Levels and health risk assessments of Phthalate acid esters in indoor dust of some microenvironments within Ikeja and Ota, Nigeria

The levels, profiles of Phthalate acid esters (PAEs) and their associated health risk in children and adults using indoor dust samples were assessed from nine (9) microenvironments in Nigeria. Six PAEs congeners were determined using Gas Chromatography–Mass Spectrometry and the human health risk assessments of PAEs exposure to children and adults were computed using the United States Environmental Protection Agency (USEPA) exposure model. The mean concentrations of the total PAEs (Σ6PAEs) in indoor dust across the study locations ranged from 1.61 ± 0.12 to 53.3 ± 5.27 μg/g with 72.0% of di-n-octyl phthalate (DnOP) as the most predominant contributor of PAEs in sample locations B, C, D, E, F and G. PAEs estimated daily intake results exceeded the USEPA value of 20 and 50 kg/bw/day for children and adults respectively in some locations. Non-carcinogenic risk exposure indicated no risk (HI < 1), while the carcinogenic risk was within the recommended threshold of 1.00 × 10–4 to 1.00 × 10–6 for benzyl butyl phthalate and bis-2-ethylhexyl phthalate. From our findings, lower levels of PAEs were observed in locations with good ventilation system. Also, the human health risk evaluation indicated indoor dust ingestion as the dominant exposure route of PAEs for both children and adults, while the children were at a higher risk of PAEs exposure. To protect children susceptible to these endocrine-disrupting pollutants, soft vinyl children’s toys and teething rings should be avoided. Appropriate policies and procedures on the reduction of PAEs exposure to humans should be enacted by all stakeholders, including government regulatory agencies, industries, school administrators and the entire community.

www.nature.com/scientificreports/ every two hours throughout the day from different sections and a composite sample for each location was used for the analysis. A total of nine indoor dust samples and six control samples were collected. After sampling, the dust particles were transferred to aluminium foil and sealed in a ziploc bag, then transferred to the laboratory 12 .
To remove unwanted substances and homogenize the sample, the collected dust samples were sieved with individual < 150 µm mesh and wrapped in aluminium foil, then stored in an airtight ziploc bag in the freezer at − 20 °C prior to further analysis [2][3][4]12 .
Sample extraction and clean up. Dust sample (100 mg) were measured with an analytical weighing balance and quantitatively transferred to a 10 mL glass tube. Exactly 3 mL of the extraction solvent (hexane and acetone) in the ratio of 1:1 was introduced into the glass and placed on the vortex mixer for 20 min at 1500 rotation per minute (rpm) to re-suspend and agitate the mixture. Afterwards, the agitated mixture was centrifuged for 20 min at 1500 rpm. This cycle was repeated two more times for better extraction efficiency, and the pool of extract was concentrated to 1 mL with the aid of a rotary evaporator and reconstituted with 2 mL of methanol 5,12 . After re-solubilisation, clean-up of samples was performed on a 30 mL glass chromatographic column packed with 10 g of activated silica gel (60-200 mesh size), and 1 g of activated anhydrous sodium sulphate (Na 2 SO 4 ) was introduced into the packed column for absorption of moisture. The packed column was pre-eluted with n-hexane at a 2 mL/min flow rate. The extract was carefully transferred into the column, and 2 mL of n-hexane was used to thoroughly rinse and transfer the extract down the column. The column was eluted with 15 mL of n-hexane, allowing the non-polar fraction of the extract in the column to be collected and discarded. After that, 20 mL of acetone was added to the column to elute the targeted analyte. The eluent was collected and transferred to a 100 mL round bottom flask and pre-concentrated to 1 mL with a rotary evaporator. The concentrated extract was rinsed with 2 mL of methanol before being transferred to an aluminium-wrapped GC glass vial for instrumental analysis. This sample preparation and extraction methods are consistent with the USEPA 8061A method for PAEs analysis.  C dust = concentration of PAEs compound in dust particles (µg/g); f = ingestion rate of dust (g /day); m = body mass (kg) IR inhalation = dust inhalation; SA = area of dermal exposure; ABF = fraction of dermal absorption; AF = dust dermal adherence factor; BW = body mass index expressed in kilogram; AT = average time (d); AT = ED × 365 (non-cancer risk); AT = LT × 365 (cancer risk); ED = exposure period (y); EF = exposure frequency (d/y); CF = conversion factor (1.0 × 10 -6 kg/mg); PEF = particle emission factor (1.36 × 10 −9 m 3 /kg); The hazard quotient (HQ) and hazard index (HI) was used to calculate the non-carcinogenic risk of indoor dust PAEs exposure as shown in Eqs. (5) and (6); where RfD is the independent phthalate compound reference dose value (mg/kg/d). The reference dose is defined as the maximum daily risk that can be imposed on the exposed population 60 . It is used as an indicator or measure to indicate the possibility of a serious health effect over the course of a person's life. While HQ signifies the health risk of each PAE to humans through the diverse exposure routes, and i signifies the various exposure routes (dermal absorption, ingestion and inhalation). HI value greater than 1 indicates that adults and children exposure to PAEs might result in a non-cancer risk. DMP, DEP, DBP, and DnOP are non-carcinogenic, whereas DEHP and BBP are carcinogenic to humans. The carcinogenic route was evaluated using the lifetime average daily exposure doses (mg/kg/d) shown in Eqs. (7) and (9) The carcinogenic risk (CR) was analyzed using Eq. (10) www.nature.com/scientificreports/ CSF stands for cancer slope factor, inclusive of CSF oral , CSF ingestion , CSF inhalation , and CSF dermal . It is a measure of the likelihood or possibility of an individual being infected by cancer due to oral intake of the chemical 61 . To calculate the possibility of cancer risk via inhalation and dermal absorption exposure pathway, it was estimated that CSF inhalation and CSF dermal are equal to CSF oral 10 . The evaluated values from cancer risk calculation are considered negligible if CR is below 10 -6 . The permissible limit is 10 -6 to 10 -4 , while values higher than 10 -4 are considered a threat to health 3,17 . Table 2 shows the parameters used to calculate the human risk assessment.
