Final Report: Tracking Semivolatile Organic Compounds Indoors: Merging Models and Field Sampling to Access Concentrations, Emissions, and Exposures

EPA Grant Number: R835641
Title: Tracking Semivolatile Organic Compounds Indoors: Merging Models and Field Sampling to Access Concentrations, Emissions, and Exposures
Investigators: Bennett, Deborah H. , Young, Thomas M , Shin, Hyeong-Moo
Institution: University of California - Davis
EPA Project Officer: Klieforth, Barbara I
Project Period: September 1, 2014 through August 31, 2017 (Extended to August 31, 2018)
Project Amount: $900,000
RFA: New Methods in 21st Century Exposure Science (2013) RFA Text |  Recipients Lists
Research Category: Safer Chemicals , Human Health

Objective:

Objective of Research: The environmental health community has growing concerns about many of the semivolatile organic compounds (SVOCs) introduced into residential environments resulting in exposures to these compounds and their transformation products. Methods for conducting rapid toxicological assessments are currently being utilized to help evaluate the hazards. Similar methods are needed in exposure science in order to determine safe levels for use by consumers. To address this need, we propose here assessments based on effective merging of measurements and models. The goals of this project are to (1) measure concentrations of a broad spectrum of target and non‐target SVOCs in indoor dust to estimate emission rates and exposures, (2) refine and evaluate a multi‐ compartment indoor fate, transport, and exposure model, and (3) evaluate air‐to‐skin transdermal uptake models

The report is organized by the objective, with the full text listed in the grant provided.

Summary/Accomplishments (Outputs/Outcomes):

Objective 1: Measure concentrations of a broad spectrum of SVOCs in indoor dust to estimate emission rates and exposures

  1. Develop a novel analytical workflow for rapid assessment of a broad spectrum of SVOCs in dust, skin, and polyurethane foam using high resolution mass spectrometry. We will use the new generation of liquid chromatography time-of-flight mass spectrometry (LC-TOF-MS) and gas chromatography (GC)- TOF-MS instrumentation to identify and quantify a far more comprehensive set of SVOCs in dust.
  2. Collect indoor dust samples and measure concentrations of a broad spectrum of SVOCs . By collecting dust from homes and using newly developed applications that combine analyses for pre-selected target compounds as well as other non-target compounds in the sample, we will identify exposure concentrations for a comprehensive set of SVOCs.
  3. Use measured dust concentrations and inverse modeling to estimate emission rates that can be used to estimate human residential exposures . Combining field data with our indoor mass-balance model allows us to determine corresponding emission rates, a critical model input with little to no existing data. With these emission rates, we will use our indoor exposure model to estimate chemical exposures that can then be compared to toxicological levels of concern and identify candidate compounds for source reduction to improve public health.
  4. Conduct factor analysis on dust concentrations to look for common source profiles. This technique will help us understand potential sources of concern and will be substantially aided by the larger set of potential marker compounds that may be diagnostic for particular sources.

Introduction

Indoor environments have been recognized as an important research area to understand the impact of chemical exposures on human health 1, because people spend most of their time indoors 2 and some studies reported consistently higher concentrations of chemicals of concern indoors versus outdoors 3,4. Consequently, the need to develop methods to better characterize indoor chemical exposures has become a high priority in the field of exposure science 5-9. In indoor environments, consumer products are major sources of human exposure to semivolatile organic compounds (SVOCs), including ingredients in insecticides, plasticizers, flame retardants, water- and oil-repellents, and personal care products (PCPs) 10. SVOCs released from their original sources are fairly persistent indoors, and are redistributed over time and partitioned to indoor air, settled dust, and other indoor surfaces 11,12. As a result, individuals are exposed to SVOCs via inhalation, dermal uptake, and non-dietary ingestion of settled dust 5.

For SVOCs with low vapor pressures, their levels in indoor dust are known as an alternate marker of chemical exposure with common indoor sources (furniture, plastics, PCPs, pesticides, etc.) 13. Studies found that concentrations in dust for some of these SVOCs were correlated with those in indoor air and other indoor surfaces, as well as those measured in biological samples 14-19. Therefore, dust concentrations have been used as a surrogate for human exposure in epidemiologic studies 20-23. Dust levels were also used to reconstruct total (non-dietary) residential exposure by applying equilibrium partitioning models to estimate corresponding concentrations in the gas-phase (both inhalation and air- to-skin dermal uptake) and in airborne particles (inhalation) 13. However, a complete picture of the chemical fingerprint of dust (i.e., identity and quantity of all chemicals present) is missing, because most previous studies analyzed known chemical classes via a targeted analytical approach 13,24-28.

With recent developments in high-resolution mass spectrometry, it is possible not only to look for known compounds (targets) for which authentic standards are available, but also to screen for expected compounds from a database or library (suspects) and even to identify previously unknown compounds (non-targets) through careful examination of the high resolution mass spectra.29

While screening methods by liquid chromatography high-resolution mass spectrometry (LC-HRMS) have been applied often for aqueous media30-35, non-target screening studies in other environmental media such as household dust are still rare. So far, the most thorough investigation of non-targets in dust has been done by Rager et al. (2016)36 who investigated more than 50 dust samples by LC time of-flight (TOF) MS. They linked the proposed formulas to EPA's Distributed Structure-Searchable Toxicity (DSSTox) database. However, their identification was only based on molecular formula match, as they did not acquire MS/MS data. A recent study by Ouyang et al. (2017)37 applied two-dimensional (LCxLC)- ToF MS in order to get a better separation of the non-target features. However, as they only looked into one dust sample, the generalizability of these results is limited. Other non-target studies specifically looked at flame retardants or brominated azo dyes in household dust38-40. Hilton et al. (2010)41 used two dimensional GCxGC-MS coupled to electron ionization (EI) to investigate non-target chemicals in dust including phthalates, polycyclic aromatic hydrocarbons, chlorinated compounds, brominated compounds, and nitro compounds.

Taking a closer look at the chemicals previously detected in dust and the analytical methods with which the chemicals were analyzed, it becomes clear that dust contains chemicals with diverse physico- chemical properties and structures. Chemicals range from very polar and non-volatile surfactants to non-polar and semi-volatile brominated flame retardants. This is also reflected in the number of analytical studies that have investigated chemicals in dust; roughly the same number of methods are based on LC-MS compared to GC-MS. To date there is no study that comprehensively investigated chemicals by both platforms (LC-MS and GC-MS) at the same time. Hence, a complete picture of the chemical fingerprint in household dust is missing, as is a comparison of the strengths and weaknesses of the two analytical approaches.

This research gap is addressed in the present study, which uses a target, suspect and non-target screening workflow for polar to semi-polar chemicals analyzed by LC-HRMS as well as a target and non- target screening method for non-polar chemicals analyzed by GC-HRMS. A total of 38 household dust samples were collected in California, and the findings of the detected chemicals are discussed. The differences between the non-target screening approaches on both platforms (LC-HRMS and GC-HRMS) are critically discussed and the complementary roles of the two platforms are acknowledged. We present SVOC concentrations analyzed in our dust samples and calculate source strengths.

Materials and Methods:

Dust sampling and Extraction

Dust samples from 38 households in the areas of Sacramento and Fresno, CA, were collected from the main living area with the High Volume Small Surface Sampler (HVS3) using a standard protocol42 and stored in a PTFE container at -20°C until extraction. Dust samples were sieved (106 µm) and 100 mg were sonication extracted using hexane:acetone (3:1 v/v) and acetone (100%). The extract was evaporated, filtered and split into a GC-fraction and a LC-fraction which were run on the corresponding instruments.

Targeted Chemical List Selection

A total of 76 chemicals to be analyzed by GC-Q/TOF and 56 chemicals to be analyzed by LC-Q/TOF were selected for this study. The selection comprised one or multiple indicator compounds from substance classes identified previously28,43-48 or compounds present in consumer products listed in the Consumer Product Chemical Profiles CPCP database49. The target list consisted of pentabromodiphenyl ether (BDE), organophosphate flame retardants (OP-FR), phenols, polycyclic aromatic hydrocarbons (PAH), phthalates, UV filters, components of fragrances, pesticides, plasticizers, parabens, biocides, polyfluorinated compounds, surfactants and skin oils.

GC-Q/TOF and LC-Q/TOF analysis

The analysis on the GC-Q/TOF was carried out on an Agilent 7890B gas chromatograph using a HP-5MS (30 m x 0.25 mm, 025 µm) column coupled to an Agilent Q/TOF 7200B running in electron ionization (EI) mode. A 78 min runtime with a linear temperature gradient from 35°C to 325°C was chosen to separate all 76 target chemicals and all major peaks in the analysis of a dust extract.

The analytical method for the LC-Q/TOF was taken from Moschet et al (2017)50 for water extracts. In brief, a C18 column (2.5x100 mm, 1.8 µm, Zorbax Eclipse Plus, Agilent Technologies, Inc.) was used for separation with the following mobile phases: positive ionization mode: A) ultrapure water plus 0.1% formic acid, B) acetonitrile plus 0.1% formic acid; negative ionization mode: A) ultrapure water plus 1 mM ammonium fluoride, B) acetonitrile. Ammonium fluoride in ultrapure water was chosen in negative mode because it had >10x higher sensitivity for phenolic compounds such as bisphenol A compared to other buffers tested. The injection volume was 10 µL. An Agilent 6530 Q/TOF (Agilent Technologies, Inc.) was used in positive and negative ionization mode. Acquisition was done in All-Ions fragmentation mode using collision energies (CE) 0, 10, 20, and 40 eV (scan rate: 4 spectra/sec) for the target and suspect screening (see below). The 0 eV channel was used to collect precursor ion information while the higher CE channels were used to obtain fragment ion information simultaneously.

Method Validation

The optimized extraction and analytical method was validated by extracting a triplicate of the NIST SRM 2585 dust (standard reference material). A spike recovery experiment was done by adding a mixture (500 ng) of all 132 target compounds to the NIST SRM 2585 dust, letting the solvent dry overnight, and extracting following the procedure described above. Absolute recovery was calculated by dividing the area of the pre-spiked sample by the area of a post-spiked sample, i.e., a NIST dust extract spiked immediately before instrumental analysis. This experiment was also conducted in triplicate. Finally, a triplicate of a method blank was extracted using an inert silica material (MIN-U-SIL®, U.S. Silica Holdings Inc., Frederick, USA) as a dust surrogate. The same method validation approach was used on both analytical platforms.

Targeted Quantification Method

Quantification of the target chemicals on both LC-Q/TOF and GC-Q/TOF employed Agilent MassHunter Quantitative Analysis (B.07). In LC-Q/TOF, the [M+H]+ or [M-H]- were used as quantifiers and - depending on the compound - the one or two most abundant fragments from the library spectra were used as qualifiers in the All-Ion scans (exact mass window ± 20ppm). In GC-Q/TOF, the most abundant fragment was used as quantifier and two further fragments used as qualifiers (exact mass window: ± 25ppm).

Non-Target Screening by GC-Q/TOF

The 38 samples including method blanks were re-run in one randomized sequence using the same acquisition method as described above. Before and after the sequence, an alkane mix was run to calculate the retention time index (RI) of all non-target features.

