Phthalate levels in Chinese residences: Seasonal and regional variations and the implication on human exposure

: Indoor pollution of manmade semivolatile organic compounds (SVOCs) such as phthalates are a growing threat to human health. Herein we summarize the dust-phase phthalate concentrations in Chinese residences reported from 2011 to 2021 and simulate corresponding airborne concentrations based on equilibrium models. The simulation considers seasonal and regional variations in indoor temperature and PM 2.5 concentration, in contrast to the common practice of using constant values. Results show that variations in these two environmental factors lead to up to ten-and six-fold variations in the monthly median gas-and particle-phase concentrations of phthalates, respectively, in residences in individual climate zones. For higher-vapor-pressure species di-n -butyl phthalate and di-isobutyl phthalate, the resultant seasonal and regional variations in aggregate non-diet intake can reach six-and three-fold, respectively. These results have important implications on exposure assessment of SVOCs and epidemiological evaluation of their health effects.


Introduction
Anthropogenic chemical pollution is recognized as a growing peril and has the potential to pose catastrophic risk to humanity [1,2].Many known hazardous synthetic chemicals, such as phthalates and polybrominated diphenyl ethers, are semivolatile organic compounds (SVOCs), which can migrate from the source materials to the surrounding air, suspended particles, settled dust, and other environmental media [3][4][5].These manmade SVOCs can enter the human body through various routes of exposure, including inhalation, skin absorption, and both dietary and non-dietary ingestion [6].The exposure primarily occurs indoors, where modern people spend about 90% of their time [7] and many SVOC-containing products are used [8].
Indoor air and indoor dust are two major environmental media contributing to human exposure to SVOCs [6,8].For less volatile SVOCs, ingestion of indoor dust is the primary non-diet exposure route, especially for young children who often crawl and play on the floor [9][10][11].For more volatile SVOCs, inhalation and skin absorption of the indoor air are the primary routes of exposure [10][11][12].Moreover, indoor air exposure is suggested to play a crucial role in the development and exacerbation of allergic diseases [13], since the associated symptoms manifest in the upper and lower airway and skin, which represent the major epithelial interface between the air and the body.Assessing human exposure to SVOCs thus requires careful characterization of SVOC concentrations in both indoor dust and indoor air.
The current approach for indoor SVOC exposure assessment typically relies on the analysis of dust samples [10,[14][15][16][17].The airborne concentrations are then estimated from measured dust-phase concentrations based on equilibrium models [10,14].In many epidemiological studies, SVOC concentrations in house dust are also directly taken as a proxy of the corresponding integrated non-diet exposure [18][19][20].While these methods reduce the complexity of sample collection, as it is more challenging to collect gas-and particle-phase SVOCs than to collect dust, their limitations still need to be investigated.
Temperature and concentration of particulate matter (PM) are two key environmental factors regulating the partitioning of SVOCs among dust, gas, and particle phases, and they are typically set as fixed values in the models which predict indoor airborne concentrations of SVOCs from the corresponding dust-phase concentrations [9,21].The use of concentrations in house dust as a proxy of environmental exposure to SVOCs is also based on the assumption of consistent temperature and PM concentrations across different residential environments.Recent observational studies have, however, revealed that the indoor airborne concentrations of SVOCs are strongly subject to variations in temperature and PM concentrations, despite that their concentrations in house dust often stay relatively constant [22,23].For example, time-resolved measurements in an office illustrated that the indoor airborne concentration of di-n-butyl phthalate (DnBP) and di-isobutyl phthalate (DiBP) could double with each temperature increase of approximately 4°C [24], and similarly strong temperature dependence was also observed in some other indoor studies for other SVOCs [25][26][27].PM concentration mainly influences the particle-phase SVOC concentration, and strong correlation between airborne SVOC concentration and PM concentration has been observed for less volatile SVOCs (i.e., those substantially partition in the particle phase) in multiple studies [25,27,28].These findings indicate the importance to consider variations in indoor environmental conditions when estimating the airborne SVOC concentrations from corresponding dust-phase concentrations in the exposure assessment.
Phthalates are among the most commonly found man-made SVOCs in the environment and have been extensively studied [8,29].The species with lower vapor pressures, such as di-2-ethylhexyl phthalate (DEHP) and butyl benzyl phthalate, are used as plasticizers in a wide range of polyvinyl chloride products, whereas those with higher vapor pressures, such as DnBP and DiBP, are often used as carrier solvents in cosmetics and personal care products.Human exposure to phthalates has been associated with many adverse health effects such as impaired reproduction development [30], male infertility [31], and children asthma and allergic symptoms [20].China is currently the most prominent market of phthalates in the world, accounting for over half of global consumption in 2020 [32].Several studies have reviewed phthalate concentrations in indoor dust reported in the literature and provided estimates for overall non-dietary phthalate exposure for Chinese [9,21].In these studies, the airborne concentrations were estimated assuming a fixed indoor temperature and a fixed indoor PM concentration across the entire country.However, recent indoor surveys have shown that temperature and PM concentration exhibit large seasonal and regional variations in residences in different climate zones in China.The monthly mean indoor temperature can differ by 13°C within a year in some regions [33], and the monthly mean indoor PM 2.5 concentration can differ by 90 μg/m 3 [34].Estimating the airborne concentrations based on fixed temperature and PM 2.5 concentration can thus potentially bring in large uncertainties in the exposure assessment.
This study aims to provide more accurate estimates of phthalate pollution levels and their non-dietary exposure in Chinese homes by accounting for seasonal and regional variations in environmental parameters.We first summarized measurements of dust-phase phthalates in the last ten years in Chinese indoor environments with a focus on residences, and then estimated the corresponding airborne concentrations based on previously reported seasonal variation in indoor temperature and PM concentrations in residences in different climate zones.The temporal and spatial variations in airborne concentrations of phthalates in homes in China were revealed and the impact on exposure assessment was evaluated.

