Occurrence of phthalates in facemasks used in India and its implications for human exposure

ABSTRACT Synthetic polymers with additives are used in the manufacturing of face masks (FMs); hence, FMs could be a potential source of exposure to phthalic acid esters (PAEs). India stands second in the world in terms of the FMs usage since the beginning of Covid-19 pandemic. However, little is known about the PAEs content of FMs used in India. Some PAEs, such as DEHP and DBP are suspected endocrine disrupting chemicals (EDCs); hence, wearing FM may increase the risk of exposure to these EDCs. In this study, we collected 91 samples of FMs from eight Indian cities and analyzed for five PAEs viz. DMP, DEP, DBP, BBP, and DEHP. The PAEs contents in FMs ranged from 101.79 to 27,948.64 ng/g. The carcinogenic risk of N 95 with filter, N-95, and cloth masks was higher than the threshold levels. The findings indicate the need to control PAEs in FMs through regulatory actions.


Introduction
The outbreak of COVID-19 in Wuhan, China, in late 2019 and its subsequent global spread have posed a significant challenge to the health sector.The virus has infected more than 536 million people and caused about 6.3 million deaths all over the world as of 7 June 2022 (Worldometer efficiency, hydrophobicity, and aesthetics.China alone produced 200 million/day of surgical masks to combat the transmission of the virus (Aragaw 2020).
Currently, three types of FMs, viz.surgical or single-use masks, respirators (N95, N99), and cloth/ fabric masks, are used across the globe (Selvaranjan et al. 2021).The surgical masks are made from nonwoven material consisting of three layers viz.an outer hydrophobic layer, a middle melt-blown layer that filters maximum particulate matter, and an inner soft fibrous layer (Prata et al. 2020).The non-woven material is generally manufactured from polymers such as polycarbonate, polypropylene (PP), polystyrene, polyurethane, polyester, or polyethylene (PE) (Aragaw 2020).The respirators generally contain multiple layers and may have a polyester cover, polyurethane nose foam, an aluminium nose clip, and a polypropylene filter (Sureka and Misra 2021).Cloth masks (single or double-layered) emerged as a supplement during the initial phase of the pandemic when the world experienced a shortage of surgical masks.Cloth masks are generally made of cotton, cotton blend, or other mixed fabrics.These can be washed and reused several times.The effectiveness of cloth masks depends on multiple factors, such as the type of fabric material, the number of layers, and the kind of fitting materials used (World Health Organization (WHO) 2020); Rana and Thappa (2020).The WHO recommended respirators for medical professionals and healthcare workers whereas, other types of masks, such as surgical and cloth masks for the general public (Rana and Thappa 2020).
The quality requirements of personal protective equipment (PPE), including FMs used in occupational settings, have been prescribed by international agencies such as the National Institute for Occupational Safety and Health (NIOSH), the US Occupational Safety and Health Administration (OSHA), and the Food and Drug Administration (FDA) (Forouzandeh et al. 2021;Wang et al. 2022).In the US, ASTM F2100 (2011) specifies the performance requirements of FMs.In the European Union (EU), EN166:2001 and EN149:2001+A1:2009 provide harmonized guidelines for the quality and testing requirements (The National Institute for Occupational Safety and Health (NIOSH) 2022).During the beginning of COVID 19, China published new guidelines for FMs, viz.GB2626-2019 (Consumer products, china mandatory standard 2022).As per EU standards, FMs are classified into Filtering Face Pieces (FFP) FFP1 (80% filtration efficiency (FE) for 0.3 μm particles), FFP2 (94% FE), and FFP3 (99% FE).The FFPs consist of four or more layers of filter media and thus have high particle filtration efficiency (PFE) (Forouzandeh et al. 2021).In the US, the NIOSH test protocols classify respirators into N, R, and P series based on FE of 95%, 99%, and 99.7%, respectively.Various types of masks provide different levels of protection against the SARS-CoV-2, and a comparison of different types of masks has been reported (Wang W. et al. 2022).In general, the efficiency of FMs is evaluated by multiple filtration efficiency tests such as particulate filtration efficiency (PFE), bacterial filtration efficiency (BFE), and viral filtration efficiency (VFE) (Forouzandeh et al. 2021).Other quality criteria applicable to FMs are water repellence, flame retardance, differential pressure (breathability), and face fit test.The NIOSH-certified respirators are used to control occupational exposure, including in healthcare settings.Based on PFE, the efficiency of FMs follows the order of respirators > surgical masks > cloth masks (Shakya et al. 2017;Forouzandeh et al. 2021).More details on the quality standards and testing requirements have been reported by Forouzandeh et al. (2021), Shakya et al. (2017), andWang W. et al. (2022).