In a first step, non-target features were extracted by spectral deconvolution using Agilent Unknowns Analysis software. The software calculates a score based on the quality of the deconvolution (component shape quality). The software runs in batch mode, but corresponding features between samples are not grouped together (no binning and alignment). In a second step, the software compares spectra for each feature using a spectral library and calculates a match score. In this study, the NIST 14 library51 was used (unit mass resolution). All compounds with a component shape quality >60, a NIST library match factor >60 and a chromatographic peak width between 3 and 15 seconds were selected (parameters calculated by the MassHunter Unknown Analysis software).

In a next step, features that were the same between the samples were grouped together manually because the software did not support this step in an automated way. This was done by comparing the candidate names as well as "mass - retention time combinations'. In general, mass deviations of ±50 ppm and RT deviations of ±0.2 min were found to be acceptable limits based on the results of the target compounds that were identified in the non-target workflow. Unfortunately, the library match name of the same compound in two samples can be different if two compounds in the library have similar fragment spectra. Also, the reference mass (ion with highest intensity) can be different if two fragments have similar intensities.

Long-chain alkanes, their acids, esters and similar compounds were neglected from the further selection as they were considered as not relevant for this project. For the selected compounds, the calculated RI was compared with the NIST library value (experimental or estimated). A deviation of ±2% in RI was considered acceptable based on the experience with the target compounds that were also detected by the non-target approach. If the NIST library only contained an estimated RI, a deviation of ±10% in RI was acceptable. The second criterion was set arbitrarily because the confidence interval of the estimation by NIST varies significantly depending on the compound properties. All compounds with intensities lower than ten times the intensity in the method blank were discarded.

Suspect Screening by LC-Q/TOF

Suspect screening was conducted using the Agilent MassHunter Qualitative Analysis software (version B.07) by applying the "Find by Formula" workflow following the method described in Moschet et al. (2017)50. Two curated spectral libraries, Agilent Forensic Toxicology Personal Compound Database and Library (PCDL) and Agilent Water Contaminant LC/MS PCDL, containing 8,000 and 1,450 compounds with MS/MS spectra were used. Briefly, compounds for which a chromatographic peak was found for their main adduct (mass accuracy: ±10 ppm) and for which the isotope pattern gave a good match (score

> 70/100)50 were selected. Next, the exact masses from the five main fragments in the PCDL's MS/MS library spectra (CE 10, 20, 40) were searched in the high energy scans by the software. If one or more fragments were present and co-eluting with the parent (determined based on a coelution score in the software), the compound was automatically flagged as "qualified". Compounds "qualified" in at least five out of the 38 samples and for which the intensity was higher than ten times the intensity in the method blank were manually inspected. If possible, a reference standard was purchased for these tentatively identified compounds (confidence level 252) for unambiguous confirmation.

Non-Target Screening by LC-Q/TOF

  1. Recursive Feature Extraction

All samples including the method blanks and the NIST reference dust were re-run in triplicate in positive and negative modes in randomized order. For five samples, three individual extraction replicates were run in addition. This resulted in 149 injections in both positive and negative modes. The acquisition followed procedures described above, but only the full scan (CE=0 eV) was acquired with a scan rate of

1.5 spectra/sec. Agilent Profinder software (version B.08.00) was used to extract non-target compounds by the "Batch Recursive Feature Extraction' workflow. In brief, the software searches and identifies molecular features in the first sample. A feature is a group of corresponding ions, i.e., adducts and isotopes of the same compound, that form a chromatographic peak at a certain RT. All detected features are stored in a list with their exact monoisotopic mass and RT. Next, the software searches all features in the subsequent samples. The detected features in all samples are compared, the exact masses and RT are aligned and the corresponding features are binned together. This results in a list of features and their presence in corresponding samples. In a second step, the average exact masses and RT of the consensus feature list are scanned in each sample to check if any feature has been missed in the first round.

  1. Selection of relevant features

The feature list was imported into Mass Profiler Professional (MPP, Version 14.0, Agilent Technologies, Inc.) which is a statistical analysis software package designed to evaluate high-resolution mass spectrometry data. To improve robustness, features were discarded if they were not found in at least two out of three replicates or if the highest intensity in the samples was less than ten times the intensity of the blank sample. In order to focus on compounds ubiquitously present in dust, all features present in at least 37 of the 38 samples were selected for further identification.

  1. Compound identification using in-silico fragmentation

The samples with the highest intensities of the selected features were re-run in targeted MS/MS mode (CE=20) by triggering the [M-H] - or [M+H] + mass of the selected feature at the measured RT. Using the MS/MS information, the features were tentatively identified (if possible) using two in-silico fragmentation software packages: Agilent Molecular Structure Correlator (MSC) and the program MetFrag 53 (online version https://msbi.ipb-halle.de/MetFragBeta/ and MetFragR http://c- ruttkies.github.io/MetFrag/projects/metfragr). Both software tools have the same principle, but a different algorithm and different filtering and weighting options. Briefly, input parameters for both programs are the exact mass of the [M-H] - or [M+H] + ion and the list of the acquired MS/MS fragment masses and relative intensities. The software searches all compounds with the corresponding exact mass (± the chosen mass error, in this case ±10 ppm) in a database. In this study, MSC searched the ChemSpider database (www.chemspider.com), and MetFrag was set to search PubChem (https://pubchem.ncbi.nlm.nih.gov). In a next step, the fragmentation pattern of every candidate is simulated based on a given fragmentation algorithm and a match score between the acquired and predicted MS/MS spectra is calculated. If no other criteria are selected, the candidates are ranked based on this fragmenter score. In MSC, candidates can be ranked by the number of data sources in ChemSpider. MetFrag has more options in this respect. The number of references and patents from PubChem can be integrated in a weighted score and the user can define the importance of each score by a weighting factor. In addition, a suspect list (csv. file containing InChIKeys) can be added. In this case, a candidate that is listed on the suspect list is ranked higher compared to a candidate that is not listed. These weighting options help in selecting the correct compound if there are multiple candidates with similar fragmenter scores due to similar structures. In this study, the weighting factor for the fragmenter score was 1.0, for the number of references 0.125, for the number of patents 0.125, and for the suspect list 0.25. The suspect list in this study was a merged list of all suspect exchange lists from the NORMAN network (http://www.norman- network.com/?q=node/236 54), an unpublished temporary list from DSSTox (desalted compounds, received from Mark Strynar, EPA) and the CPCP database 49 (>18,000 compounds). The online version of MetFrag cannot be used in a batch mode. Therefore, MetFragR was used and a batch version was programmed to run batches of 20 compounds.

The candidate list for each feature was manually checked and the most plausible structure was selected. In cases where it was determined to be necessary or advantageous, additional lines of evidence were considered, e.g. plausibility check of the retention time based on the predicted logK ow value or the MS/MS match score from the fragmentation prediction software CFM-ID (http://cfmid.wishartlab.com) 55.

Chemical use categorization

Chemicals identified in the current study were classified into the most appropriate and primary use categories to understand distribution and concentration by different use categories. For most compounds, we relied on databases such as CPCPdb and the U.S. National Library of Medicine's Household Product Database ( https://householdproducts.nlm.nih.gov/index.htm) to find the most common and primary use of the compounds. For compounds with multiple uses, we also relied on web searches to categorize them.

Statistical analysis

Statistical analyses were performed using Microsoft Excel 2017. For concentrations between the limit of quantification (LOQ) and the limit of detection (LOD), we assigned a value of LOQ divided by 2. For concentrations below LOD, we assigned a value of LOD divided by the square root of 2 56. The variability of chemical concentrations in residential dust across homes provides important insight into sources of particular compounds. Chemicals with low variability are likely to be found in a wide range of products and/or in widely used products, while those with high variability are likely to have more specific sources or to be associated with more episodic activities (e.g., house-level pest control activities). We examined the variability of concentrations for all compounds that were confirmed with a reference standard and detected in more than 50% of samples.

Source Strength

We then use a fugacity-based indoor mass balance model to estimate the whole-house emissions rates of SVOCs that would account for the measured dust concentrations. The method follows Shin et al. (2013)57.

Results and Discussion:

Method Validation Procedure for Targeted Analytes

Method validation showed good absolute recoveries (extraction recovery of a triplicate spike sample) above 75% for more than 80% of the 76 GC-Q/TOF target compounds and above 50% for more than 80% of the 56 LC-Q/TOF compounds. The extraction was somewhat biased towards non-polar compounds by the selected solvents (hexane:acetone 3:1 and 100% acetone). However, more polar solvents such as methanol would not have meshed well with a single sequential extraction because of its immiscibility with hexane and its elevated boiling point. Nevertheless, method detection limits (MDL) were generally lower for LC-Q/TOF compounds. In total, 50% of all compounds had MDLs below 10 ng/g dust and 80% had MDLs below 100 ng/g dust. These MDLs are comparable with MDLs from previous literature studies58-61. Compounds with higher MDLs either had extremely high concentrations in the dust (phthalates, organophosphate flame retardants, skin oils) and background contamination in the method blank or they had limited sensitivity in GC EI-MS mode (pyrethroids, phenols). Precision (standard deviation of replicates) was <20% for 95% of all compounds. In addition, the accuracy of the concentrations could be checked for 14 compounds for which certified concentrations in the SRM 2585 dust were available (i.e., deviation of the measured value from the certified value). For 12 out of 14 compounds, accuracy was < 25%. The only exceptions were phenanthrene and pyrene, for which concentrations were underestimated by 33% and 26%, respectively. In LC-Q/TOF, ion suppression due to matrix load was low for most compounds measured in negative ionization mode (90% < factor of two), but higher for compounds measured in positive ionization mode (60% > factor of two). Therefore, the use of appropriate internal standards was absolutely necessary to accurately quantify the compound concentrations.

Overall, the quality control parameters of this simple and extremely broad extraction method followed by two untargeted analytical methods show that the compound classes previously described to be present in dust can be efficiently and accurately extracted and detected.

Results of GC-Q/TOF Non-Target Screening

The deconvolution of the GC-EI-MS chromatograms produced about 3000 to 5000 features per sample; roughly 300 of these compounds per sample had a NIST library hit (component shape quality > 60, library match factor > 60). An example compound, octyl methoxycinnamate (CASRN: 5466-77-3) - a UV- filter later confirmed with a reference standard - is shown in Figure 1.1A. The perfect deconvolution (component shape quality: 99) is indicated by the co-elution plot of the five main fragments. The good match with the hit in the NIST library (match factor: 87.5) is underlined by the differential plot. In addition, the experimentally derived RI in the NIST library perfectly matches with the measured RI in this study. The compound was detected in 36 out of 38 dust samples.

The manual grouping and prioritization (see method section) led to 75 compounds with detections in multiple samples. Twenty-six of them were discarded due to high presence in the blank or RI deviation above the selected criterion. Twenty-two of the remaining 48 compounds were target compounds (BDE, OP-FR, pyrethroids, phthalates) that were already confirmed; the remaining 27 were identified uniquely through this non-targeted workflow. For instance, 7,9-Di-tert-butyl-1-oxaspiro(4,5)deca-6,9-diene-2,8- dione (CASRN: 82304-66-3) - a leachable from plastics - has not been reported in dust before, but is of increasing interest because of detections in water (leached from pipes) and in airborne particles62. For 17 of the non-targets, a reference standard could be purchased and all identifications were confirmed by matching RT. For the remaining 10 compounds, no reference standard could be purchased; they remain tentatively identified with confidence level 252 based on matching EI spectra and RI values.