Compilation of literature on dust-phase phthalate concentrations
We conducted a literature search for phthalate concentrations in settled dust using the "Web of Science (WOS)" database and employing the following keywords: "phthalate"OR "PAE" OR "semivolatile organic compound" OR "SVOC" and "indoor" OR "building" OR "dwelling" OR "house" OR "home" OR "apartment" OR "office" OR "school" OR "dormitory" OR "university" OR "campus".In total, 1659 articles published prior to January 2022 were obtained.Nearly eighty percent (1319) were published between 2011 to 2021.Thus, we limited our focus to articles published in the last ten years to capture the recent phthalate pollution levels.As the next step, we screened the title and the abstract of all these papers, to select those containing original indoor dust measurement data in China.In this process, review articles, commentary articles, and conference abstracts were excluded.We found most of the data were measured in three types of microenvironments, including residences, offices (stores), and schools, and only a few papers targeted on other microenvironments such as factories, laboratories, hospitals, and vehicles.We thus narrowed our focus to measurement data in residences, offices, and schools.This approach reduced the initial list of articles to 57.Next, we carefully reviewed the full texts of the 57 articles, and excluded those with small sample sizes (<5), duplicate measurements, or only graphic representations of phthalate concentrations.As a result, 27 articles with relevant phthalate measurement data in China were identified, which formed the basis of our data synthesis.
Medians (or geometric means) of the phthalate concentrations in settled dust reported in individual articles were taken for further statistical analysis.Here we used the medians instead of the means because that concentrations of indoor air pollutants including phthalates usually follow a lognormal distribution [21,35,36].The medians (or geometric means) were not reported in four articles [37][38][39][40].We estimated the median values in three articles from the reported means and standard deviations based on Eqs.(S1) and (S2) in the Supplementary information [37][38][39].One article only reported mean values and thus was eliminated in this step [40].Therefore, 26 studies that contain measurement data of dust-phase phthalate concentrations were selected for further analysis.
Among these studies, the number of measurements conducted in residences was significantly larger than that in other building types.The total sample size was 2006 in residences, nearly 28 and 4 times higher than that in office buildings (70) and schools (434), respectively.Our analysis thus focuses on measurements in Natl Sci Open, 2023, Vol.2, 20230011 residences.Results from other types of buildings are presented in the Supplementary information.Moreover, DEHP, DnBP, and DiBP are the most frequently detected phthalate species in indoor settled dust and are selected for further analysis.

Temperature and PM 2.5 concentrations in Chinese residences
The observational data of indoor temperature and PM 2.5 concentrations reported in recent surveys in residences in China are used in the current analysis.The temperature data are obtained from a yearlong measurement project spanning from 2018 to 2019 [33], and the PM 2.5 data set is from a similar project from 2016 to 2017 [34].In both studies, over 100 homes were selected for continuous measurements, and the results were summarized by climate zones according to the thermal design code for civil buildings in China (GB50176-2016) [41].That is, the whole country is divided into five climate zones: severe cold (SC), cold (C), hot summer and cold winter (HSCW), mild (M), and hot summer and warm winter (HSWW).The monthly statistics of indoor temperature and PM 2.5 concentration in each climate zone are used to predict the corresponding airborne phthalate concentrations.