Synthetic polymers used in PPEs are responsible for the generation of huge plastic waste and biomedical waste during the pandemic (Khoo et al. 2021).Surgical FMs are one of the sources of microplastics; hence, improper disposal of used FMs is a source of secondary microplastics (Shubhanshu and Singh 2021).Also, the continued use of FMs poses several issues to human health, such as nasal dryness, headache, epistaxis, shortness of breath, adverse skin reactions, dryness of eyes, and the environment (Battista et al. 2021;Panda et al. 2021).Various additives such as flame retardants, plasticizers, antioxidants, surfactants, and stabilizers are incorporated into plastics to enhance product quality, functionality, and compliance with quality requirements (Hahladakis et al. 2018;Jin et al. 2021).Most of these additives are semi-volatile organic compounds (SVOCs); hence, FMs may increase human exposure to these SVOCs.PAEs or phthalates are high production volume chemicals used as plasticizers in plastics and additives in textiles as adhesives, stabilizers, and solvents (H-L et al. 2019;Dobaradaran et al. 2020).FMs are designed to fit close to the mouth and nose of the user.Therefore, the probability of inhalation exposure to PAEs is high during FM use.Also, studies indicated that PAEs can easily leach out from cotton compared to other fabrics such as rayon and polyester (Andjelković et al. 2021).PAEs can enter the human body through two different routes; inhalation (Tran and Kannan 2015;Tran et al. 2017;Tang et al. 2020) and dermal routes (H-L et al. 2019).The presence of PAEs in food, water, and the environment is reported in the literature, and inhalation of PAEs from the FMs may further increase the daily intake and result in adverse health effects in humans (Kiani et al. 2018;Mehraie et al. 2022).PAEs have been classified as endocrine-disrupting chemicals and are found to associate with various adverse effects on the male reproductive system (Radke et al. 2020).PAEs such as bis(2-ethylhexyl)phthalate (DEHP), di-n-butyl phthalate (DBP), and benzyl butyl phthalate (BBP) are reported to cause prostate cancer in males (Morrison et al. 2015).The United States Environment Protection Agency (USEPA) classified BBP and DEHP as class C and class B2 human carcinogens, respectively (Selvaraj et al. 2015).The International Agency for Research on Cancer (IARC) has classified DEHP as a group 2B carcinogen (possibly carcinogenic to humans) (Chung et al. 2019).
Phthalates are identified as one of the priority pollutants of ubiquitous occurrence by the US EPA, EU, and China (Katsikantami et al. 2016).Several regulatory bodies, such as the Consumer Product Safety Commission (CPSC) of the USA, EU, and Health Canada, banned the use of phthalates >0.1%in child care articles (Eupropean Council 2005;Canada G of 2016).PAEs in consumer products have been identified as one of the threats by the Strategic Approach to International Chemical Management (Strategic Approach to International Chemicals Management (SAICM) 2020).In India, PAEs are regulated in articles indented for children's use by the Indian toy and children products regulation (BIS 2017), which ban the use of DBP, BBP, DEHP, DNOP (Di-n-octyl phthalate), DiNP (Diisononyl phthalate), and DiDP (Diisodecyl phthalate) in quantities >0.1%, either as a single compound or a mixture of PAEs.However, FMs are not covered under any of these regulations.Recent studies reported the occurrence of PAEs in different types of FMs manufactured in China, Canada, Europe, Japan, Korea, and the United States of America (USA) (Vimalkumar et al. 2022;Wang X. et al. 2022;Xie et al. 2022).The results indicated widespread occurrences of PAEs in FMs, and wearing FMs could be a moderate non-dietary source of PAEs exposure.For a comprehensive assessment of PAEs risk, it is essential to know the PAEs content of different types of FMs used in various countries, especially in highly populated countries such as India.
In this context, the present study investigated the PAEs content in different types of FMs used in India and its exposure levels and probable human health risks.The data will further advance the existing knowledge and help policymakers and regulators to manufacture safe products in times of public health emergency.