Results of LC-Q/TOF Suspect Screening

The screening of the LC-Q/TOF chromatograms acquired in the All-Ions fragmentation workflow with two PCDLs containing almost 10,000 chemicals with MS/MS spectral information led to 97 tentatively identified compounds after applying the automatic filter criteria and after manual inspection. The approach is discussed in detail for water samples in Moschet et al. (2017)50. Seventeen of them were already quantified in the target screening, six were target chemicals from the GC-Q/TOF and two were non-targets identified by GC-Q/TOF. For 52 suspects, a reference standard was purchased; 43 of them were unambiguously confirmed by matching RT; 9 were rejected due to non-matching RT. The remaining 28 compounds remain tentatively confirmed with confidence level 252.

One example of a positively identified compound is the fungicide imazalil (CASRN: 35554-44-0, molecular formula C14H14Cl2N2O), which is used to preserve citrus fruits and is likely to be carcinogenic to humans (Figure 1.1B). The mass error was +4.5 ppm and the isotope pattern explains the two Cl-atoms which resulted in an isotope match score of 97 out of 100. In addition, the five main fragments (see SI-6) co-elute with the [M+H]+. The compound was unambiguously confirmed by matching retention time of a reference standard. It was detected in 29 of the 38 dust samples.

These results show the efficiency of suspect screening using curated spectral libraries and automated software workflows, allowing identification of dozens of compounds without procuring thousands of standards or optimizing target methods. For example, 21 pharmaceuticals - e.g. diphenylhydramine (CASRN: 58-73-1), miconazole (CASRN: 22916-47-8), diclofenac (CASRN: 15307-86-5) - which have previously not been investigated in house dust, were detected by this approach.

Figure 1.1. A) Example of the UV-filter octyl methoxycinnamate (CASRN: 5466-77-3) detected by GC- Q/TOF (non-target screening). Top: Co-elution plot of five main deconvoluted fragments in a real sample. Bottom: Differential plot between deconvoluted spectra and the NIST library spectra. The match factor was calculated by MassHunter Unknown Analysis software. The identity of the compound was later confirmed by a reference standard. RI: retention time index, exp.: experimental; B) Example of the fungicide imazalil (CASRN: 35554-44-0) detected by LC-Q/TOF (suspect screening). [M+H]+ and five main fragments from the All-Ions scans in a real sample (top) and in the reference standard (bottom). Inset: Isotope pattern match including the monoisotopic mass [M+H]+and five isotopes (M+1 to M+5). Black lines reflect the measured isotopes, red boxes reflect the theoretical isotope pattern.

Results of LC-Q/TOF Non-Target Screening

  1. Quality Control in Non-Target Screening using LC-Q/TOF

The recursive feature extraction of the 149 triplicate injections detected 13,340 individual features in negative mode and 14,588 features in positive mode, respectively. Features that were only found in one out of three replicates (roughly 30% of total features) and features present in the blank (roughly 10%) were discarded leading to a new total number of features of 7,701 in negative mode and 9,326 in positive mode. Identification of all these features is not feasible35so a statistical analysis, explained below, was used to focus on relevant compounds.

As classical quality control using validation parameters (e.g. recovery, accuracy) is not possible when doing non-targeted analysis, data quality assessment needs to be demonstrated differently; two proxies for this are described below. Only after the reproducibility and accuracy of the screening approach is verified should statistical analysis or compound identification be performed.

The first approach is to examine the reproducibility of features among replicate samples, e.g., using principal component analysis (PCA; see Figure 1.2 for negative ionization mode). The plot shows that the injection replicates cluster close together (triangles with the same color). The clustering is clearly visible, although components 1 and 2 only explain <10% of the variation, which is due to the fact that several thousand features are compared. This shows that i) RT and mass accuracies were stable over the four day run of the 149 injections and ii) that the recursive feature extraction algorithm grouped the features accurately and reproducibly. The RT shifts of the internal standards throughout the sequence were <0.2 min in negative mode and <0.4 min in positive mode.

The extraction replicates of the five selected samples (squares in Figure 1.2, indicated with a colored circle) also grouped within similar distances to the injection replicates of the same samples. This means that dust is homogenous enough to obtain similar results when extracting different sub-samples multiple times.

The second approach is to check for known chemicals in the untargeted feature list. Most of the targets and suspects that were detected in the dust samples were found as features in the unfiltered feature list (40 out of 48 compounds with >5 detections in negative mode, 21 of 30 compounds in positive mode). Reasons for missing a compound could be that it fell below the selected intensity threshold or due to occurrence in the blank. The automated criteria were set more stringently than when manually evaluating the data; thus, the manual evaluation leads to lower detection limits.30

However, the fact that most target compounds were found shows that relevant compounds were isolated using the recursive extraction algorithm and not just compounds with much higher intensities such as surfactants (see next section).

Figure 1.2. Principal component analysis (PCA) of the detected non-target features in 38 dust samples and the NIST SRM 2585 dust sample on the LC-Q/TOF in negative ionization mode. Different colors indicate different samples. Each sample was injected in triplicate (triangles). Total number of features: 7,701 (blank subtracted). Samples indicated with a colored circle had additional extraction replicates (squares). The indicated light blue sample is the NIST SRM 2585 reference dust sample.

Homologous Series Identification

The total ion chromatogram of the dust samples suggested that homologous series of compounds were present, so all features were searched for homologues using the software EnviHomolog (www.envihomolog.eawag.ch)63. Interestingly, 50% of the features in negative mode and 30% of the features in positive mode were identified as homologues by the software (SI-8). Most prominent in negative mode were homologues with a mass defect of 44.0262, i.e. (-CH2CH2O)n (~50% of the homologues). The most prominent homologues in positive mode were identified with a mass defect of 14.0156, i.e. (-CH2)n (~40% of the homologues). Cleaning agents usually contain surfactants with a homologous series of compounds. They have been detected in different environmental media (e.g.64), and it is expected that surfactants end up in the dust. Examples of surfactants with (-CH2CH2O)n chains are alcohol polyethoxylates (AEOs) and polyethyleneglycols (PEGs); examples of surfactants with (-CH2)n chains are linear alkylbenzenesulfonates (LASs) and sulfophenyl carboxylic acids (SPCs)31,65,66 One way to identify expected surfactants is to use the NORMAN exchange list54 that contains surfactants previously identified in waste water31 using a suspect screening approach. Although numerous other types of surfactants and also other compounds with homologous series such as polyfluorinated or polyhalogenated compounds might be present in dust67, the identification of these individual compounds is outside the scope of this study.

  1. Compound Identification using In-Silico Fragmentation Software

As it is impossible to identify several thousands of non-target features35, we have chosen to prioritize features that were ubiquitous in the dust samples. Therefore, features present in ≥ 37 out of 38 samples were selected and identified, if possible (see method section). These included 611 features in negative mode and 284 in positive mode. To further refine that selection, ubiquitous features with stable intensities amongst the samples (coefficient of variation CV<75%) and features with significant intensity fluctuations amongst the samples (CV>200%) were selected. This led to 129 features in negative mode and 99 features in positive mode. Good MS/MS spectra were acquired for 57 features in negative mode and 75 features in positive mode. The remaining features had insufficient intensity or no/poor fragments.

The features with good MS/MS spectra were examined using the two in-silico fragmentation software packages MSC and MetFrag. An example of a compound identified by using both packages is the ionic surfactant N-Lauroylsarcosine (mass 270.2068, RT 10.5 min), which is used in shampoos and shaving foam and which has not been detected in house dust before. MS/MS information first helped to confirm the molecular formula with both MSC and MetFrag producing top candidates with the formula C15H29NO3. The isotope pattern score of 99.7 in the MS full scan supported this. In MSC, the top candidate N-Lauroylsarcosine (CASRN: 15535-18-9) had a fragment match score of 87.5 (mass error -1.2 ppm) with 95% of the fragments being explained by its structure. In MetFragR, the compound had a fragmenter score of 70.7, only rank 33 among all candidates. However, the compound had the highest number of references and patents and was listed on the custom suspect list. Therefore, the compound had the highest rank with the weighted score. The estimated logKow (4.5, MetFrag output) is consistent with the measured RT of 10.5 min. N-Lauroylsarcosine reference standard was purchased, and its identity was confirmed with matching RT and MS/MS spectra.

This example illustrates how multiple pieces of evidence support correct identification. MetFrag was frequently favored over MSC due to its wider range of program capabilities. Nonetheless, both software packages are very useful in identifying chemicals through the non-targeted workflow.

Another tentative compound identification later confirmed with a reference standard was the emerging organofluorine compound 6:2/8:2 diPAP (Polyfluoroalkyl phosphoric acid diester, CASRN: 943913-15-3, see Figure 1.3). Although the isotope pattern of 6:2/8:2 diPAP is not distinctive because it does not contain Cl- or Br-atoms, the negative mass defect indicates the presence of multiple F-atoms. The eight top fragments of 6:2/8:2 diPAP could all be explained by its structure, with the compound receiving the highest fragmenter score by MetFrag. In this case, neither the suspect list nor the number of references/patents helped because none of the candidates had any entries. Two other emerging organofluorine compounds - 6:2 fluorotelomer sulfonic acid (6:2 FTSA, CASRN: 27619-97-2), 6:2diPAP (CASRN: 57677-95-9) - were detected by the non-target approach and later confirmed by a reference standard. Emerging organofluorine compounds, especially diPAPs, have recently been found in dust samples in high concentrations and detection frequencies68 and are an underestimated source of human exposure to polyfluorinated compounds.

Another example is the fungicide metabolite 4-hydroxychlorothalonil (CASRN: 28343-61-5), which has not been detected in house dust before. The isotope pattern indicated the presence of three Cl-atoms and the top five fragments could be explained by its structure. However, three structural isomers had the same fragmenter score by MetFrag. Of these, only 4-hydroxychlorothalonil was on the suspect list, accompanied with the highest number of references and patents.

With this approach, 75 compounds were tentatively identified with a proposed structure either in negative or positive ionization mode. Four of them were already identified by the target or suspect approach (bisphenol A, dexpanthenol, fipronil-sulfone, triclocarban). For 16 non-target candidates, a reference standard could be purchased. Twelve compounds were confirmed by matching RT and matching MS/MS spectra. In addition to the aforementioned compounds, these were: vanillin (CASRN: 121-33-5), genistein (CASRN: 446-72-0), palmidrol (CASRN: 544-31-0), linolenic acid (CASRN: 463-40-1), palmitic acid (CASRN: 57-10-3), leucine (CASRN: 61-90-5), and piperine (CASRN: 94-62-2). Four compounds were not confirmed (methyl-2-octynoate, cinnamic acid, diphenyl phosphate, dibutyl- phthalate). The remaining 55 compounds remain tentatively confirmed with confidence level 352. Sixteen additional compounds were identified only by a proposed molecular formula.

