor environment of the residential houses, indoor air inhalation intake was 1.88 and 1.29% of the total in male and female adults, respectively, and the house dust was the main exposure pathway of the non-dietary PBDEs intake.
To the best of our knowledge, this was the first study addressed the investigation of the POPs in the indoor environment of the gyms especially for PM2.5-bound PBDD/DF/DEs and these organobromines in indoor dust.It was studied that the increment of respiratory ventilation and the inhalation of air through the mouth rather than the nasal, which has particle-filtering apparatus, during the exercise was accelerated in the case of gym users with exposure to the environmental contaminants via inhalation or ingestion.Finally, this study also assessed the health risks of the gym users and employees after they were exposed to PBDD/DF/DEs.

Sampling Areas and Sample Collection
The gyms, which were the commercial gyms such as Les Mills or Equinox with various exercise devices (e.g., treadmill, exercise bike, and fitness equipment), including 4 gyms in Kaohsiung and 2 gyms in Pingtung were in southern Taiwan and the indoor dust, indoor PM2.5, and outdoor PM2.5 was collected between November 2018 and October 2019.All the gyms were nearby the road to gather the indoor dust and the proximities of 4 gyms (3 gyms in Kaohsiung and 1 gym in Pingtung) to the main traffic roads were chosen for the collection of samples for indoor dust and indoor PM2.5 and outdoor PM2.5.
The samples for indoor dust, indoor PM2.5, and indoor PM2.5 were simultaneously collected for two weeks in a gym.Indoor dust on the floor was collected by using a vacuum cleaner (Nilfisk Advance Euroclean UZ934 HEPA canister vacuum cleaner) equipped with an independent filter bag as described previously (Gou et al., 2016a).The indoor dust sample was immediately transported to the laboratory at the National Pingtung University of Science and Technology (NPUST) after the daily dust (5-10 g) was gathered in the gym during the sampling period of two weeks.Prior to the chemical analysis, each dust sample was sieved through 100 mesh (< 0.149 mm) by shaking for 10 min, with the removal of hair, trash, leaf, and possibly interfered materials before all the dusts were collected from a gym was pooled together.
The indoor and outdoor PM2.5 was followed by TEPA standard methods (NIEA A205.11C).PM2.5 samples were gathered on quartz fibre filters under the pre-heated condition at 600°C for 2 h prior to sampling.The PM2.5 filter paper was conditioned in an electronic desiccator at the temperature of 20-23°C and humidity of 30-40% and weighed by a balance with an accuracy of 0.1 mg before and after the collection of PM2.5 samples.The indoor PM sampler, such as Model 200-Personal Environmental Monitor™ (PEM™) for PM2.5 (2.5 µm) (MSP Corporation, Minnesota, USA) was used to collect PM2.5 indoors in the gyms based on our previous report with minor modification (Chao et al., 2016).After two-week indoor air sampling with the flow rate of 5.00 L min -1 (7.2 m 3 day -1 and 100.8 m 3 for two weeks), the PM2.5 samples were pooled together from the four continuous 3-day sampling samples by considering the collection of sufficient PM2.5-bound PBDD/DF/DEs for further chemical analysis.Collection of the outdoor PM2.5 referenced by our previous study was done by using a high-volume air sampler, such as SIBATA HV-1000R (Saitama, Japan) following U.S. EPA Reference Method TO9A (Chung et al., 2019).The outdoor PM2.5 in the gyms was gathered for 48 h with a flow rate of 800 L min -1 for a triplicate repeat during the twoweek sampling and three outdoor PM2.5 samplers were pooled for the further chemical analysis.The PM2.5 sample to be conditioned in an electronic desiccator after the sampling was immediately transferred into the lab at NPUST.