Prediction of airborne phthalate concentrations from dust-phase concentrations
Assuming equilibrium partitioning between the gas phase and the settled dust, the gas-phase SVOC concentration C g (μg/m 3 ) can be estimated from the dust-phase concentration C d (g/g) as follow [4]: where K d (m 3 /μg) is the dust-gas partition coefficient.
Similarly, the particle-phase concentration C p (μg/m 3 ) can be estimated from the gas-phase concentration C g assuming equilibrium partitioning [42]: where K p (m 3 /μg) is the particle-gas partition coefficient, PM (μg/m 3 ) is the particle mass concentration.Both K d and K p are temperature-dependent.The temperature dependence can be described using the van't Hoff equation [36,43]: where ∆H d_g (J/mol) and ∆H p_g (J/mol) are the phase change enthalpies from the dust to the gas and from the particle to the gas, respectively, K d (298 K) and K p (298 K) are the dust-gas and particle-gas partition coefficient at 298 K, respectively.Eqs. ( 1)-( 4) allow for predicting gas-and particle-phase SVOC concentration C g and C p from the dustphase concentration C d at a given temperature and PM concentration.To use these equations, K d (298 K) and K p (298 K) can be estimated using empirical equations.Wei et al. [44] compiled the estimates of K d (298 K) and K p (298 K) from 29 published articles for commonly occurring SVOCs, and herein we took the median of Natl Sci Open, 2023, Vol.2, 20230011 all estimates for individual compounds.Meanwhile, ∆H d_g and ∆H p_g of a compound can be approximated using its vaporization enthalpy ∆H vap [36].Values of K d (298 K), K p (298 K), and ∆H vap for DnBP, DiBP, and DEHP used in this study are presented in Table S1.The method and parameterization adopted herein to predict C g and C p from C d have been previously validated using two observational studies that provided simultaneous measurements of C g , C p , and C d [36].
Monte Carlo simulation was used to characterize the uncertainties of the indoor airborne concentrations predicted from the indoor dust-phase concentrations in individual climate zones.Values of all three input parameters, including C d , T, and PM, were sampled from the probability density functions of their individual values.The ensemble of median C d values reported in individual surveys was approximated by a log-normal distribution (as presented in the next section), and the geometric mean and geometric standard deviation were estimated from the collected data.The indoor temperature in each climate zone was assumed to follow a normal distribution and the indoor PM 2.5 concentration was assumed to follow a log-normal distribution based on observational results [33,34,36]; the corresponding distribution parameters were acquired from the measurement data previously reported [33,34].

Exposure assessment
The aggregate non-dietary intakes of phthalates in residences were estimated following the methods used in the literature [11,21,45].Inhalation, dust ingestion, and dermal absorption through direct air-through-skin were considered.Given that the current study focuses on the influence of temporal and spatial variations in temperature and PM concentration on the phthalate exposure assessment, we took the exposure of male adults in the age of 18 to 64 as an example.Detailed methods and parameters used in exposure assessment are shown in Tables S1-S3.As a further note, the uncertainties for non-dietary intake were also characterized by Monte Carlo simulations, considering the uncertainties in the reported dust-phase concentrations and the simulated airborne concentrations.