Quality assurance/quality control. The suitability of the analytical and instrumental processes were ensured by avoiding all forms of plastics throughout the laboratory procedures. The glassware used for the work was thoroughly oven-baked for 6 h at 120 °C in line with the USEPA 8061A method for the treatment of glassware and rinsed with acetone and hexane thereafter. The instrumental system was calibrated by preparing an equal volume of the external and internal standards in 5 mL methanol. A five-point calibration curve was plotted between the areas of peak and calibration concentrations resulting in an overall linearity range of 0.996-1 for the six PAEs compounds. To test the suitability of each of the analytical procedures, blank samples were prepared, namely, field blank, sieve blank, method blank, and reagent blank. The six phthalate compounds for the study were found in the blank samples and deducted from the sample concentrations to obtain the exact value of PAEs in the sample set. The method precision from the duplicate blank was 11.14%. This value is within the expected range of < 20% and is consistent with other reported works on PAEs in indoor dust particles such as 2,12,59,62 . The recovery study was done to ascertain the extraction efficiency, and the value ranged from 70.9 to 72.6%. This range is in agreement with the expected range of 70-120% for PAEs recovery study in indoor dust particles 12,29 . The computation of the limit of detection (LOD) and limit of quantification (LOQ) was done using a signal-tonoise ratio of 3:1 and 10:1, respectively. The LODs and LOQ of PAEs ranged from 0.005 (DEP) to 0.025 (DBP) and 0.016 (DEP) to 0.083 (DnOP), respectively, as shown in Table S2.
Statistical analysis. Descriptive statistics were done using Microsoft Excel 2013 and Origin Pro 8.5. The Shapiro-Wilk test was applied to check the normality of the data. The data were not normally distributed as such, the non-parametric Kruskal-Wallis test was used to compare PAEs concentration differences between the various microenvironments. A p-value < 0.05 was considered significant.

Results and discussion
Concentrations of PAEs in indoor dust. The concentrations and detection frequency of PAEs from the sampled sites (Ikeja and Ota) are shown in Figure S1. There was a 100% detection frequency for the six PAEs compounds. The concentrations of total PAEs compound in each location are shown in Fig. 2 and Table S3. DnOP was the most abundantly detected compound in all nine samples, followed by DMP and DBP. The mean  Table S1. Materials found in indoor environments like medical devices in the children's ward in a hospital, floor tiles, wall covering decorative items, chairs, tables, children's play toys and articles in preschool centres, television sets, adhesives, wires and cables have DnOP as the major plasticizers used in their production, which led to DnOP being the dominant pollutants found in seven out of the nine sampled locations. The highest level of Σ 6 PAEs was found in indoor dust from location G representing a paediatrics' ward in the hospital, with a mean value of 53.3 ± 5.27 μg/g, followed by two preschool centres in locations B and F with mean concentrations of 33.6 ± 3.10 and 15.3 ± 0.62 μg/g. The high level of PAEs in these locations is attributed to inadequate ventilation as a result of little or no exposure to sunlight because the window blinds were permanently closed, thereby preventing the free flow of air and the penetration of sunlight 63,64 . Also, Liao et al. 65 findings observed reduced levels of indoor PAEs, using mechanical ventilation and air cleaner than with only natural ventilation. Hence, they recommended the use of mechanical ventilation or natural ventilation along with air cleaners to reduce the concentration of PAEs indoors. The lowest Σ 6 PAEs concentrations were from locations A (2.34 ± 1.86 μg/g) and C (1.61 ± 0.12 μg/g) day-care centres in Ikeja, Lagos, Nigeria. Both locations were observed as the most ventilated and spacious locations among all the microenvironments. Furthermore, levels of total PAEs reported in this work were lower than those reported in other regions, such as Greece 797 μg/g 66 , China 592 μg/g 12 , Saudi Arabia 663 μg/g 10 , and USA 214 μg/ g 55 . The gas/particle partitioning of PAEs is integral to their indoor fate and health risks. However, the effects of indoor environmental parameters such as temperature, humidity and the partitioning of PAEs between air and particles is rarely known 64,67 . In addition, it is worth stating that the particle-air partition coefficient and vapour pressure properties affect the existence of PAEs congeners in both phases of individual PAEs. The equilibrium typically favours the gas phase for PAEs with higher vapour pressure and a lower particle air coefficient, while the particle phase is in parity with PAEs with lower vapour pressure and a higher particle air coefficient 29,55 . The high concentration of DEP and DBP in the hostel and chapel indoor dust might be connected to their solvent-carrying properties in the production of several household and skin care materials such as body lotion and deodorants, insecticides and pesticides, plastics and sealants during production in spite of the high volatility of DEP and their reported existence predominantly in the gas phase than dust phase 12 . The location with the highest concentration of DEHP was found in a preschool centre with PVC flooring. This result is in conformity with studies carried out by 12,29,68,69 . As they all found a significant correlation between PVC flooring and the concentration of DEHP in indoor dust. The variation in the mode of sampling, extraction and clean-up procedures, climatic factors, the lifestyle of the occupants, indoor maintenance, flooring type, building characteristics, and decorative wall items are all factors contributing to the levels of PAEs in indoor dust 11,29,[68][69][70] . Apart from different analytical procedures, the lack of inter-laboratory comparisons of PAEs results in dust and the inability to analyze with standard reference material makes the comparison of results difficult. Consequently, a high variation was observed between concentrations determined in different study locations and similar neighborhoods in the same country.
One-way ANOVA of the parametric method could not be used since the data were not normally distributed 58,71,72 . Hence, the Kruskal-Wallis test of the non-parametric method was used to compare PAEs concentration differences between the various microenvironments. The result presented in Table 3 showed that there was a statistically significant difference between PAEs compounds and microenvironments since the p-value (0.001) < 0.05 level of significance. Kruskal-Wallis' mean rank for each of the PAEs compounds is presented in Table 4.
The Σ 6 PAEs concentrations in indoor dust from selected reports are presented in Table 5. Globally, PAEs are among the identified indoor chemical pollutants that thrive in children learning environments 69  www.nature.com/scientificreports/ studies with similar trends of result 16,42 . As indicated in Table 5, the highest DEHP concentration of 3030 ug/g reported in South Korea far outweighs the 0.018-1.41 ug/g concentration obtained in this study. Since this is the first study of PAEs in indoor dust on the shores of Africa, there was no basis for comparison within the continent. Human health risk estimation. Tables 6 and 7 present the results of the human health risk assessments for this study. Table 6 reveals the estimated daily intake (EDI) of total PAEs in each microenvironment. Location G representing the children's ward in the hospital, had the highest concentration of EDI, 691 and 86.4 μg/g, respectively, for both children and adults. Exception of locations A and C, individual PAEs compounds recorded in other investigated locations (B, D, E, F, G, H, and I) were above the threshold of 20 kg /bw/day for children. Furthermore, the EDI for adults were above the stipulated limit of 50 kg /bw/day as estimated by USEPA in location B and G. This indicates that the daily exposure to PAEs compounds in these locations was very high, thus harmful to the occupants of the environment. The results of the average daily dose (ADDs) of six investigated PAEs via the three major exposure routes in indoor dust for children and adults are indicated in Table 7. PAEs total intake of ADDs (ADD ing + ADD der + ADD inh ) from indoor dust for children and adults was found to be 3.37 × 10 -1 and 1.70 × 10 -4 mg/kg/d, respectively. This result implies that children are more susceptible to PAEs in indoor dust than adults. Children's ADDs from ingestion, dermal absorption, and inhalation of indoor dust were 9.84-, 6.31-, www.nature.com/scientificreports/ and 6.74-times higher than those of adults. This may be attributed to the frequent hand-to-mouth activities, increased exposure duration and children's lower body weight. The ADDs of PAEs for children and adults via dust ingestion exposure were significantly higher than other exposure pathways in the study locations. The ADDs of PAEs for children and adults through dust ingestion were 1.46-1.56 orders higher than that of dust inhalation and dermal adsorption. Similar results have been reported and documented by other researchers and reviewers 12,76,78-80 .  