Sample collection
We collected 91 FM samples of different brands from eight cities of India, viz.Varanasi (Uttar Pradesh), Hyderabad (Telangana state), Delhi, Kakinada (Andhra Pradesh), Palampur (Himachal Pradesh), Madurai (Tamil Nadu), and Nagpur (Maharashtra), Pune (Maharashtra).The geographical location of the cities is shown in Figure 1.FMs were purchased from shops/malls located in these cities.A few locally manufactured FMs (unbranded) were also collected.No information about the type of polymer used was available in these unbranded FMs.The labels displayed on the outer packaging of the masks suggested the use of non-woven polypropylene (PP), polyethylene (PE), cotton, and polyethylene terephthalate (PET).A few brands were labelled with nylon/polyester spandex blends, which were mainly used in-ear loops.All mask samples were stored in aluminium foil packs to minimize volatilization of PAEs and their crosscontamination from atmospheric air.All masks were categorized as N95, N95 WF, surgical/single-use, and cloth masks.Though ten types of surgical FMs are indicated, we classified all single-use/disposable masks as surgical FMs only (Aragaw 2020).Other details about FM samples, such as the type of mask, weight, and the city of collection, were provided in the supporting information Table ST1.

Standards and reagents
All solvents used, viz.dichloromethane, ethyl acetate, and hexane, were of pesticide residue grade/GC-MS grade and were purchased from Honeywell (Germany).PAEs neat standards viz.dimethyl phthalate (DMP), diethyl phthalate (DEP), DBP, BBP, and DEHP, and internal standard benzyl benzoate were procured from Sigma Aldrich (USA).A stock solution of 1000 mg/l of PAEs mixture was prepared in hexane by dissolving the required amount of neat standard.Working standards were prepared by dilution of the intermediate stock solution of 100 mg/l.All other chemicals used were of AR grade or higher.Glass fiber filter paper of 0.1 µm was used to filter extracted samples.

Sample preparation
The FMs samples were prepared in a clean work area.The sample preparation room was kept free from plastic articles to minimize possible sample contamination.All glassware was cleaned first with detergent, followed by hot water, then heated in a muffle furnace at 350°C for 30 minutes.The cleaned glasswares were rinsed with hexane and ethyl-acetate before use.Samples were processed immediately after collection to minimize probable cross-contamination during storage.Initially, mask parts such as metal nose strips, ear loops, and valves were removed and weighed.The weighed mask (3 or 5 layers) was cut into small pieces (1-2 cm) with the help of a stainless-steel scissor.PAEs in the FMs were extracted following the procedure of (Xie et al. 2022).Briefly, a known weight (about 0.5 g) of the FM was taken in a 40 ml glass vial, and 20 ml of ethyl-acetate: dichloromethane (1:1) was added.Ultrasonic extraction (LAB MAN Ultrasonicator, 40 Hz, India) was used to extract PAEs.The samples were extracted twice, and the combined extracts were evaporated under a gentle stream of N 2 gas (EV-Plus-50, PCI Analytics, India).The residue was dissolved in 500 µl of hexane.Concurrently, a blank sample for each batch (city-wise) was prepared by following the same procedure except using mask pieces.The blank sample was injected into GC-MS before injecting every batch of samples.The sample results were blank corrected and used for data analysis.

Instrumental analysis
PAEs were analyzed by gas chromatography-mass spectrometry (GC-MS, Perkin Elmer, Clarus SQ8C) equipped with an SLB 5 MS capillary column (30 m × 0.25 mm × 0.25 µm).The optimized GC parameters were as follows: injection volume: 2 µl; injector: 310 ℃, oven temperature initial: 100 ℃ for 0.5 min, then increased to 320 ℃ at a rate of 15 ℃/min, hold time at 320 ℃ for 2.5 min.Helium was used as carrier gas at the flow rate of 1 ml/min.The mass spectrometer parameters were as follows: inlet line temperature: 300 ℃, ion source temperature: 280 ℃.The selected ion monitoring (SIM) mode was used for quantitation.The most abundant ion of PAEs was m/z 149, except for DMP, where the most abundant ion was m/z 163.The chromatograph of GC-MS is incorporated in supporting information Figure SF1.In addition, other ions such as m/z 163, 222, etc., were also used as qualifier ions.Information about quantifier, qualifier, retention time (RT), LOD, and LOQ are provided in Table 1.