Overview of dust-phase phthalate concentrations in residences in China
Table 1 summarizes median concentrations of DEHP, DnBP, and DiBP in settled dust surveyed in residences in China.Each line represents a survey (covering at least 5 residences) conducted in a specific geographic region.In some studies, separate results were reported for two seasons in one region, often with different sample sizes.We take them as separate records.There are in total 37 records in Table 1, corresponding to a total sample size of 2006 residences.All surveys were conducted in urban areas, mostly in a specific city and some in multiple cities.Thus, the data reported herein only represent urban residential environments.As presented in Table 1, there are 6, 10, 13, and 6 surveys conducted in the SC, C, HSCW, and HSWW climate zones, respectively, while no survey focused on the M zone.In addition, there are two nationwide surveys covering all the climate zones [46,47].
Figure 1 presents the box-whisker plots of median dust-phase concentrations obtained from surveys conducted in individual climate zones in China as well as across the entire country, for DEPH, DnBP, and DiBP, respectively.Due to the lack of data, the M zone is not included in the analysis.For all three phthalates, the median concentrations vary by two orders of magnitude among the surveys conducted across the whole country and by one order of magnitude among surveys in most individual zones.The wide variation in median concentrations reported by different surveys might reflect large differences in source strengthens of phthalates in residences, which are determined by factors such as building and furnishing materials, household products, and living habits of occupants [65].Meanwhile, lack of standardized sampling and a The median values were calculated from the arithmetic mean and standard deviation.b Not measured.c The average of median concentrations measured in the living room and bedroom.d Below limit of detection (LOD), and LOD / 2 is used instead [11].
Natl Sci Open, 2023, Vol.2, 20230011 analysis procedures regarding places where the settled dust is collected, tools used (brush or vacuums, types of vacuums) [46,52], size of sieves to remove large particles [50,53], and QA/QC for phthalate measurements, might also contribute to the variation of the data [65].The latter factor might explain systematic differences in phthalate concentrations between some studies (indicated by colors in Figure 1).To avoid amplifying the potential bias of individual studies, in the following analysis we consider equal weight for each record in Table 1 regardless of the corresponding sample size.Despite the order-of-magnitude differences in median dust-phase phthalate concentrations reported in individual surveys, the central tendencies in differing climate zones are comparable.The zonal medians range from 328 to 582 μg/g for DEHP, 45.4 to 174 μg/g for DnBP, and 17.3 to 55.0 μg/g for DiBP.For DiBP  1.
Natl Sci Open, 2023, Vol.2, 20230011 and DEHP, there is no significant difference in dust-phase concentrations across the four climate zones.For DnBP, the concentrations in the C, SC, and HSCW zones are comparable, but the concentrations in the HSWW zone might be lower.Given the overall small regional differences in dust-phase phthalate concentrations shown in Figure 1, we use the median of all the records across the whole country to represent the dust-phase phthalate concentration in each zone and focus more on how the temporal and spatial variations in environmental factors drive variations in the airborne concentration.As shown in Figure 1, the national median concentrations obtained herein are 510, 97.0, and 23.9 μg/g for DEHP, DnBP, and DiBP, respectively, which are comparable with estimates in some previous review articles [9,21,66].For example, Liu et al. [66] integrated the articles published from 2000 to 2017 and estimated that the weighted median concentrations in indoor dust were 634, 96.3, and 94.1 μg/g for DEHP, DnBP, and DiBP, respectively.
Seasonal variations of phthalate concentrations in indoor settled dust are also evaluated based on the literature data compiled in Table 1.Although the dust samples were collected across seasons in many surveys, there are at least 5 records reporting the concentrations of DEHP and DnBP in the summer and winter, respectively (Table 1).As shown in Figure S1, the concentrations of both species are found to be distributed across a wider range during winter, but the difference in median concentrations between the two seasons is only 10% and not statistically significant (Mann-Whitney U test; p > 0.05).Overall, these results lent further support to the assumption that phthalate concentrations in indoor settled dust stay relatively stable across seasons.This invariance makes sense, as temperature variations across seasons lead to opposite changes in source-gas and gas-dust partition coefficients, which cancel each other out.