Figure 3. Composition profiles of PAEs in indoor dust. www.nature.com/scientificreports/ The degree of PAEs exposure in adults was assessed via two approaches, the carcinogenic and the noncarcinogenic routes, as shown in Table 7. The findings revealed that the non-carcinogenic risks for adults and children were less than the stipulated limit of (HI > 1). Humans are considered exposed to non-cancer risks if the value of the hazard index (HI) is greater than 1. All the HI values were below one (1), indicating the absence of non-carcinogenic risk. However, the HI of DEHP (3.07 × 10 -2 ), DBP (2.67 × 10 -3 ) and DnOP (1.40 × 10 -3 ) via indoor dust ingestion were close to the threshold value. Also, the hazard quotient of non-cancer risk for both children and adults for the six priority PAEs compounds in the sampled locations were in the ascending order of DEHP > DnOP > DBP > BBP > DEP > DMP. DEHP (3.07 × 10 −2 ) and DnOP (2.58 × 10 -4 ) contributed the most to the degree of hazard index among the nine study locations for children and adults, respectively. Table 7 indicates that the carcinogenic risk for BBP and DEHP in indoor dust for adults through the three key predominant routes were 1.11 × 10 -9 and 2.42 × 10 -10 , respectively. These values were below the carcinogenic threshold limits of 1.00 × 10 -4 to 1.00 × 10 -6 for PAEs exposure in humans. As a result, the carcinogenic risk of human exposure to BBP and DEHP in indoor dust was negligible. This is consistent with other reported research works on PAEs in indoor dust for BBP and DEHP, respectively: Li and Wang 61  However, the three predominant exposure routes for children were significantly higher than those of adults in both carcinogenic and non-carcinogenic risk estimation. This increase may be attributed to the explorative nature of children, their frequent hand-to-mouth activities, lower body weight and higher exposure duration. On the contrary, the dermal absorption and ingestion exposure routes were relatively similar and comparable for adults. This study's results clearly highlight dust ingestion as the predominant route of PAEs in children and adults. Also, the total exposure of children to PAEs via ingestion route in indoor dust particles, as shown in Fig. 4, indicated that DEHP had the highest contribution of 87.2%, followed by DBP (7.61%) and DnOP (4.0%). DMP, DEP and BBP made an insignificant contribution of 0.01%, 1.23% and 0.60%, respectively. Similarly as shown in Fig. 5 for adults, DEHP (54.6%) contributed the highest, followed by DBP (29.9%) and DnOP (9.7%). While DMP, DEP and BBP had 0.01%, 3.25%, and 2.51% contributions, respectively. This study's results corroborate with other reported work 12,29,69 . Conclusion and recommendations. This study is the first to report the occurrence of six predominant PAEs compounds, namely DnOP, DBP, DEP, BBP, DMP and DEHP and their human health risk assessments using indoor dust as the environmental monitoring tool. Fifteen samples were collected from nine different indoor microenvironments in Ikeja, Lagos and Ota, Ogun States, Nigeria. DnOP was the most predominant contributor of PAEs, in indoor dust except for locations A, H and I, where DEP was the dominant contaminant. The mean levels of Σ 6 PAEs in indoor dust ranged from 1.61 ± 0.12 to 53.3 ± 5.27 μg/g across the different microenvironments. The highest level of Σ 6 PAEs in the microenvironments was in the order of location G (53.3 ± 5.27 μg/g) > location B (33.6 ± 3.10) and > F (15.3 ± 0.62 μg/g). The highest contributions of DnOP and DEP could be linked with plasticizers and personal care products containing DnOP and DEP, in addition to the contribution from flooring types, cleaning habits, and building materials in the indoor environment. The human health risk evaluation indicated ingestion of indoor dust as the dominant exposure route of PAEs for both children and adults, while the children were at a greater risk of PAEs. The non-carcinogenic risk exposure was below the threshold limit of (HI > 1), indicating no risk. Also, the result of the carcinogenic risk exposure of www.nature.com/scientificreports/ DEHP and BBP from indoor dust by adults was within the recommended threshold of 1.00 × 10 -4 to 1.00 × 10 -6 . More attention should be paid to the indoor environment, and appropriate policies and procedures should be enacted by all stakeholders in order to reduce PAEs exposure to humans. A major limitation is the difficulty in obtaining the required permissions to access preschool centres and financial constrains in the analysis of samples thus resulting in the small sample size reported in this study. As a result, studies with a large pool of samples are required for a comprehensive understanding of the occurrence and distribution of PAEs in children's microenvironments and the availability of research grants specifically for PAES analysis in children's microenvironments is required to conduct a better comparative analysis of result in Africa. Also, future studies should include a comparative analysis of PAEs in both indoor and outdoor environments, source apportionment to ascertain the primary sources of PAE emissions and the environmental fate of PAEs in indoor environments. This will provide a scientific and robust understanding of the occurrence of PAEs in indoor environments.

Data availability
The data supporting the findings of this study are available on request from the corresponding author.