Figure 1.3. Chromatogram (top), isotope pattern (inset) and annotated MS/MS spectra (bottom) of 6:2/8:2 diPAP (CASRN: 943913-15-3) identified by the in-silico fragmentation software MetFrag by a complete non-target approach in LC-Q/TOF negative ionization mode. MetFrag fragmenter score: 177, number of explained peaks: 8, number of references/patents (PubChem): 0/0, suspect list: no, mass error of precursor mass: 1.3 ppm, mass error of fragment masses: 0.2 ppm (m/z 542.9659) to 4.7 ppm (m/z 78.9590). The estimated logKow (10.6, Jchem for Excel) is consistent with the measured RT of 14.1 min.. The feature was detected in ≥37 out of 38 samples and was later confirmed by a reference standard.

Comparison of LC-Q/TOF and GC-Q/TOF workflows for detecting unknown chemicals

The identification of compounds using GC and LC techniques provides complimentary yet unique capabilities while providing a complete chemical profile of dust samples. Both analyses provide several thousands of detected non-target features and it is important to prioritize the most relevant features35 either by statistical analysis or by previous knowledge about suspected occurrence of certain compounds.

The fact that LC-ESI-MS provides the molecular ion information while GC-EI-MS generally does not, necessitates distinct non-targeted screening workflows. Both platforms have advantages and drawbacks. The biggest advantage in GC-EI-MS is that the fragment spectra are very reproducible and that libraries containing over 200,000 compounds are available. In addition, relative RT are very reproducible, so that normalized RI can be calculated when using a standard column and a simple temperature gradient. Both help to tentatively identify known unknowns with high confidence when the compound is in the library, saving significant time, labor and cost by avoiding the need to procure, prepare and analyze every analytical standard. In this respect, a good deconvolution software package and/or a good separation is essential to obtain the correct spectrum. As a drawback, relatively few curated and reliable accurate mass library spectra are presently available. Another drawback is the low or missing molecular ion, which would otherwise allow use of the suspect screening approach used for LC-Q/TOF. Also, this makes it much more difficult to detect unknown unknowns, i.e., compounds that do not have an EI spectrum in the library.

LC-Q/TOF software processing tools are more advanced, making it easier to bin and align non-target features in multiple samples. For compound identification - or for compounds lacking MS/MS spectra - the approach of acquiring MS/MS spectra and running an in-silico fragmentation is promising, though still largely a manual effort. Interestingly, the hardware and software tools from classical GC-MS and LC- MS are increasingly being integrated into a single, simpler platform to maximize data processing and integrity. For example, a GC-HRMS operated with a soft ionization source (low energy EI, PCI, APCI)40,69 generates data that are comparable with LC-ESI-MS/MS data. On the other hand, the All-Ions fragmentation workflow uses elements from the GC-EI-MS where multiple fragments are co-eluted to form a specific spectrum. Further advances in joint non-target screening by LC-HRMS and GC-HRMS depend critically on the optimization of the different software tools.

The results of this study show that both instrument types - LC-Q/TOF and GC-Q/TOF - are indispensable for a complete identification of chemicals in a sample. In the 38 dust samples, we detected and identified 86 compounds by GC-Q/TOF (59 targets and 27 non-targets) as well as 204 compounds by LC- Q/TOF (42 targets, 79 suspects, 83 non-targets,)). The actual number of compounds present in the dust is much higher, though. For example, many hydrocarbons were detected on GC-Q/TOF, and numerous surfactants were detected by LC-Q/TOF. These chemicals were not further investigated.

The detected and identified compounds are shown in Figure 1.4. As expected, GC-amenable compounds are generally in the higher logDow range than LC-amenable compounds. However, there are quite a few exceptions: e.g. diPAPs, dioctadecylamine with high logDow detected by LC-Q/TOF, or triethyl citrate and coumarin with lower logDow detected by GC-Q/TOF. There is also an overlap of compounds that can be detected by both instrument types. Sixteen of the 86 compounds that were detected by GC-Q/TOF were also detected in a comparable number of samples by LC-Q/TOF (Target or Suspect screening approaches, see Figure 1.4). Vice versa, three compounds that were detected by LC-Q/TOF were also detected in a comparable number of samples by GC-Q/TOF. The chemicals detectable on both platforms were phthalates, organophosphate flame retardants, UV filters and fipronil and its metabolites, which is consistent with results from a collaborative trial in water samples.70

Figure 1.4. Physico-chemical properties of detected chemicals by LC-Q/TOF, GC-Q/TOF and compounds detected by both platforms with the different identification workflows (target, suspect, non-target). The logDow (at pH 7) was calculated by ChemAxon (JChem for Excel). The homologous series between mass 200-800 at logDow <0 are polyethylene glycol (PEG) surfactants that were detected by the LC-Q/TOF non- target approach.

Detected compounds in dust

By applying target, suspect screening, and non-target approaches to GCMS and LCMS, 271 compounds were detected in at least one of the 38 household dust samples. For 16 compounds, identification was not possible and only molecular formula (e.g., C4H7FO) could be assigned with high confidence 71. Among the 257 compounds that could be identified with structure and formula, 163 were unambiguously confirmed and quantified by reference standards, and 94 compounds were tentatively assigned by matching literature or library spectra. We confirmed that 30 target compounds were not detected in any samples. The majority (~70%) of these non-detected compounds were biocides, including fungicides, herbicides, and insecticides.

Using targeted analysis with GCMS and LCMS, we could definitively detect and quantify concentrations of chemical classes commonly found in indoor dust samples, including fragrances, parabens, insecticides, phthalates, OP-FRs, PBDEs, other FRs and plasticizers, PAHs, PFAS, phenols, and skin oils (see Table 1.1). BDE-29 has been widely detected in California house dust 72,73. However, we could not detect BDE-29 in our samples on our instruments, because it is a very large (molecular weight

= 959 g/mole) and completely non-polar (logP = 11.25) compound and is not volatile enough to get mobilized on the GC column we used. Thus, it could not be efficiently measured on our instruments. For plasticizers and skin oils (e.g., squalene) that are also ubiquitous in blanks, detections and concentrations reported in the current study were further confirmed by discarding all concentrations that were not at least 10 times higher than those in the blanks 71.

Tentatively identified compounds using the three screening approaches (GC non-target, LC non-target, LC suspect) were largely those used as cosmetic ingredients, natural compounds originating from plants, animals, or humans, food-related compounds, fungicides, pharmaceuticals, and surfactants. A large portion of previously unknown UV filters and compounds used in PCPs was also detected via suspect and non-target screening approaches.

Table 1 .1. Summary of detected compounds by different analytical instruments and approaches. Values in parentheses indicate the number of compounds confirmed by a reference standard.

Newly-measured compounds in household dust

In the present study, 118 compounds were identified and quantified for the first time in household dust (Figure 1.5). Some of these compounds were previously measured in U.S. wastewater samples via target analysis, but had not been measured in indoor dust. The majority of these compounds was detected via LC non-target (46%) and LC suspect (29%) approaches (see inset of Figure 1.5). These newly-measured compounds mainly comprised surfactants (n = 24), pharmaceuticals (n = 19), compounds with unknown use information (n = 16), and human metabolites (n = 12). We also identified 10 biocides (4 insecticides, 6 fungicides) and 6 phenols (some are also used as biocides) in dust for the first time via GC target and/or LC target analyses.

Figure 1.5 . Number of newly-measured compounds in household dust (n = 118) by use category and by different analytical instruments and approaches.

Measured dust concentrations

Eighty-eight compounds were detected in more than 50% of samples, 57 compounds in more than 90% of samples, and 33 compounds in all samples. To investigate which compound classes were in high concentration, we summarized measured median concentrations for the 88 compounds in more than 50% of samples by five levels (<500, 500-1,000, 1,000-5,000, 5,000-10,000, >10,000 ng/g of dust; Figure 1.6). A large number of UV filters, phthalates, and OP-FRs were detected in our dust samples and some of them were at concentrations above 10,000 ng/g of dust. Fungicides, PBDEs, insecticides, PCPs, PFAS, and pharmaceuticals were also abundant in our samples, but most measured at concentrations below 500 ng/g of dust. Median concentrations greater than 10,000 ng/g of dust were observed in four compounds that comprise or are found in skin surface lipids (cis-hexadec-6-enoic acid, squalene, cholesterol, vitamin E); two phthalates (bis(2-ethylhexyl) phthalate; DEHP), dioctyl terephthalate; DOTP); two cosmetic ingredients (linoleic acid, glycerol tricaprylate); one OP-FR (tris(1-chloro-isopropyl) phosphate; TCIPP); and one UV filter (oxybenzone). Solid consumer products (i.e., DEHP, DOTP, TCIPP) and PCPs (i.e., oxybenzone) are well-known sources of SVOCs. High concentrations of other compounds in the current study highlight the roles that humans and their activities, and possibly pets, play as sources of SVOCs in the indoor environment.

Figure 1.6 . Summary of median concentrations (ng/g of dust) for 88 compounds (target + suspect + non- target) detected in more than 50% of samples.

Variability and magnitude of individual compounds across samples

Figure 1.7 displaying distributions of targeted compounds only. This figure excludes two phenols (tetrachloro phenols, cresol) that were detected in most samples (above LOD) but that were below LOQ. Overall, dust concentrations varied by almost three orders of magnitude across household samples and by almost four orders of magnitude across compounds. PFAS were measured in the lowest concentrations and had relatively large variability in concentrations among target compounds. DEHP was shown to have the smallest variability (CV = 0.35) across the samples.

Unexpectedly, three compounds, including one used as a surfactant (alcohol ethoxylates), one used in various solid items including rubber footwear and automobile tires (1,3-diphenylguanidine), and one used as a disinfectant or a surfactant (dodecyl dimethyl ammonium chloride) were also measured in high concentrations, with medians greater than 2,500 ng/g of dust (empty boxes in Figure 1.7). Except for tri-n-butyl phosphate (TNBP), OP-FRs were measured in higher concentrations than PBDEs, and bisphenol S (BPS) was measured in higher concentrations than bisphenol A (BPA), consistent with recent changes in consumer use due to changes in product formulation and regulations affecting PBDEs and BPA.

Figure 1.7 . Distributions of dust concentrations (ng/g of dust) for 56 target compounds detected in more than 50% of samples. Two skin oils (cis-hexadec-6-enoic acid, squalene) were grouped into "cosmetics' with linoleic acid in this figure. Compounds with asterisk (*) indicate the first measurement in household dust (n =13).

Calculated Source Strengths

Based on the measured concentrations, the source strength of the compounds was calculated using a fugacity-based model, as shown in Figure 1.8.

Figure 1.8 . Calculated source strengths of targeted compounds measured in house dust.

Implications

A major achievement of this study was the identification and quantification of a large number of SVOCs with high confidence (78% of compounds were fully confirmed with a reference standard or tentatively confirmed with a matching mass spectral library). We were able to expand the list of compounds present in indoor dust because both LCMS and GCMS with all available analytical approaches were applied at the same time. For example, 75% of newly-measured chemicals were observed via LC non- target or LC suspect approaches and the newly-measured compounds in our study mainly comprised surfactants, pharmaceuticals, and human metabolites.