Estimation of airborne phthalate concentrations from dust-phase concentrations
The month-to-month variations of airborne phthalate concentrations in residences in differing climate zones are predicted based on Eqs. ( 1)-(4), using the national median dust-phase concentrations presented in the previous section as well as monthly mean indoor temperature and median PM 2.5 concentrations in each zone reported earlier [33,34].The results are presented in Figure 2, with red and blue lines representing the bestestimated gas-phase and particle-phase phthalate concentrations, respectively, and light shades representing the corresponding uncertainty ranges predicted using Monte Carlo simulation.DnBP and DiBP are mainly present in the gas phase, and the particle-phase concentration contributes to less than 3% of the total.In contrast, for the lower-vapor-pressure species DEHP, the particle fraction can vary from 11% to 93%.Overall, the predicted gas-particle partitioning of the three phthalates largely agrees with those observed in real indoor environments (3%-16% in particle phase for DnBP and DiBP, 8%-100% for DEHP) [24,25,28].The gas-phase concentrations of DEHP, DnBP, and DiBP are predicted to range from 0.002 to 0.02 μg/m 3 , 0.5 to 3.4 μg/m 3 , and 0.3 to 1.8 μg/m 3 , respectively.The predicted gas-phase concentrations of DnBP and DiBP are comparable to the median values reported in previous measurements conducted in Chinese residences (0.3-5.2 μg/m 3 for DnBP and 0.4-0.8μg/m 3 for DiBP) [49,54,58,61,62,67], but those of DEHP are lower than limited measured values (0.03-0.37 μg/m 3 ).Since the predicted gas/particle partitioning of DEHP is consistent with measurements, the disagreement between predicted and measured gas-phase concentration of DEHP is less likely due to bias in parameters used in the model.Instead, it is more likely due to the mismatch between dust-phase measurements used for the prediction (Table 1) and limited gas-phase measurements.
The simulated gas-phase phthalate concentrations in residences vary closely with indoor temperature and exhibit strong seasonal variations.The monthly median indoor temperature peaks at ~31°C in the summer (July or August) across all four zones.In contrast, the lowest temperature, which occurs in January or February, ranges from 17°C in the HSCW zone to 23-26°C in the other three zones.Driven by the seasonal variation in indoor temperature, the simulated gas-phase phthalate concentrations are substantially higher in the summer than in the winter.The monthly median gas-phase DEHP concentration in residences in the HSCW zone varies by 10 times throughout a year, ranging from 0.002 to 0.021 μg/m 3 .The corresponding variations in other zones are 2-3 times.For DnBP and DiBP, seasonal variations in gas-phase concentrations are smaller due to their lower enthalpies of evaporation (based on Eq. ( 3)).There is still a six-fold difference in the HSCW zone, and two-fold differences in other zones.The particle-phase phthalate concentrations in residences are proportional to the indoor PM 2.5 concentration and also exhibit strong seasonal variations.In all four zones, indoor PM 2.5 concentrations are higher in the winter than that in the summer, likely due to overall heavier outdoor PM 2.5 pollution in the winter [34].Consequently, the simulated particle-phase phthalate concentrations are higher in the winter.The strongest seasonal variation occurs in the HSCW zone, where the particle-phase concentration of DEHP varies six-fold, ranging from 0.004 to 0.026 μg/m 3 .The seasonal variation is the smallest in the C zone, with a two-fold change.It is worth noting that the seasonal trend of PM 2.5 concentration is opposite to that of indoor temperature, which dampens the seasonal variation of total airborne DEHP concentration.As shown in Figure 2, the simulated total airborne concentrations of DEHP peak during both summer and winter, with maximum seasonal variations of two folds across all regions.

Implication for phthalate exposure in indoor environments
To investigate the influence of temporal and spatial variations in airborne phthalate concentrations on the exposure assessment of phthalates, we calculated the aggregate daily non-diet doses of male adults in residences in each month, which include the contributions from dust ingestion, inhalation, and dermal absorption.In the assessment, all the other parameters are kept the same over time and space, except for airborne phthalate concentrations.As shown in Figure 3, the best-estimated daily non-diet intake of DEHP is about 0.2 μg/(kg day), and stays relatively constant regardless of time and space.The reason is that 97% of the aggregate dose comes from ingestion of indoor dust, in which the phthalate concentration is assumed constant.For the more volatile DnBP and DiBP, dermal adsorption and inhalation of the indoor air account for 60% and 30% of the daily intake, respectively, while dust ingestion only accounts for 10%.As shown in Figure 3, the seasonal variations in daily intake of DnBP and DiBP are similar to the variations in their gasphase concentrations.The daily intakes of DnBP and DiBP peak at ~1.6 and ~0.8 μg/(kg day) in the summer (July or August), respectively.In the winter (January or February), the intake of DnBP falls to 0.3 μg/(kg day) in the HSCW zone and 0.6-0.8μg/(kg day) in the other three zones, and the intake of DiBP falls to 0.1 μg/(kg day) in the HSCW zone and 0.3-0.4μg/(kg day) in the other three zones.Therefore, the largest seasonal variation in the daily intake occurs in the HSCW zone, by up to five and six times for DnBP and DiBP, respectively.In other zones, the maximum difference in daily intake of DnBP and DiBP within 12 months is about three folds.In terms of regional variations, the daily intakes of DnBP and DiBP in the HSCW zone are only 30%-50% of those in the other three zones in the winter, while their peak values in the summer are close.On the time scale of one year (365 days), the summed intakes vary from 317 μg/(kg year) in the HSCW zone to 414 μg/(kg year) in the HSWW zone for DnBP, and 161 μg/(kg year) in the HSCW zone to 211 μg/(kg year) in the HSWW zone for DiBP.It should be noted that the aggregated exposure dose reported herein only accounts for that occurred in the residential environments, and it represents the lower limit of total exposure dose, given that on average adults spend 62% of their time in residences [21].