Quality assurance and quality control (QA/QC)
Stringent quality control measures were followed to minimize background contamination of PAEs.Only glass and stainless-steel containers and tools were used during sample preparation and analysis.To avoid adsorption of PAEs from laboratory air, mask samples were stored in aluminium foil packages and taken out only during extraction.Each sample was prepared in duplicate and analyzed in triplicates.The analytical results were within the relative standard deviation (RSD) of 10-20%.Procedural blanks and matrix spiked samples were analyzed for every 10 samples.Spike recoveries were in the range of 81-121%.A 7-point calibration in the range of 50-500 µg/L with a correlation coefficient of >0.99 was obtained for all analytes.Quantification was done using the internal standard (benzyl benzoate) method.All solvents had trace quantities of all PAEs, i.e. 0.5-1 µg/L, and blank samples had all five PAEs in the range of 1-3 µg/L.Previous studies also reported background contamination in blanks and solvents (Guo et al. 2012).The measured concentrations of samples were corrected for blank and spike recovery.LOD was in the range of 1.10-3.54ng/g (Table 1).Concentrations reported as "not detected (ND)" were substituted by LOD/2 for statistical analysis.

Exposure assessment
The estimated daily intakes (EDI) of PAEs due to wearing FMs were calculated by the following equation (Hyun and Byung 2004;Xie et al. 2022).where C i is the concentration of PAEs in the mask (ng/g); N is the number of masks used per day (one mask/day was considered); M is the weight of mask (g) (Supporting Information Table ST 1); and "A" denotes the absorption rate of PAEs from the mask by humans (20%).For the calculation of dermal absorption rate of PAEs, previous studies on personal care products and cosmetics (5%) (Koniecki et al. 2011;Gkrillas et al. 2021), sanitary napkins (0.5-10%) (Gao et al. 2020;Gao and Kannan et al. 2020) and baby diapers (0.5-10%) (Ishii et al. 2015) were considered.While wearing a mask, PAEs exposure occurs via two routes viz.(i) dermal absorption -as the mask is often tightened around the face, direct contact of the mask and skin is imminent, and (ii) inhalationgenerally mask is fitted over the mouth and nose; thus, the probability of direct inhalation of SVOCs such as PAEs from the mask is expected high.Yubin et al. reported that the release of PAEs from clothes would be higher due to sweating; hence, intake of PAEs via mask is expected to be higher during summer (Liu et al. 2020).By considering these scenarios and taking into account of the recent study (Xie et al. 2022), we considered an absorption rate of 20% for adults with a mean body weight of 60 kg (Nutrient Requirements And Recommended Dietary Allowances For Indians National Institute of Nutrition, Indian Council of Medical Research (ICMR) 2010).The product of M, C, and A was related to the intake of the PAEs from one FM per day.

Risk assessment
The non-carcinogenic risk associated with PAEs exposure via masks was estimated using hazard quotient (HQ), and the cumulative risk was estimated by hazard index (HI).HQ is defined as the ratio of EDI to the reference dose (RfD) (Zhang et al. 2020;Xie et al. 2022).
Exposure to individual PAEs was calculated using EDI.The RfD is equivalent to the tolerable daily intake (TDI) values of PAEs established by the European Food Safety Authority (EFSA) and the US Environmental Protection Agency (USEPA).The RfD of DBP, DMP, DEP, BBP, and DEHP were 100, 100, 800, 200, and 20 µg/kg-bw/day, respectively (David 2000;Guo et al. 2012;Alnaimat et al. 2020).The carcinogenic risk (CR) of the PAEs wass computed by the following equation (Wei et al. 2020).

Data analysis
Data analysis was done by using SPSS version 26 (IBM, SPSS statistics).In order to determine whether the data were normally distributed, the Shapiro-Wilk test was used.In cases where the data did not meet the normal distribution, a nonparametric test was used to compare various groups.Multiple groups were compared using a Kruskal-Wallis H, and two groups were compared using the Mann-Whitney U test.