Our study also supports the idea that dust can serve as a marker of use. The presence of food additives or preservatives in dust indicates that they exist outside their intended use, which is to be consumed via direct food intake. Skin oils and cholesterol were ubiquitously measured in Danish homes and daycare centers 74. In addition to these compounds, we observed cosmetic ingredients and vitamin E with median concentrations greater than 10,000 ng/g and some food additives (e.g., caffeine, sorbic acid). In a separate study in which we analyzed skin wipe samples 75, 11 compounds (triethyl citrate, butylated hydroxytoluene, cholesta-3,5-diene, vitamin E, cholesterol, tridecanoic acid, arachidonic acid, palmidrol, palmitic acid, pentadecanoic acid, linolenic acid) were detected, and they were also detected in our dust samples. This indicates that human activity, including cosmetics, skin sloughing, dropping food residue or debris unintentionally on floors, could be sources of these compounds.

The measured dust concentrations of 144 compounds in the current study can be used to improve our understanding of residential chemical exposure via non-dietary exposure routes 76. Some compounds detected in our study have exposure routes via direct oral intake (e.g., pharmaceuticals), direct dermal uptake (e.g., cosmetics, fragrances, pharmaceuticals, PCPs, UV filters) or direct food intake (e.g., food additives, preservatives). Except for these compound classes, non-dietary exposure routes for others that occur indoors include inhalation (gas-phase, airborne particles), air-to-skin dermal uptake (gas- phase), and non-dietary dust ingestion (settled dust). Chemical properties (volatility, octanol-air partition coefficient) determine the compound's distribution among gas-phase, airborne particles, and settled dust. Accurate determination of concentrations in each of these three indoor media are essential to characterize indoor residential exposure. In the absence of measurements in the gas-phase and for airborne particles, measured dust concentrations can be used to estimate concentrations by applying the partitioning relationships between settled dust and the gas-phase and between the gas-phase and airborne particles 11,12. Subsequently, total exposures via inhalation, transdermal uptake, and dust ingestion can be estimated by applying standard human exposure factors (inhalation rate, exposed skin surface area, etc.) to the measured dust concentrations and the estimated gas- and particle-phase concentrations 13. In addition, estimated total exposures from measured dust concentrations can be used in high-throughput screening and prioritization of indoor chemicals by exploring their association with toxicity potential.

Our findings have important implications. Because some compounds are used variously in indoor environments and do not fall into a single use category, it is difficult to characterize their sources and source strengths. However, measured dust concentrations can be used to determine overall source strength of SVOCs aggregated from all sources, because dust is known to be a reservoir of chemicals used indoors 57. In addition, our results can be representative of those in the general population and thus can be used to compare with those reported in other studies to examine overall time trends in dust concentrations or spatial variability 77.

Objective 2: Refine and evaluate a multi-compartment indoor fate, transport, and exposure model

  1. Provide data to evaluate indoor models by measuring half-lives of penta-brominated diphenyl ethers (BDEs) after source removal . One shortcoming of indoor modeling is the inability to conduct model evaluation. Thus, a small, focused field study will be conducted where couches with penta-BDEs, the primary source of these compounds, will be removed and decreases in indoor levels will be observed over time. This will provide a much-needed evaluation data set to compare to model estimates.
  2. Collect field data to improve estimates of dust removal rates and surface dust loading . Model predictions indicate that the least volatile compounds have among the highest levels of human exposure per unit released. Dust loading and dust removal rates are critical input parameters that influence exposure to these compounds. As there is little existing data, we will collect data for these parameters to improve exposure estimates.
  3. Update equations for partitioning, transport, and exposure processes in the model . We will evaluate current model assumptions for a number of processes against recently collected equations and field data available in the literature.

The subtasks of this objective are each very different and thus are discussed independently. The information for Objective 2a has all been produced and is discussed below, but the modeling is still ongoing as it was not as straightforward as originally thought. The samples for Objective 2b were collected, but have not yet been evaluated and thus are not presented. Two separate modeling activities were completed for Objective 2c, and are discussed separately, as Modeling Exposure from Cleaning Products and Modeling Exposure from Dust Concentrations.

Objective 2a: One way to evaluate environmental models is by removing a source of a compound to a home and evaluating the rate of decrease of the concentration of the compound in the home, and determine if it decreases at the same rate as predicted by the model. Couches have traditionally been a source of PBDE and other flame retardants in the home. California passed regulation allowing sale of couches without chemical flame retardants. By recruiting homes planning on removing their old couch, a source of flame retardants from their home, and replacing it with a couch that is not a source, a data set can be assembled to evaluate the change in concentration over time.

Methods

Recruitment and enrollment

Households interested in replacing the couch in their main living room in their home were recruited for this study. All households were in the Bay Area in California. Recruitment consisted of convenience sampling with the assistance of local non-profits. This study was approved by the UC Davis IRB.

Each interested resident was screened for eligibility using a screening questionnaire and inspection of their current furniture in the main living are of their residence. The criteria for eligibility included, adults 18 years and older, able to provided informed consent in either English or Spanish, and having at least one couch or sofa in the main living area of their home that likely would contain flame retardants (not antique or appearing to be manufactured prior to approximately 1970 and not tagged with a label indicating that the couch does not contain added flame retardant chemicals. In addition, the participant was planning to remove their couch with in the following year.

Once enrolled in the study, an initial home visit was scheduled to collect and initial dust sample. At 6- months, 12-months, and 18-months following the removal of the source in the homes, a home visit was conducted to collect additional dust samples from the home.

Dust Sample Collection

Dust samples were collected using a Eureka Mighty-Mite vacuum cleaner (Eureka Model 3670), with a crevice tool attachment that was modified to collect dust in a cellulose extraction thimble (Whatman International, Ltd.), a protocol used in prior studies 28,78-81. The main living room of the home was sampled. The equivalent of the floor's entire surface area was sampled by gently drawing the crevice tool attachment across all horizontal surfaces in the room. Dust was not collected from any of the furniture that was to be replaced (initial visit) or from the replacement furniture (6, 12, and 18 month visits) or from under any furniture. After sample collection, the cellulose thimble was removed, wrapped in pre-cleaned foil, placed in a zip top bag, and transported to the field office on ice. Samples were then stored at -20°C. All equipment was cleaned between sampling events.

Chemical Analysis of Dust

The dust was analyzed for the compounds listed in Table 2.1 by the methods discussed in Objective 1 of this report.

Table 2.1. 5 PBDEs and 7 OP-FRs were included as chemicals of interest in this study.

Results

A total of 28 households were recruited and completed the initial visit. Twenty-two households successfully replaced their couch within the study timeframe and completed the 6-month follow-up dust collection. Twenty-one households completed the 12-month study visit and the 18-month study visit.

Concentrations at the initial visit and the visit 6-months following the removal of the source were compared using a t-test. All of the PBDEs had a statistically significant decrease in concentration, as shown in Figure 2.1. Most of the other flame retardants also showed a decrease, with the exception of TBOEP and TDTBPP. Using the simple linear regression method, we estimated the monthly percent change for the compounds that have decreasing trends verified from the t-test. Comparisons of the distributions of measured concentrations can be seen in Figures 2.2 and 2.3.

Figure 2.1. This is the results of t-test to confirm the difference in means between 0 and 6 months. Most of the compounds have decreasing trends during the period.

Figure 2.2 . This is for the comparison of PBDEs' geometric means (GMs) between 0 and 6 months. Every individual PBDE compounds as well as the average of total PBDEs exhibit significant differences in their GMs over the period.

Figure 2.3 . This is for the comparison of OP-FRs' geometric means (GMs) between 0 and 6 months. They look like there are significant overlaps in 95% CIs, but anyhow the first 5 compounds as well as the average of total OP-FRs have statistically significant differences in their GMs over the period according to the t-test.

We aimed to determine how effective the couch replacements are for reducing the dust concentrations of PBDEs and OP-FRs after 18 months. The GMs in this figure are based on not individual compound but average of total PBDEs or OP-FRs. As you can see in the figure, the total PBDEs and OP-FRs concentrations were reduced by 52.5% and 12.2%, respectively, after 18 months since the couch replacements. One can see that the majority of the decrease occurred in the first 6 months.

Figure 2.4. Distributions of concentrations over the full 18 month study period.

Objective 2c: Models of exposure were advanced in two ways. First, a model framework for integrating multiple exposure pathways for chemicals in household cleaning products was developed. Second, a model for estimating exposures based on indoor dust concentrations was developed.

Objective 2c: Household cleaning products

The indoor use of household cleaning products is intended to promote hygiene and comfort in the indoor environment 82, but can lead to exposure to primary and secondary air pollutants 83-85. Exposure models for cleaning product use are available 86-88. However, chemical properties are not parameterized as input parameters in the first two models 86,87, which thus only allow calculating exposure of a product from a specific cleaning activity (e.g., daily intake rate of a dish detergent from hand dishwashing). We note that exposure from cleaning product use also varies among different chemicals because chemical properties primarily determine the degree of volatility and the portion disposed down the drain 89, and dermal permeation 90. The U.S. Environmental Protection Agency (EPA) Office of Pollution Prevention and Toxics developed a consumer exposure model to estimate potential exposures from the use of various consumer products, but the model only captures the exposure during the use phase 88. This highlights the need for the development and evaluation of exposure models that parameterize key chemical properties for individual compounds and capture exposure from other overlooked exposure phases (e.g., exposure due to down-the-drain disposal). To provide a realistic magnitude of exposure from cleaning activities and capture all plausible sources and routes of exposure to cleaning product chemicals, a novel model framework that integrates exposure during cleaning activities, exposure due to indoor releases from the activities, and exposure due to down-the-drain disposal is needed.

In this study we propose models which can be used to predict improved estimates of exposure to organic compounds formulated in three types of cleaning product uses (i.e., laundry detergents, dish detergents, surface cleaners). The objective of this study is to develop a model framework that integrates chemical exposure during each of the five cleaning activities (i.e., hand dishwashing, machine dishwashing, machine clothes washing, countertop surface cleaning, and floor surface cleaning). To meet this goal, we integrate exposures from various fate and exposure models.

Scope and overview of the model

In this study, we assume that exposure occurs in three phases as a result of cleaning product use: (1) exposure during/after product use, (2) exposure due to indoor releases, and (3) exposure due to down- the-drain disposal. Using compounds associated with hand dishwashing as an example, a person who washes dishes with bare hands might be exposed to chemicals via dermal uptake and inhalation. In addition, the portion that is released indoors from the hand dishwashing activity will be another exposure source to other occupants in the same residence. Likewise, the portion ventilated outdoors and the portion released to local outdoor environments after down-the-drain disposal could be additional exposure sources to those who are not associated with that particular hand dishwashing activity. We also considered "exposure after product use' because people can be exposed via dermal uptake from wearing clothes after being washed with a product. Figure 2.5 provides a conceptual overview of an integrated model framework that describes how models are connected and how outputs from one model are used as an input for other models.