Conclusions and perspective
In this study, we assessed the airborne concentrations and non-diet exposure of phthalates in residences in China, by simulating gas-and particle-phase concentrations based on reported dust-phase concentrations from the literature, accounting for the seasonal and regional variations in indoor temperature and PM concentrations.Our results demonstrate that these environmental factors can lead to substantial variations in Natl Sci Open, 2023, Vol.2, 20230011 monthly median concentrations of phthalates, with up to ten-fold difference in gas-phase concentrations and six-fold differences in particle-phase concentrations.For the lower-vapor-pressure species DEHP, the major exposure pathway of which is dust ingestion, the aggregate daily dose stays relatively stable across seasons and regions.In contrast, for the higher-vapor-pressure species DnBP and DiBP, the major exposure pathways of which are inhalation and dermal adsorption of indoor air, the aggregate daily doses vary by several folds in different seasons and regions.
The findings of this study have important implications for exposure assessment and epidemiological studies of phthalates and other SVOCs.Firstly, this study suggests that the commonly used practice of assuming a constant temperature and PM concentration in estimating airborne concentrations of SVOCs from their dust-phase levels across different seasons and regions could result in substantial biases.Therefore, future exposure assessments should consider the effects of variations in indoor temperature and PM concentration to improve the accuracy of estimates of SVOC concentrations.Secondly, our results call for cautions on using indoor dust-phase concentration as a proxy for total non-diet exposure for more volatile SVOCs, such as DnBP and DiBP.Exposure to these SVOCs is primarily through inhalation and skin absorption of indoor air and is thus more affected by variations of indoor temperature over time and space.As a result, previous epidemiological studies that examined associations between concentrations of DnBP, DiBP or other more volatile SVOCs in house dust and health outcomes should be interpreted with caution.In contrast, dust-phase concentrations can be a reasonable proxy of environmental exposure for less volatile SVOCs such as DEHP, because dust ingestion is the primary exposure route.This distinction might help explain some of the previously confusing comparison results between internal and external exposure of phthalates.For example, a cross-sectional study conducted in six regions of Japan showed that the concentrations of urine metabolites, measures of total internal exposure in a recent period, were correlated with the concentrations in house dust for DEHP, but the correlations were poor for DnBP and DiBP [68].Looking into the future, we strongly recommend including indoor temperature measurements in addition to dust collection in studies investigating the health effects of more volatile SVOC species.

Figure 1
Figure 1 Summary of median phthalate concentrations in house dust reported in surveys in individual climate zones in China and across the entire country.Panels from the top to bottom show the concentrations of DEHP, DnBP, and DiBP, respectively.The bottom, middle, and top of the boxes represent the 25th, 50th, and 75th percentiles, respectively.Significant differences in dust-phase phthalate concentrations between different zones are labeled on the figure (Mann-Whitney U test; p < 0.05).Herein the whole country is divided into five climate zones in reference to the thermal design code for civil buildings in China (GB50176-2016) [41].SC, C, HSCW, and HSWW represent SC, C, HSCW, and HSWW climate zones, respectively.Data are missing for the M zone and thus not shown in the figure.Data points are colored by surveys in order of appearance in Table1.

Figure 2
Figure 2 Seasonal patterns of indoor environmental factors and predicted airborne phthalate concentrations in Chinese residences in individual climate zones.Panels from the top to bottom show the indoor temperature, PM 2.5 concentration, and predicted gas-(red) and particle-(blue) phase concentrations of DEHP, DnBP, and DiBP, respectively.Panels from left to right show results in SC, C, HSCW, and HSWW zones, respectively.The shaded regions represent the interquartile ranges.For DEHP, the predicted total airborne concentration is shown in grey.The dust-phase concentrations of individual phthalates are treated as constants across seasons and regions, and thus the variations in airborne concentrations are merely driven by environmental factors.

Figure 3
Figure 3 Seasonal patterns of predicted daily intakes of phthalates in individual climate zones.Panels from the top to bottom show daily intake of DEHP, DnBP, and DiBP, respectively.Panels from left to right show results in SC, C, HSCW, and HSWW zones, respectively.The shaded regions represent the interquartile ranges.The dust-phase concentrations of individual compounds are treated as constants across seasons and regions, and thus the variations in exposure doses are caused by the variations in concentrations in the indoor air.

Table 1
Summary of median phthalate concentrations measured in settled dust (μg/g) in residences in China.Each record contains at least measurements in 5 residences.