Occurrence of PAEs in FMs
The PAEs contents in the 91 FM samples collected from eight cities were given in supporting information Table ST2.The summary of the results is given in Table 2, and more details on PAEs distribution were given as the box-whisker plot (Figure 2).A Shapiro Wilk test revealed that concentrations of most PAEs in facemask were not normally distributed (p < 0.05).The ∑ 5 PAEs content in FMs ranged between ND and 27,948.64ng/g (median 2578.94ng/g).Significant difference between the total concentrations of PAEs among the different facemasks was observed (p < 0.05).Total PAEs content of the four types of masks followed the order N95 WF > N95> cloth > surgical masks.Total PAEs content showed 1-2 orders of magnitude variation in the four types of masks (Figure 2).The ∑ 5 PAEs found in FMs of this study were broadly in agreement with similar two studies published recently.Xie et al. reported ∑ 11 PAEs in the range of ND and 37,700 ng/g (median 1950 ng/g) (Xie et al. 2022), and Vimalkumar et al. reported ∑ 9 PAEs in the range of ND and 7,670 ng/g (median 1300 ng/g) (Vimalkumar et al. 2022).The former study reported FM samples from China, USA, and Canada, whereas the latter collected FMs from China, EU, Japan, Korea, and the USA.These two studies analyzed nine and eleven PAEs, respectively, as against five PAEs of this study.However, except for di-iso-butyl phthalate (DiBP), we analyzed all the dominant PAEs reported in these two studies.Therefore, we expect that the actual PAEs content would be higher than the results reported here.Also, the mean PAEs content of 3850.27ng/g reported in this study is higher than that of Xie et al. ( 2022) and Vimalkumar et al. (2022).
Also, the median concentration of total PAEs (2578.94ng/g) was higher than that of FMs from China (1950 ng/g) (Xie et al. 2022) and single-use FMs from the USA and a few other countries (1300 ng/g) (Vimalkumar et al. 2022).The median PAEs concentration was also higher than the feminine hygiene products such as sanitary pads (362 ng/g), panty liners (1830 ng/g), tampons (1130 ng/g), wipes (546 ng/g), bactericidal cream (47.9 ng/g), deodorant spray (26.7 ng/g) (Gao and Kannan et al. 2020), and sanitary napkins collected from six cities of USA (1534 ng/g), UK (1700 ng/ g), Australia (2441 ng/g) and Germany (2197 ng/g) (Tang et al. 2019).This shows FMs used across the globe are one of the important sources of PAEs exposure to humans.In the present work, the median concentration of PAEs in the surgical mask was less than that of N95, N95 WF, and cloth masks.This could be due to the multiple layers (up to five) present in N-95, and N-95 WF masks, compared to the three layers present in the surgical masks.Our findings were similar to that of Wang et al., who also found about twofold higher PAEs levels in N 95 masks compared to surgical masks (Wang X. et al. 2022).In the present study, the median content of DEHP (594.48 ng/g) and DBP (668.64 ng/g) were found to be one order of magnitude higher than that of the other three PAEs.Hence, the use of hot-melt adhesive to bind the different polymeric layers is a major source of PAEs (Aragaw 2020).The cloth mask, which is usually made up of cotton and polyester fabrics (Khan and Parab 2020), had a higher PAEs concentration (median 2902.16ng/g) than that of the surgical mask (median 1415.99ng/g).However, the median concentration of PAEs in cloth masks (2902.16ng/g) was lower compared to the median concentration of six PAEs (5748 ng/g) in children's clothing from India (Tang et al. 2020).The higher concentrations of PAEs found in FMs are due to the use of synthetic polymers such as PP, PE, and hot melt adhesive, which are commonly used in the production of a surgical, N 95, and N 95-WF FMs.In hot-melt adhesives, PAEs are added during their production to enhance the ductility and gel properties (Gao et al. 2020).More quantity of hot melt adhesive is required to bind five layers of PP, which further increases the concentration of PAEs in N 95 masks.Previous studies indicated that most of synthetic polymers (PP, PE, PET) and hot melt adhesives use DEHP as the plasticizer (Gao et al. 2020;Cao et al. 2022;Xie et al. 2022).The higher concentration of PAEs in clothing has been attributed to the use of other materials in textile industries, such as synthetic leather, buttons, coated fabric, plastisol, and dye printing which act as a primary sources of PAEs (Walters et al. 2005;Nijssen-Wester 2021).Dyes composed of PAEs are commonly used to impart color to textiles, thus increasing the overall PAEs concentrations.Infant cotton clothing from China also reported a median value of 4150 ng/g, which is 2.5 times higher than that of FMs of this study.Further, the PAEs content of clothes does not decrease appreciably even after repeated washing (H-L et al. 2019).This indicates that washing of masks does not reduce the PAEs content of cloth masks and hence the exposure levels.Another source of PAEs in clothing is sorption from ambient air during manufacture and storage; however, PAEs contribution from ambient air to FMs is reported to be minimal (H-L et al. 2019).