To provide the logistics of how this model framework can be used, we (1) apply cleaning product volatilization models to estimate f volatilized and f down -the-drain, (2) apply cleaning product exposure models to estimate iF during/after cleaning product use, (3) apply an indoor fate and exposure model to estimate the fraction ventilated outdoors (f ventilated) and iF due to indoor releases, (4) apply an outdoor fate and exposure model to estimate iF due to outdoor air and surface water releases, and (5) apply a wastewater treatment plant (WWTP) fate model to estimate the fraction volatilized from a WWTP (f volatilized , WWTP) and the fraction discharged to surface water as effluents (f discharged). The last step is to integrate results from indoor and outdoor fate modeling and WWTP fate modeling with those from cleaning product exposure and volatilization models to show overall impact of the three phases of exposures (i.e., exposure during/after product use, exposure due to indoor releases, exposure due to down-the-drain disposal) on the model results (i.e., iR).

Selected compounds

We selected a suite of organic compounds that are commonly used in three types of cleaning products (e.g., laundry detergents, dishwashing detergents, and surface cleaners) 91. Our selected compounds include four glycols, four glycol ethers, four volatile organic compounds (VOCs), and three other non- VOCs. A list of selected compounds is provided in Table 2.2 along with key chemical properties, which were obtained from the U.S. EPA Estimation Program Interface SuiteTM (EPI Suite) 92.

Models Used

To estimate f volatilized during the operation of residential dishwashers and clothes washers, we applied the volatilization models developed by the University of Texas Corsi research group 93-95. We made a rough estimate for volatilization during dishwashing based on the chemical-specific overall mass transfer coefficient (K v, m/hour) from a liquid phase to a gas phase. To estimate f volatilized during two surface cleaning activities, we used a regression model previously fitted to the results of experimental studies that measured emission rates from the use of surface cleaning products 96.

To estimate the fraction exposed through dermal uptake during hand dishwashing, surface cleaning, and wearing clothes washed with the product, the chemical-specific skin permeation coefficient, Kp (cm/hour), was derived from the ten Berge model (2009). We then applied dermal exposure parameters (e.g., area of skin contact, product use frequency, film thickness on the skin) aggregated from multiple sources 86 to estimate the total mass absorbed through the skin. The iF is calculated by dividing the total absorbed mass by the product amount per use. To estimate the fraction exposed through inhalation during hand dishwashing and surface cleaning, we multiplied f volatilized from the cleaning product volatilization models by the average inhalation rate (m3/hour) and the duration of exposure (hour), and then divided by the recommended volume (m3) of a breathing zone 97.

We previously developed an indoor fate and exposure model for organic compounds released to the indoor residential environment 98,99. For chemicals released to outdoor environments (i.e., ventilated outdoors and volatilized to outdoor air and discharged to surface water from a WWTP) as a result of cleaning activities, we applied a CalTOX multimedia fate and exposure model 100 to estimate the iF due to outdoor air releases and surface water releases. For chemicals disposed down the drain as a result of the cleaning activities, we used a WWTP fate model 101 to estimate the fraction of chemicals volatilized (f volatilized,WWTP) and discharged to surface water (f discharged) from a WWTP.

Model Integration

For each of the five cleaning activities (i.e., hand dishwashing, machine dishwashing, machine clothes washing, countertop surface cleaning, and floor surface cleaning), we estimated iF during/after product use, iF due to indoor releases, and iF due to down-the-drain disposal. For each of the three exposure phases, we considered all plausible exposure routes (e.g., inhalation, dermal uptake, etc.).

Results

Contribution of individual exposure phases to the total exposure

The model framework presented in this study integrates chemical exposures during various cleaning activities and thus allows us to understand the overall contributions of the three phases of exposures (i.e., exposure during/after product use, exposure due to indoor releases, and exposure due to down- the-drain disposal) to the total exposure. Figure 2.6 depicts the percent contribution of the three exposure phases to the total intake for five cleaning activities considered in this study. In the case of hand dishwashing, humans are primarily exposed during dishwashing activities for most of the compounds.

For machine dishwashing, the main exposure pathway is either due to indoor releases or due to down- the-drain disposal depending on the chemical properties. Specifically, for those chemicals with large vapor pressure (VP) and small K ow, the total exposure is driven by indoor releases during machine operation. The exposure contribution for machine clothes washing is similar to that for hand dishwashing.

Estimated intake rates by cleaning activities

Figure 2.7 shows integrated intake rates, iR (mg/kg/day), for each of the study compounds by five cleaning activities. Overall, the estimated daily iRs are higher for those compounds with large VP (four VOCs) during countertop surface cleaning and floor surface cleaning.

Table 2.2. Key chemical properties of selected compounds from the U.S. EPA EPI Suite TM

Figure 2.5. Scope of this study. Black boxes indicate either volatilization models or fate and exposure models and red thick and rounded boxes indicate exposure phases considered in this study.

Figure 2.6. Percent contribution of individual exposure phases to the total exposure for five cleaning activities. The exposure scenario for "iF after product use' used in this study is dermal uptake from wearing clothes washed with a laundry product.

Figure 2.7. Estimated intake rates (mg/kg/day) for each of the five cleaning activities from this study. Red dots are the estimated intake rates (mg/kg/day) from the Shin et al. (2015)76 study with assumptions that a maximum of 10% of the total product mass is available for dermal uptake and a product applied on the skin leaves until being washed like body lotion, regardless of type of cleaning activities.

Objective 2c: Estimating Exposures Based on Dust Concentrations

Humans are exposed in indoor residential settings to various semivolatile organic compounds (SVOCs) in insecticides, plasticizers, flame retardants, water- and oil-repellents, and personal care products (PCPs).10 SVOCs are an emerging public health concern based on toxicological findings in laboratory animals. To characterize risk for human exposures to these compounds in indoor environments, several models were developed to assess and prioritize a large number of indoor chemicals.5-8,102 These models require product use (e.g., carpet cleaner) and chemical release information (e.g., emission rate in unit mass per time) to simulate the fate and transport of chemicals released into the indoor environment.

However, little is known regarding SVOC source strength including quantity introduced (e.g., number of products and proportion of chemicals of interest in each), product use, and chemical release/transfer rates. Due to information gaps, assuming no release indoors underestimates exposure by several orders

of magnitude; conversely, assuming release of all product chemicals using existing models produces exposure estimates (i.e., external dose) several orders of magnitude larger than those inferred from biomarkers (i.e., internal dose),76 requiring alternative methods.

Among SVOCs with a known indoor presence but little information about their source strength, indoor dust concentrations can serve as a marker of exposure. Once chemicals are released indoors, chemicals are redistributed among gas-phase and particle-phase (i.e., airborne particles), settled dust, and other indoor surfaces; thus humans are exposed via inhalation, dermal uptake/skin absorption, or non-dietary dust ingestion. For this reason, several studies8,13,103,104 applied two partitioning relationships between gas-phase and settled indoor dust, as well as between gas-phase and airborne particles, to facilitate a rapid exposure calculation using measured dust concentrations. However, this method was not fully evaluated for a larger number of indoor SVOCs, because most previous studies compared exposures reconstructed from measured indoor dust concentrations and measured biomarkers with a limited number of phthalates. Also, variability and uncertainty of model input parameters were not fully taken into consideration in exposure calculations, limiting model validity.

We propose methods for improved estimates of SVOC exposure compared to currently available indoor exposure models. The study objectives are to employ our methods to estimate exposure to SVOCs known to have indoor sources using measured concentrations of indoor dust and biomarkers.

Materials and Methods

Scope and overview

We used daily intake rates (iR; μg/kg/day) as a primary exposure metric to validate our methods. To

account for exposure in a residential indoor environment, we assumed that exposure primarily occurs via three routes: inhalation (gas-phase and particle-phase); air-to-skin dermal uptake; and non-dietary dust ingestion.

To integrate exposures from the three routes, we first obtained measured dust concentrations (Cdust;

µg/g) of chemicals of interest having indoor sources. Second, we applied a dust-air partition coefficient (Kdg = Cdust/Cg) to estimate a gas-phase concentration (Cg; µg/m3) and then a particle-air partition coefficient (Kp = Cp/Cg) to estimate a particle-phase concentration (Cp; µg/m3). Third, we estimated iRs from the three routes using standard exposure factors (e.g., inhalation rate, dust ingestion rate).

iR calculation

Methods used to 1) estimate gas-phase and particle-phase concentrations from measured indoor dust levels and partitioning relationships and 2) estimate iRs via inhalation, dermal uptake, and non-dietary dust ingestion are described here.

Partitioning relationships : To estimate Cgand Cpamong the selected SVOCs, we applied partitioning relationships between settled dust and gas-phase, and between gas-phase and particle-phase. Briefly, the dust/air (i.e., dust/gas-phase) partition coefficient (Kdg; m3/mg), defined as the ratio of the mass associated with 1 mg of dust to the mass associated with 1 m3 of air, was derived to compute a predicted equilibrium mass fraction in air.11 The gas/particle partition coefficient (Kp; m3/µg), defined as the ratio of mass associated with 1 µg of particulate matter to the mass associated with 1 m3 of air, was derived to account for the distribution of chemical compounds between gas-phase and the surface of airborne particles.12 These relationships were previously used in other exposure assessment studies8,13,103,104 to estimate a concentration in one phase (e.g., gas phase) from that in the other (e.g., settled dust). With the chemical-specific estimates of Kdg and measured dust concentrations, we estimated Cgby dividing Cdust by Kdg.11 With the chemical-specific estimates of Kp and the estimated Cg, we estimated Cpby multiplying Cg, Kp, and TSP (total suspended particles; mass concentration of airborne particles, assumed to be 20 µg/m3).12

Exposure calculations : After estimating SVOC concentrations in gas- and particle-phases, we calculated iRs for each of the three exposure routes. For inhalation, we multiplied the chemical concentration in each phase (Cgor Cp; µg/m3) by inhalation rate (InhR; m3/day) and then divided by average male and female adult body weight (BW; 80 kg).105 For non-dietary dust ingestion, we multiplied the chemical concentration in dust (Cdust; µg/g) by dust ingestion rate (DIngR; g/day) and then divided by BW. To calculate exposure via dermal uptake from the gas-phase, we used the air-to-skin transdermal uptake theory.106 Briefly, air-to-skin transdermal uptake of SVOCs can be estimated from three resistances in series: 1) resistance from bulk air to air at the skin surface, 2) resistance through the stratum corneum, and 3) resistance through the viable epidermis.8,106 Thus, we multiplied Cgby overall permeability coefficient (Kp_g; m/day) by skin surface area (SA; m2) and then divided by BW. Exposure factors (e.g., InhR, DingR) were obtained from the U.S. Exposure Factors Handbook105.

Objective 3: Evaluate air-to-skin transdermal uptake models

  1. Measure concentrations of a broad spectrum of SVOCs in human skin . Concentrations of a broad spectrum of SVOCs will be measured in skin samples (obtained as surgical waste).
  2. Compare measured concentrations to model predictions to evaluate air-to-skin transdermal uptake models. Measured skin concentrations, compared to levels predicted in dermal models, are critical because air-to-skin transdermal uptake has not been evaluated with actual measurements and these models estimate dermal as a significant pathway.

We determined the primary objective could be met more effectively by taking multiple wipe samples from people's skin as opposed to using skin obtained as surgical waste. This is because we were concerned about the process from cleaning skin prior to surgery and how this would impact results.