Extraction and Clean-up
Soxhlet extraction and multi-clean-up processes, including a multi-layered silica column, alumina column and activated carbon column, were followed based on the previous study (Wu et al., 2014) and our previous report (Chao et al., 2016;Gou et al., 2016a) with minor modification.In brief, the isotopes and surrogate standards were spiked in the PM2.5 filter paper for estimating the organobromines loss during the aerosol sampling process before PM2.5 sampling indoors and outdoors.To understand the PBDD/DFs and PBDEs loss in the extraction and clean-up procedures, the isotope internal standards were spiked into dust and PM2.5 samples were mixed well with toluene before extraction.PM2.5 filter samples and 20 g of indoor dust were extracted with 250 mL of toluene by the Soxhlet extractor for 24 h before evaporation.The elute was concentrated to 2 mL for the following multi-clean-up procedures to be eluted by 15 mL hexane, 25 mL dichloromethane/hexane (1/24, v/v), 5 mL toluene/methanol/ethyl acetate/hexane (1/1/2/16, v/v/v/v), and 40 mL toluene.The final elute was concentrated to approximately 1 mL in a vial.The elute was evaporated to near dryness by a gentle stream of gaseous nitrogen to quantity the final volume of 0.2 mL.
The analytical assurance and quality control (QA/QC) of the analytical method used in this study were referenced by U.S. EPA method 1614 and TEPA NIEA M803.00B to be briefly shown as follows.The blanks including field and lab blanks (i.e., solvent or glassware blanks) were regularly tested in each batch of the analyses.Recoveries of spiked blanks, prelabelled internal standards, surrogate, and precision and recovery (PAR) of all the relevant standards were within the acceptable criteria (70-130%).The limits of detection (LODs) were defined as the ratio of signal to noise (S/N) over 3 and method detection limits (MDLs) were obtained from LODs to be calculated as 2.5-5 folds of the estimated LODs.The limit of quantification (LOQ) was defined as the ratio of S/N over 10.MDLs values of PBDEs and PBDD/Fs ranged from 4.33-290 pg g -1 and 0.0310-25.6pg g -1 , respectively.
Measurements of PBDEs and PBDD/Fs below MDLs were recognized as zero for further statistical analysis.Descriptive analysis was used to determine the means and standard deviations (SD) of indoor dust, indoor PM2.5, and outdoor PM2.5.Mann-Whitney U tests were used to determine the differences in levels of PBDD/Fs and PBDEs on PM2.5 indoors and outdoors because PM2.5-bound PBDEs and PBDD/Fs were not normally distributed and the sample size was small.The correlations or sources of PBDEs and PBDD/Fs in the gyms were examined by the principal component analysis (PCA).For the assessment of risk in the gym users and the staff in the gyms, the Monte Carlo method was used to simulate the non-dietary intakes of PBDEs and PBDD/Fs via ingestion or inhalation.The critical values of HQs and Rs were set as 1.00 and 1.00 × 10 -6 .All statistical analyses were performed using Statistical Product and Service Solutions, version 12.0.
The indoor dust level of ∑12 PBDD/Fs was 7090 ± 4680 pg g -1 and its PBDD/Fs-TEQ level was 37.8 ± 13.7 pg WHO2005-TEQ g -1 in the gyms (n = 6).Several studies investigated and reported on PBDD/Fs in indoor dust (Ma et al., 2009;Takigami et al., 2009;Suzuki et al., 2010;Tue et al., 2010Tue et al., , 2013;;Gou et al., 2016a).In our previous study (Gou et al., 2016a), the dust PBDD/F levels in the normal and computer classrooms of the elementary schools in southern Taiwan were reported as 4.72-7.51and 3.60-6.28ng g -1 respectively, while the dust levels of PBDD/F-TEQs were 0.0401-0.0636ng WHO2005-TEQ g -1 in the normal classrooms and 0.0281-0.0474ng WHO2005-TEQ g -1 in the computer classrooms.A comparison between the present work and the previous work reported by us (Gou et al., 2016a), indicated that the indoor dust levels of PBDD/F WHO2005-TEQs (0.0378 pg WHO2005-TEQ g -1 ) in the gyms were lower when compared with those in the elementary schools from our previous study.Based on the current global data concerned with PBDD/Fs in indoor dust, it was suggested that the PBDD/F concentrations (7.09 ng g -1 ) in the gyms had a distinctly lower magnitude when compared with those from the floor dust in the e-waste plant of China (89.3-143 ng g -1 ) (Ma et al., 2009) and house dust in the e-waste sites of Vietnam (7.70-63.0ng g -1 ) (Tue et al., 2010) as well as the comparable magnitude with those from the indoor dust in the classrooms (1.46-6.39ng g -1 ) (Gou et al., 2016a), house dust in the US (0.33-150 ng g -1 ) and Japan (1.10-12.1 ng g -1 ) (Suzuki et al., 2010;Tue et al., 2013), indoor dust in the Japanese offices (1.62-31.3ng g -1 ) (Suzuki et al., 2010), and indoor dust in a Japanese hotel (0.740-15-4 ng g -1 ) (Takigami et al., 2009).Based on the distribution of PBDD/F profiles in PM2.5 and in indoor dust, it was observed that PM2.5-bound PBDD/Fs were also similarly distributed in the gyms indoors and outdoors, but the indoor dust did not have the similarity of PBDD/F profiles with indoor PM2.5 (Fig. S1).