PAEs profiles in FMs
The PAEs profiles in the different types of FMs are given in Table 2 and Figure 2. The mean detection frequencies (%DF) of DMP, DEP, DBP, BBP, and DEHP were 73.63%, 81.32%, 82.42%, 70.33%, and 71.43%, respectively.Among the five PAEs analyzed, DEHP accounted for 39.15% (mean value) of total PAEs concentration, followed by DBP (28.65%),DMP (19.24%),DEP (8.22%), and BBP (4.74%), indicating that the three phthalates viz.DEHP, DBP, and DMP were the dominant phthalates present in FMs used in India.The maximum concentrations of DMP, DEP, DBP, BBP, andDEHP observed were: 4562.5, 2248, 9303, 1418, and20,173.93ng/g, respectively (Supporting Information Table ST2).The maximum concentrations of DMP and DBP were found in N 95 WF, and cloth masks, respectively, whereas maximum concentrations of DEP, BBP, and DEHP were found in N 95 masks.There was no significant difference between the total concentrations of DMP, DEP, DBP, BBP, and DEHP among the facemasks (p < 0.05).The PAEs distribution followed the order DEHP > DBP > DMP > BBP > DEP.In general, PAEs profiles observed in this study were similar to that of Vimalkumar et al. (2022), Wang X. et al. (2022), and Xie et al. (2022); however, one striking difference in the PAEs profiles observed in our study was the higher DF (73.63%) and concentration of DMP (mean: 741.14 ± 1048 ng/g) compared to the other two studies (DF: 76.8 and 26, respectively and mean: 22.8 and 34 ng/g, respectively).Wang et al. also reported higher DF of DMP (88%), but the authors reported the PAEs content as ng/mask basis; hence, we could not make a comparison with our study (Wang X. et al. 2022).Vimalkumar et al. (2022) noticed that surgical FMs from various countries showed the different compositions of all PAEs, including DMP.In our study, DMP content in surgical FMs showed wide variation similar to the other two studies (range: ND-571 ng/g; median 241.25 ng/g) (Vimalkumar et al. 2022).One probable reason is the difference in the polymeric material used in surgical FMs of different brands.Vimalkumar et al. observed that surgical FMs from different countries use different polymers; therefore, it is possible that the observed variation in PAEs content of the same type of mask could be due to the use of different polymers by different manufacturers (Vimalkumar et al. 2022).The DMP content of cloth masks was about threefold less than that of surgical FMs (range: ND-369.46ng/g; median ND).On the other hand, DMP content of N 95 WF, N 95 was higher by orders of magnitude than cloth and surgical masks.This suggests that the nature of synthetic polymer present in FMs was the source of variation of PAEs compositional profiles.Cloth masks are predominantly made of cotton and other synthetic fibres, which may contribute less PAEs than that of polymers used in surgical FMs.Further, Vimalkumar et al. observed that surgical FMs from different countries use different polymers, resulting in wide variations in PAEs profiles (Vimalkumar et al. 2022).Wang et al. also reported wide variations in PAEs compositional profiles among the same types of masks (surgical and N 95) from different countries (Wang X. et al. 2022).
Among the five PAEs analyzed, the concentration DBP and DEHP together accounted for 68% of the total PAEs content indicating that these two are widely present in polymers used in FMs (Figure 2).PAEs profiles in N 95 and N 95 WF differed significantly (p < 0.05).Similarly, significant differences between the two N 95 types and cloth and surgical FMs were observed (p < 0.05).For example, in cloth masks, DBP was predominant with a median concentration of 1097.25 ng/g, whereas median concentration of DEHP was 188.42 ng/g.In N 95 and N 95 WF masks, next to DEHP, DMP was dominant, accounting for 28% and 26% of the total PAEs.The reason for the higher levels of DMP in N 95 type FMs could not be established, but it could be due to the additional layers and/or materials used in the manufacturing process.In surgical masks, the median concentrations of DEHP (558.23 ng/g) and DBP (466.09ng/g) were more or less similar, with the contribution of 38% and 32% to total PAEs.
Hot melt adhesives, a common material used in many disposable hygiene products, are widely used in face mask to bind the different layers of facemask.It has been reported that DEHP was the major plasticizer used in the hot melt adhesives to enhance the ductility and gel properties of FMs (Gao et al. 2020;Xie et al. 2022).Thus, relatively higher concentrations of DEHP would be expected in N 95 type FMs compared to surgical and cloth masks.In majority of FMs, the PAEs content were directly proportional to the weight of facemask as well as the number of layers in facemask.Overall, we observed significant differences in PAEs concentrations between the different types of FMs.This shows the difference in the PAEs levels in the raw materials and the manufacturing process.Also, the adsorption or contamination of PAEs from ambient air during different manufacturing stages may serve as a secondary source of PAEs in masks.However, the concentrations of this secondary source are reported to be negligible (Morrison et al. 2015).On the other hand, the studies by Xie et al. and Vimalkumar et al. did not find significant differences in PAEs profiles among different types of masks (Xie et al. 2022;Vimalkumar et al. 2022).