Semi-volatile organic compounds (SVOCs) are ubiquitous in the indoor environment and a priority for exposure assessment because of the environmental health concerns that they pose. Direct air-to-skin dermal uptake has been shown to be comparable to the inhalation intake for compounds with certain chemical properties. In this study, we aim to further understand the transport of these types of chemicals through the skin, specifically through the stratum corneum (SC). Our assessment is based on collecting three sequential forehead skin wipes, each hypothesised to remove pollutants from successively deeper skin layers, and using these wipe analyses to determine the skin concentration profiles. The removal of SVOCs with repeated wipes reveals the concentration profiles with depth and provides a way to characterize penetration efficiency and potential transfer to blood circulation. We used a diffusion model applied to surface skin to simulate concentration profiles of SVOCs and compared them with the measured values. We found that two phthalates, dimethyl and diethyl phthalates, penetrate deeper into skin with similar exposure compared to other phthalates and targeted SVOCs--an observation supported by the model results as well. We also report the presence of statistically significant declining patterns with skin depth for most SVOCs, indicating that their diffusion through the SC is relevant and eventually can reach the blood vessels in the vascularized dermis. Finally, using a non-target approach, we identified skin oxidation products, linked to respiratory irritation symptoms, formed from the reaction between ozone and squalene.

Dermal exposure to indoor chemicals can occur either by direct contact with a surface or by direct transdermal uptake from air 107-109. This latter process occurs similarly to the process of partitioning between air and organic matter on indoor surfaces 107. After partitioning directly from air into the skin, SVOCs can permeate through the skin, and depending on their chemical properties, go deep enough to pass into the bloodstream. These models have been confirmed with experiments quantifying the direct dermal uptake of airborne SVOCs 109,110.

Another recently-studied phenomenon is the impact of occupants on indoor chemistry 111 . It has been reported that skin lipids react with indoor ozone to form oxidation by-products 112,113 that can alter the indoor chemistry. These substances can affect the oxidative capacity of indoor air, since the by-products react with different radicals and oxidants present (such as hydroxyl radicals), reducing their concentrations 111. As a result of these skin-lipid reactions, indoor ozone concentrations decrease, which can in turn impact the composition and concentration of secondary organic aerosols (SOA) 114, an important fraction of particulate matter indoors. Furthermore, the more volatile oxidation by-products, such as 4-oxo pentanal, can concentrate in the air and act as respiratory irritants 115. Other low-volatility products can accumulate in skin oils and act as skin irritants, with the potential of being absorbed into the bloodstream 116. By recognizing the reaction between squalene and ozone, we found that the identification of the oxidation products and the analysis of their concentration gradient as revealed by

consecutive skin wipe from our experiments offers an additional opportunity to study this emerging issue of indoor chemistry.

In this study, we use multiple consecutive forehead wipes to further study and better understand transport of SVOCs through the SC and the potential for their absorption into the bloodstream. We use a diffusion model 117 to describe the transport of selected SVOCs through the SC, the layer that offers the highest resistance to diffusion and transport through the skin. We then compare empirical data collected from individuals to the model predictions and estimate the depth of sampling associated with each wipe using the distribution of squalene concentrations measured in the participants. Lastly, we used a non-target approach to determine the presence of squalene oxidation products in the wipes to explore this new area of indoor air chemistry to determine if these compounds penetrate into the skin. To the best of our knowledge, this is the first study that utilizes quantitative information from multiple skin wipes to explore the fate of SVOCs in skin.

Methods Study design

We recruited 13 subjects for our wipe sampling study. The subjects were all adults 18 years of age or

older from Northern California. The group included 7 men and 6 women and represented a convenience sample. The same individual collected all samples following a standardized protocol to reduce experimental variability. The study was conducted under the University of California, Davis Institutional Review Board review and approval.

Sample collection and analysis

To study the passive transfer of chemicals to skin from air, we collected sequential skin wipe samples from the forehead, one after another with no time interval in between. The forehead is an area thought to have primarily passive exposure, although hand-to-head contact is a possible route influencing the forehead levels. We refer to the first wipe collected as FH-1, with subsequent wipes being designated as FH-2 and FH-3.

Before use gauze pads (MG Chemicals, Surrey, BC, Canada) were Soxhlet extracted in hexane:acetone (1:1 v/v) for 24 h in batches of 50 wipes. After sample collection, a mixture of different isotope- labelled internal standards was added, the wipes were extracted using a 5 mL, 3:1 mixture of hexane and acetone, and sonicated 118. After transferring the supernatant, wipes were extracted again by adding 5 mL of acetone and sonicating. The resulting extracts were combined, the sample was evaporated to a final volume of 1 mL, filtered, and split into two fractions, one for the GC-MS analysis, the other for the LC-MS analysis.

We analyzed the GC-MS fraction on a 7200B GC-Q/TOF (quadrupole time-of-flight, high-resolution mass spectrometer, Agilent Technologies Inc.) using a standard column in 80 min chromatographic runs 118.

With this analysis, we quantified 17 targeted SVOCs (Table 3.1). For the LC-MS analysis, we solvent transferred the samples to a methanol:water solution (1:1 v/v). We analyzed the LC-MS fraction on a 6530 LC-Q/TOF (Agilent Technologies Inc.) high-resolution mass spectrometer using a reversed phase column in a 24 min chromatographic run and electrospray ionization in positive and negative ionization modes 118. We also used the data obtained from both the LC and GC in a non-target screen for potential oxidation products 112,113.

Estimation of the depth of each skin wipe and depth patterns of the chemicals

The depth of sampling of each wipe is needed to evaluate the permeation profile of chemicals through the skin. To deduce the depth from which a chemical is removed from the skin by wiping, we compared measured squalene mass recovered in successive wipes to concentrations reported in the literature and then used squalene as a removal marker for the wipes to determine equivalent depth.

After we estimated the depth of sampling for each wipe, we studied the concentration patterns for each chemical with depth by calculating correlation coefficients between concentrations and estimated sampling depth, ρ, across all subjects, and estimated their significance with a Fisher transformation. We also normalized the FH-2 and FH-3 sample concentrations to the FH-1 sample concentrations. These two methods allow us to explore chemical penetration through the skin.

Model of transport of chemicals in the stratum corneum (SC).

There are two transport processes for a chemical to reach the bloodstream: first, partitioning from air to skin, and second, transport across the SC to the viable epidermis to reach blood vessels.

To derive further insight from the experimental work, we used the model developed by Weschler and Nazaroff 108,117 to estimate the time required for compounds to move from air to skin followed by the transport of chemical substances through the skin based on Fickian diffusion within the SC, specifically evaluating phthalates.

In this approach, we first calculate the equilibrium concentration ratio between the top 1 lllm of skin lipids and air, which is derived from the octanol-air partitioning coefficient with the modifications proposed by Weschler and Nazaroff 108,117. We also calculated the time to equilibrium for the chemicals to partition from air to surface skin lipids, τ. Calculating first the time for a thin top layer of skin to reach equilibrium with the air, and then separately modeling transport through the skin is a similar concept as the existence of a thin skin surface lipid layer as proposed by Morrison et al.119.

We next modeled the movement of chemical into skin using a dynamic mass-transfer model 108,117 that produces a concentration-depth profile within the SC. The SC offers the greatest resistance to movement of chemical through skin as compared to other skin layers. Because there is a lack of direct measurements of skin diffusion coefficients (Dsc) available for compounds with log Kow values in the range of most SVOCs, we use permeability coefficient estimation equations from the U.S. EPA dermal exposure guidelines 120. These equations use Kow and molecular weight values 121,122 to obtain estimated diffusion coefficients (Table 3.1).

To compare the relative diffusion through skin to measured results, we calculate the concentrations in the SC after 24 hours of exposure, and we take values at the experimental depth of each wipe. Then we normalize those concentrations to that of the skin surface and compare them to the measured values.

Table 3.1. Chemical properties and parameters used in the model for the compounds modeled.

Results and discussion

Stratum Corneum (SC) concentration distributions

The following compounds were present in more than 90% of the samples collected: octocrylene, homosalate, galaxolide, di-methyl phthalate, di-ethyl phthalate, di n-butyl phthalate, bis-2-ethyl hexyl phthalate, di-octyl terephthalate and squalene, as indicated in Table 3.2. For the compounds measured in more than 50% of the samples, we include the median and mean concentrations on a per area basis.

In comparing the magnitude of the standard deviations to the mean values listed in Table 3.2, we see a high variability in the concentrations across the subjects for many of the compounds. A high variability is apparent for most of the phthalates, including low molecular weight phthalates used in personal care products, and high molecular weight phthalates used as plasticizers and in building materials. This variability is likely explained by differences in concentrations to which people are exposed, as well as differences in use patterns of consumer products that contain these chemicals of interest. Squalene variability is quite significant among all subjects as well.

Table 3.2. Percent of detection (%), average, median and standard deviation (in μg/m2) of area-based concentrations of chemicals measured in FH-1.

Analysis of squalene concentrations and estimation of depth of sampling.

With all squalene concentrations from every subject and assuming a constant concentration with depth, we calculated the log-normal mean depth of wipe sample extraction. We estimate that the first wipe removed chemicals from the first 0.6 µm of the SC (SD: 0.3-1.1 range), the second wipe removed chemicals from 0.6 to 0.9 µm, an additional depth of 0.3 µm (SD: 0.2-0.5), and the third wipe removed chemicals from 0.9 to 1.1 µm, an additional 0.2 µm depth (SD: 0.1-0.4). The use of squalene to estimate the depth of sampling has limitations. There is both variability in squalene levels between subjects and also by body location 123. As a result, the assumption of a constant squalene concentration to estimate the depth of sampling may produce depths that do not reflect the true sampling depth given that each subject has different levels. Moreover, a high inter-subject variability in the depth of sampling due to the use of squalene as a marker may be expected; this variability may be compounded by the small sample size of the present study.

Analysis of the depth patterns of the chemicals

To further understand transport processes through skin, we investigated the concentration patterns of both SVOCs and squalene with each subsequent wipe. Because sequential wipes extract chemicals deeper in the SC, we determined the correlation between the estimated depth and concentration levels for every subject. Results are shown in Table 3.3. We also note the low values of the SD relative to the average correlation coefficients. The first column is the correlation coefficient of the concentration with depth for each SVOC, and the results are all negative, indicating that the compounds diffuse through the skin towards the dermis with a decreasing concentration pattern with depth.

The two compounds with the largest correlation coefficients out of the compounds assessed are di- methyl and di-ethyl phthalate, indicating that they penetrate more deeply into the skin over shorter time periods and are found in higher concentrations at depth compared to the other SVOCs. There is a strong negative correlation for homosalate, octocrylene and galaxolide, indicating a steep decline with depth. The same holds true for bis-(2-ehtylhexyl) phthalate and di-octyl phthalate.