Levels of PBDEs in the Gyms' Environment
PBDE levels on indoor and outdoor PM2.5 and indoor dust in the Taiwanese gyms are presented in Table 3. ∑14PBDEs in outdoor PM2.5, indoor PM2.5, and indoor dust in the gyms were expressed as Table 3. Levels of PBDEs on PM2.5 and indoor dust collected from the gyms.

Congeners
PM2.5 (pg m -3 ) Indoor Dust (pg g -1 ) Outdoor PM2.5 Mean ± SD (n = 4) Indoor PM2.   , and 2670000 ± 2330000 pg g -1 (2670 ± 2330 ng g -1 ), respectively.The results indicated that PM2.5-bound PBDEs and BDE-209 (4.85 pg m -3 ) in indoor PM2.5 was higher than that of those in outdoor PM2.5 (BDE-209: 4.80 pg m -3 ), however their difference was found to be not significant.Based on the obtained results, it was suggested that BDE-209 was the predominated congener consisting of 58.0% and 50.8% of the total PBDEs in outdoor and indoor PM2.5, respectively.The study conveyed that for indoor dust in the gyms, BDE-209 was also consisting of 70.4% of the total PBDEs.According to the homologue distribution of PBDEs in the gyms including outdoor and indoor PM2.5 and indoor dust as shown in Fig. S2, the distribution of PM2.5-bound PBDEs in the gyms indoors and outdoors was very similar, but PBDE homologues in indoor dust were different from those in PM2.5.
Based on the current global data in Table 4, it was claimed that not many studies were addressed PM2.5-bound PBDEs in the outdoor and indoor environment (Beser et al., 2014;Wang et al., 2014;Li et al., 2015;Xu et al., 2015;Chao et al., 2016;Deng et al., 2016;Liu et al., 2016;Xu et al., 2016;Guo et al., 2020;Tames et al., 2020).According to the current data, it is very difficult to conclude whether indoor levels of PM2.5-bound PBDEs are higher or lower than outdoor levels of PM2.5-bound PBDEs or indoor and outdoor PM2.5-bound PBDEs are highly correlated.The results of this present study conveyed that the data of PM2.5-bound PBDEs indoors and outdoors in the gyms were not lower when compared with those from the certain outdoor and indoor environments, such as residential houses, rural areas, and urban areas.
A large variation of indoor PBDE concentrations in the gyms was found in this study.Based on the current global data it was emphasized that indeed this was the first report of indoor PBDEs in the gyms.Indoor dust BDE-209 as shown in Fig. S2 was the most abundant congener among PBDEs in the gyms which was very similar to the other microenvironments, such as offices, daycare centres, houses, cars, classrooms, and the public places (Gou et al., 2016a;Brits et al., 2019;Jin et al., 2019;Lee et al., 2020;Al-Harbi et al., 2021;Klinčić et al., 2021).Although the results obtained in the present study were in the same order of magnitude when compared with the other microenvironments (offices, medical and dental clinics, the classrooms of the elementary schools, and residential houses), the pattern of PBDE homologue in indoor dust from the gyms was not similar to the other Taiwanese microenvironments according to our previous studies (Chao et al., 2014c;Shy et al., 2015;Chou et al., 2016;Gou et al., 2016a).It was demonstrated that the unique PBDE homologue pattern in indoor dust of the gyms might be due to three reasons: (1) the electronic equipment used in the gyms was quite different from the one used in other indoor environments; (2) the gyms we chose were established and opened within these 5 years whereas the microenvironments we used for our previous studies were opened before 2015 (Chao et al., 2014c;Shy et al., 2015;Chou et al., 2016;Gou et al., 2016a).( 3  d The workplaces in the present study were library, rail station, department store, office, hospital, and supermarket.latest announcement of persistent organic pollutants on Stockholm Convention in 2019 to list BDE-209 or commercial decaBDE mixture as Annexure A for elimination, the new building materials, electronic equipment, or decoration might be largely used for PBDE-replacement formulation of fire retardants in the gyms.