PAEs in FMs collected from different cities of India
The PAEs contents of FMs collected from eight different cities of India were given in supporting information Table ST1.A comparison of mean PAEs concentrations of the four different types of masks collected from different cities is provided in Figure 3.The median concentrations of PAEs in surgical mask in different cities were found in the following order; Varanasi (2439.00ng/g) > Pune (2040.16ng/g) > Delhi (1822.97ng/g) > Madurai (1650.56ng/g) > Palampur (1621.46ng/g) > Hyderabad (1396.81ng/g) > Nagpur (1284.87ng/g) > Kakinada (815.69 ng/g).Similarly, the median concentrations of PAEs in N-95 mask was found to be in the order Delhi (8324.94ng/g) > Pune (6426.45ng/g) > Varanasi (6364.99 ng/g), Kakinada (5914.49ng/g) > Madurai (5251.42ng/ g) > Nagpur (2793.20 ng/g) > Hyderabad (2313.89ng/g) > Palampur (988.91 ng/g).The N-95 WF masks were collected from only two cities, viz.Kakinada and Varanasi and the median PAEs concentrations were 8425.15 and 7472.77ng/g, respectively.In cloth masks, the median concentration of PAEs was found in the order: Varanasi (6093.16ng/g) > Kakinada (3098.94ng/g) > Hyderabad (2736.70 ng/g) > Nagpur (2678.29 ng/g) > Delhi (2404.94ng/g) > Pune (1144.49ng/ g) > Palampur (101.80 ng/g).The highest mean PAEs concentrations were found in the surgical mask and N-95 WF mask samples collected from Varanasi.Similarly, the lowest total and individual PAEs concentrations were found in the cloth and N-95 masks collected from Palampur (Table 3 and Figure 3).Overall, statistically significant differences (p < 0.05) were observed in the PAEs contents of FMs collected from different cities.
It is to be noted that except for a few locally manufactured FM samples, all other FMs were manufactured in places other than the city of collection.Hence, the difference in the PAEs content found between samples of different cities is related to PAEs content of raw materials and manufacturing process rather than due to geographical locations.

Estimation of PAEs exposure
The EDIs of PAEs exposure from FMs were computed, and the results were given in supporting information Table ST3.The Σ EDI values ranged between 1.70 and 331.57ng/kg-bw/day, with a median value of 28.65 ng/kg-bw/day.Among the five PAEs, the exposure due to DEHP was higher, accounting 41.53%, followed by DBP 28.22%, DMP 18.06%, BBP 3.85%, and DEP 8.34%.The highest EDI value of 331.57ng/kg-bw/day was observed for the N95 mask collected from Pune city.Similar to the PAEs content, the mean EDI value followed the order N 95 WF > N 95 > Cloth > Surgical masks (Figure 4).
Xie et al. reported the EDI value in the range of 3.71-639 ng/kg-bw/day for eleven PAEs in FMs of China (Xie et al. 2022).Vimalkumar et al. reported the EDI in the range of 3.5 to 151 ng/kg-bw /day for nine PAEs in single-use FMs collected from the USA, Canada, and other countries (Vimalkumar et al. 2022).Though we analyzed only five PAEs, the EDI found in our study is comparable to these two studies.The exposure to PAEs from FMs was compared with the other exposure pathways to see whether wearing FMs is a significant source of PAEs exposure.For example, the EDI of fifteen PAEs through food (288.12µg/kg-bw/day), water (0.095 µg/kg-bw/day), indoor air (43.19 µg/kg/day), out-door air (1.245 µg/kg-bw/day) of Delhi, India has been reported (Das et al. 2014).These EDI values are 2-4 orders higher than the EDI values of this study (mean 0.062 µg/kg-bw/day) except through drinking water.Also, the median EDI value (28.65 ng/kg-bw /day) of this study was lower than the dermal route, infant clothing (776 ng/kg-bw/day) (H-L et al. 2019), indoor dust (2022 ng/kg-bw/day), out-door dust (53460 ng/kg-bw/day) (Das et al. 2014) and composite sanitary napkin (86.00 ng/kg-bw/day) (Gao et al. 2020).These findings indicate that wearing FM is a minor source of PAEs exposure compared to major pathways such as food and dust ingestion.

Risk assessment
The carcinogenic (CR) and non-carcinogenic (Non-CR) risk of PAEs exposure through FMs were estimated and was provided in supporting information Table ST3.A potential Non-CR risk to humans can be expected if the value exceeds 1.The mean and median values of HI risk due to exposure PAEs via FM were 1.60 × 10 −3 and 4.45 × 10 −4 (non-CR), respectively.The observed values were 3-4 orders of magnitude lower compared to the safe value.The highest HI value of 1.28 × 10 −2 observed was well below the acceptable range; therefore, using the facemask is not likely to pose non-carcinogenic risks.The carcinogenic risk is considered significant when the CR value is greater than 1 × 10 −6 .In this study, the mean CR value from exposure to PAEs was 6.29 × 10 −6 .The mean CR values for the N-95 mask, N-95 WF, and cloth mask were 1.08 × 10 −5 , 1.77 × 10 −5 , and 6.13 × 10 −6 , respectively, which were higher than the acceptable value (1 × 10 −6 ).On the other hand, the mean CR value of 9.72 × 10 −7 for the surgical mask was lower than the threshold value.The maximum CR value of 3.19 × 10 −5 was observed for the N-95 WF mask from Varanasi city.Therefore, wearing FMs with PAEs levels observed in this study may pose a potential carcinogenic risk.Adequate risk reduction strategies such as minimization of PAEs levels, use of less harmful and non-phthalate alternatives, may be considered.

Need of regulatory guidelines for additives in FMs
The presence of hazardous additives such as high molecular weight PAEs, flame retardants, and volatile organic compounds (VOCs) in FMs has been the subject of interest after the outbreak of the  ST2).However, given the continued use of FMs by health professionals and frontline workers, PAEs exposure via FMs could be a potential source; hence regulatory agencies may consider setting appropriate standards for PAEs and other additives.

Limitations of study
This study gathered preliminary information about the PAEs contents of FMs used in various cities of India.Although the data provide insights on PAEs concentrations in FMs, EDI, and probable risks, the following limitations should be noted (i) only five major PAEs were studied, and other PAEs might be present in appreciable concentrations (ii) the findings may not represent the whole FMs market of India due to the limited number of samples (iii) FMs collected from a particular city may not have been manufactured in that city; hence the reported concentrations are not representative of the cities where samples have been collected (iv) variations in the size, weight, and the number of masks worn per day influence exposure levels (v) assumptions on inhalation and skin absorption rate and its dependency on individual PAEs, the effect of temperature and moisture, confounding factors, etc. were not considered for the calculation of EDI.

Conclusions
During the ongoing COVID-19 pandemic, wearing FM is the best possible intervention to protect humans from contracting the virus.This is the first study reporting PAEs content in FMs used in India.
Five PAEs were determined in 91 FM samples collected from eight cities.The total concentrations of the PAEs in the masks ranged from 101.79 to 27,948.33 ng/g, with a mean of 3850.40 ng/g.The EDI values for Σ PAEs ranged between 1.17 and 331.57ng/kg-bw/day, with a mean value of 98.36 ng/kg-bw/day.The non-carcinogenic risk based on HI was far less than the threshold level; however, the carcinogenic risks of cloth, N 95, and N 95 WF masks were higher than the acceptable limit.Our results show that though the exposure due to PAEs in FMs is considerably less than in other pathways, prolonged wearing of FMs is a cause of concern given the different additives present in FMs.Currently, the market is flooded with different types of FMs, and no regulations exist on the safety of additives present in FMs.Hence, regulatory agencies need to take appropriate actions to reduce the PAEs and other additives in FMs.
Further research based on urinary biomonitoring could provide more insights on PAEs exposure via FMs.

Figure 1 .
Figure 1.Geographical status of Indian cities selected for face masks study.

Figure 4 .
Figure 4.Estimated daily intakes of PAEs from FMs.

Table 1 .
GC-MS parameters used for the quantitation of PAEs.

Table 2 .
Descriptive statistics of PAEs concentration in different types of face masks.

Table 3 .
Median concentration of PAEs in different FMs.
(Canada G of 2016)rnández-Arribas et al. 2021;Jin et al. 2021;Vimalkumar et al. 2022; Wang X.  et al. 2022).Hazardous additives present in consumer products and human exposure have been identified as the top priority area of SAICM (Strategic Approach to International Chemicals Management (SAICM) 2020).Though various regulations such as the European Union (EU) REACH (EU Council 2005), US CPSC (U.S. Consumer Product Safety Commission (USCPSC) 2014), Health Canada(Canada G of 2016), and Bureau of Indian Standards (BIS 2017) restrict the use of PAEs to <0.1% for child care products and toys, no specific regulations exist for chemical additives present in PPE such as masks.The total PAEs content of all mask samples in this study was <0.1% (Supporting Information Table