Correlations of PBDEs and PBDD/Fs in the Gyms' Environment
Fig. 1 shows the correlations of PBDD/Fs or PBDEs in the gyms indoors and outdoors after examination of PCA tests.Firstly, PM2.5-bound PBDD/Fs indoors and outdoors were examined by PCA to classify two groups as follows: ∑12PBDD/Fs, PBDDs, PBDFs, and certain PBDD/Fs in outdoor PM2.5 in a group and indoor PM2.5-bound PBDD/Fs including PBDDs, PBDFs, and certain PBDD/Fs in the other group (Fig. 1(A)) (percentage of variance (σ) = 94.9%).It was stated that outdoor PM2.5-bound PBDD/Fs including PBDDs, PBDFs, and certain PBDD/Fs were independent of those in indoor PM2.5, indicating that indoor PM2.5-bound PBDD/Fs were not associated with outdoor PM2.5-bound PBDD/Fs and it could also be explained as the different emission sources of PBDD/Fs between indoors and outdoors.Fig. 1(B) shows the statistically independent correlation of PM2.5bound PBDEs between indoors and outdoors and this result could thus explain the 94.9% of the data (σ = 94.9%).It clearly showed that the outdoor PM2.5-bound PBDEs were not related to indoor PM2.5-bound PBDEs.According to our previous study (Gou et al., 2016b), airborne PBDEs from Di-BDEs to Nona-BDEs including gas and particle-phase were significantly independent both indoors and outdoors of the vehicle dismantling factories, but the outdoor BDE-209 was significantly correlated with indoor PBDEs from Hexa-BDEs to Deca-BDE.

Risk Assessments in the Gyms' Environment
The risk factors and the exposure levels were used to assess the risk assessments, such as daily intakes, non-cancer risks (HQs), and cancer risks (Rs), via non-dietary routes based on two exposure scenarios including the gym users and employees (Table 1 and Table 5).According to Table 1, the active inhalation rates were 1.69-2.10times higher than the inactive inhalation rates in Taiwanese.The values of daily intakes, HQs, and Rs for the gym users and staff through ingestion of indoor dust were found to be notably higher than that of those through inhalation of PM2.5 as shown in Table 5.The Korean study also announced that the contribution of dust ingestion was still main after the estimation of multiple-exposure assessment of PBDEs (Lee et al., 2020).The summary and individual calculation of HQs and Rs from certain PBDEs and ∑PBDD/Fs for the gym users and the employees were still under the critical values of HQ = 1.00 and R = 1.00 × 10 -6 .Thus, it was concluded that no non-cancer and cancer risks were found in the gym users and employees through indoor PM2.5 and indoor dust in the gyms.A comparison with our previous report revealed that the schoolage children at the age of 7-9 consumed between 9.40 × 10 -12 and 1.34 × 10 -11 mg kg b.w. - day -1 from the classroom dust (Gou et al., 2016a).However, in the present study, the estimated daily intakes of ∑PBDD/Fs (1.29 × 10 -13 -5.42 × 10 -13 mg kg b.w. - day -1 ) for the adults through the indoor dust of the gyms were found to be low.The gym-dust daily intakes of PBDEs for the male and female employees were found to be 1.61 × 10 -7 and 2.11 × 10 -7 mg kg b.w. - day -1 (0.161 and 0.211 ng kg b.w. - day -1 ), respectively.Several studies reported the estimated daily intakes via indoor dust for adults as BDE-209 of 0.2 ng kg b.w. - day -1 in the South African dust (Brits et al., 2019), PBDEs of 0.56-0.57ng kg b.w. - day -1 in the Korean dust (Lee et al., 2020), PBDEs of 0.557-0.727ng kg b.w. - day -1 in the Taiwanese dust (Shy et al., 2015), and PBDEs of 0.283-0.311ng kg b.w. - day -1 (Xu et al., 2015).The results represented that the data of HQs and Rs in the present study was consistent with our previous study (Shy et al., 2015) and also the Korean report (Lee et al., 2020) suggesting that the non-cancer and cancer risks have not occurred.

Table 1 .
The parameters of risk assessments in this study a,b .
a DIdust (ng kg b.

Table 2 .
PBDD/F concentrations on PM2.5 and indoor dust collected from the gyms.

Table 5 .
Risk assessment of PBDEs and PBDD/Fs in the indoor environment of the gyms.