For squalene, there is a high variability in the observed levels among subjects, given its inherent biological variability. The negative concentration gradient of squalene reflects that squalene is a major constituent of sebum but not keratinocytes, and by moving deeper into the SC the ratio of sebum-to- keratinocytes-related substances decreases 124, therefore the squalene concentrations should be decreasing. Although we assumed a constant squalene concentration in calculating the depths, there is not a perfect correlation ρ between squalene and depth because ρ represents the average correlation across all subjects between individual squalene levels and a common estimated depth, which is in turn the geometric mean across all subjects. Sapienic acid (cis-6-hexadecenoic acid), which is a unique human lipid, makes up about 5.6% 123 of the total human skin lipids. It shows a strong negative gradient with depth across all 13 subjects (Table 3.3).

Table 3.3 also lists the average percent and the range of chemical load in the uppermost wipe, calculated by summing the total mass of chemicals removed from FH-1, FH-2 and FH-3 and dividing the recovered mass in FH-1 by the total mass. FH-1 removes between 40-65% of chemical, averaging 50%. For the compounds that exhibit a significant concentration gradient with depth, the percentage removed by FH-1 should be greater than for other compounds, since these compounds do not diffuse as much through the SC, with a low percentage of removal by subsequent forehead wipes. We note that there is a correlation between the percentage of chemical load in FH-1, the variation with depth, ρ, and the chemical properties of the compound, as represented by DSC. For instance, the two compounds with the largest diffusion coefficients (DSC), di-methyl and di-ethyl phthalates, have concentration gradients that are not statistically significant with depth, reflecting more even distribution throughout the SC, and as such, are removed to a greater extent by subsequent forehead wipes and making the chemical load in FH-1 lower.

Table 3.3. Correlation coefficients, ρ, of the chemical concentrations with depth of each forehead wipe and p-values associated through a Fisher's transformation. The percentage of chemical load (calculated by dividing the mass determined in FH-1 by the total mass of chemical measured in all three wipes) removed by FH-1 is shown as well.

Model Results

The model has two components, transfer from the air to skin and diffusion through the skin. For 9 of the 14 compounds, the surface of the skin can reach 85% of the equilibrium concentration in less than one hour, much less than the assumed 24-hour exposure period, and thus for these compounds the air to skin transfer rate does not control the resulting profile through the skin. Additionally, some compounds we measured may have been applied directly to skin, bypassing the partitioning step. Therefore, we use the model to calculate the concentration profile in the skin after 24 hours of exposure to a constant air concentration, calculated based on the measured skin's concentration in FH-1 and the Klg of each compound. For one compound, DEHP, the time for the skin surface to reach equilibrium with air is 2.8 hours, and thus the assumption of constant air concentration on the skin surface will produce some error. For homosalate, galaxolide and di-n-octyl terephthalate, the time for the skin surface to reach equilibrium with the air exceeds the 24-hour model simulation period, a limitation that should be noted when interpretation the results. The assumption of a 24-hour modeling period is still used for these compounds, as it is assumed that the skin surface is cleaned every 24 hours.

Figure 3.1 provides the resulting concentration profiles where the thicker line represents the model prediction after 24 hours of a constant skin surface concentration and the other lines represent measured concentrations for each individual subject. We report the modeled normalized concentrations in the skin as a percentage of the concentration calculated in the uppermost wipe (FH-1) and compare the modeled estimates to the normalized measured concentrations in the wipes in Table

    1. The compounds are listed in order of descending calculated diffusion coefficient, DSC , and we see that the modeled concentrations are driven primarily by this property and thus the normalized modeled concentrations in FH-2 are in an almost perfect descending order. For example, the two compounds with the highest DSC, present the highest surface-normalized concentrations at depth, which indicates that they migrate through skin faster than others due to their chemical properties and are at greater concentrations in subsequent wipes.

Figure 3.1. Comparison of the results of the model (thicker line) to the measured concentrations in the wipes (thinner lines) through normalization of the concentrations, both measured and modeled, to the modeled and measured surface wipe FH-1 concentrations (%), with the calculated depth estimates (in μm).

When we observe the modeled normalized concentrations on the right-hand side of the table, we also find higher percentages in subsequent wipes (FH-2 and FH-3) for compounds with a higher DSC value. However, the values do not decrease in the exact order. For example, galaxolide shows lower than expected values in wipes 2 and 3, possibly because this for compound, it takes over 24 hours for the skin surface to equilibrate with the air. However, we also note that there is considerable variation in reported Kow values for this compound 107, which influences the DSC value. In contrast, homosalate, which also takes more than 24 hours for the skin surface to come to equilibrium with the air, has higher than expected normalized concentrations in FH-2, potentially because it is a UV product and may be applied directly to the skin surface. However, we also note that the two compounds with the highest DSC values, are also two of the most likely to have been applied directly to skin, given their predominance in consumer products.

Table 3.4. Summary of model results; to the left, the first three columns represent modeled normalized concentrations to the first wipe closest to the surface (FH-1) after simulation in the model for 24 hours of exposure; to the right, the three columns represent the percentage of measured concentrations normalized to the FH-1 for each depth at which the wipe removes pollutant. Compounds modeled are ordered by descending DSC value.

Non-target approach for oxidation products identification.

With non-target methods, we tentatively identified three primary oxidation products formed by squalene ozonolysis 112: 1-hydroxy-2-propanone (hydroxyacetone), 6-methyl-5-hepten-2-one (6-MHO), and 6,10-dimethyl-5,9-undecadien-2-one (geranyl acetone). We subsequently confirmed these compounds using authentic standards. We found geranyl acetone in most wipes, but it did not exhibit a significant declining concentration pattern with depth as expected. We found 6-MHO in 1-3 wipes in most individuals, most often including the surface wipe. We only detected hydroxyacetone in one person. These compounds tend to be more volatile, and thus may escape into the gaseous phase 112, bypassing detection in the wipes.

From the LC-Q/TOF non-target analysis, we tentatively identified two oxidation products: 4,9,13,17- tetramethyl-octadeca-4,8,12,16-tetraenoic acid (C-22 tetraenoic acid) and 5,9,13-trimethyl-tetradeca- 4,8,12-trienoic acid (C-17 trienoic acid). With no standards available, we did not make a full confirmation. However, the compounds were tentatively identified with fair confidence due to the following MS information: i) matching exact mass and isotope pattern of the precursor (confirmation of the molecular formula); ii) matching MS/MS fragments with predicted fragments from in-silico software tools (CFM-ID, MassFrontier). For C-17 trienoic acid, three MS/MS fragments were matched to the known spectrum, whereas for C-22 tetraenoic acid, only one MS/MS fragment was identified.

We used the area under the curve as a proxy for the relative concentration in the wipes. From these data we could assess the decrease in concentration of the oxidation products in each wipe sample. We also calculated the linear correlation coefficient between the area under the curve and penetration depth of each wipe (Table 3.5). For C-22 tetraenoic acid, the concentration with each subsequent wipe was decreasing, implying that the ozonolytic reaction occurs closer to the skin surface and, as the next wipe extracts chemicals and skin lipids deeper from the skin, this relatively high Koa compound partitions farther into skin. For C-17 trienoic acid, with a slightly lower Koa, the trend did not reach statistical significance.

We tentatively identified two other oxidation products through the NIST Library in GC-Q/TOF (with no standards available for confirmation): C-17 trienal found on wipes from 4 people and C-22 tetraenal found on wipes from 5 people as summarized in Table 3.5.

Table 3.5. Squalene oxidation products screened in the non-target analysis of the forehead wipes for both LC- and GC-Q/TOF, along with their estimated log Koa through EPI Suite™ by using their SMILES fragments.

In conclusion, squalene measurements offer key insights for interpreting dermal wipe samples while measurements of its oxidation products provide valuable information about the production and fate of these respiratory and skin irritant products in the dermal layer. Further work is needed to identify these compounds reliably, subject to the constraint that there are not readily available standards, to better characterize their presence and relative levels, both in the skin and in the indoor environment. Further non-target analyses should be performed to identify related compounds in the wipes, as well as SVOCs and other anthropogenic oxidation products, and propose pathways of oxidation other than ozone, such as microbes in the skin, or other processes that may alter and transform their original chemical structures.

References:

  1. Mitchell, C. S., Zhang, J. F. J., Sigsgaard, T., Jantunen, M., Lioy, P. J., Samson, R. & Karol, M. H. Current state of the science: Health effects and indoor environmental quality. Environmental Health Perspectives 15, 958‐964, doi:0.1289/ehp.8987 (2007).
  2. Klepeis , N. E., Nelson, W. C., Ott, W. R., Robinson, J. P., Tsang, A. M., Switzer, P., Behar, J. V., Hern, S. C. & Engelmann, W. H. The National Human Activity Pattern Survey (NHAPS): a resource for assessing exposure to environmental pollutants. Journal of Exposure Analysis and Environmental Epidemiology 11, 31‐252, doi:10.1038/sj.jea.7500165 (001).
  3. Adgate , J. L., Church, T. R., Ryan, A. D., Ramachandran, G., Fredrickson, A. L., Stock, T. H., Morandi, M. T. & Sexton, K. Outdoor, indoor, and personal exposure to VOCs in children. Environmental Health Perspectives 112, 186‐1392, doi:10.1289/ehp.7107 (2004).
  4. Esplugues , A., Ballester, F., Estarlich, M., Llop, S., Fuentes‐Leonarte, V., Mantilla, E. & Iniguez, C. Indoor and outdoor air concentrations of BTEX and determinants in a cohort of one‐year old children in Valencia, Spain. Science of the Total Environment 09, 63‐69, doi:10.1016/j.scitotenv.2010.09.039 (2010).
  5. Shin, H. M., McKone, T. E. & Bennett, D. H. Intake Fraction for the Indoor Environment: A Tool for Prioritizing Indoor Chemical Sources. Environmental Science & Technology 46, 10063‐10072, doi:10.1021/es3018286 (2012).
  6. Zhang, X., Arnot, J. A. & Wania, F. Model for screening‐level assessment of near‐field human exposure to neutral organic chemicals released indoors. Environmental Science & Technology 48, 12312‐12319, doi:10.1021/es502718k (2014).
  7. Wenger, Y., Li, D. & Jolliet, O. Indoor intake fraction considering surface sorption of air organic compounds for life cycle assessment. International Journal of Life Cycle Assessment 1, 919‐931, doi:10.100/s11367‐012‐0420‐0 (2012).
  8. Little, J. C., Weschler, C. J., Nazaroff, W. W., Liu, Z. & Hubal, E. A. C. Rapid methods to estimate potential exposure to semivolatile organic compounds in the indoor environment. Environmental Science & Technology 46, 11171‐1117, doi:10.1021/es301088a (2012).
  9. Isaacs, K. K., Glen, W. G., Egeghy, P., Goldsmith, M. R., Smith, L., Vallero, D., Brooks, R., Grulke,

C. M. & Ozkaynak, H. SHEDS‐HT: An Integrated Probabilistic Exposure Model for Prioritizing Exposures to Chemicals with Near‐Field and Dietary Sources. Environmental Science & Technology 48, 12750‐12759, doi:10.1021/es502513w (2014).

  1. Weschler, C. J. & Nazaroff, W. W. Semivolatile organic compounds in indoor environments.

Atmospheric Environment 42 , 9018‐9040, doi:10.1016/j.atmosenv.2008.09.052 (2008).

  1. Weschler, C. J. & Nazaroff, W. W. SVOC partitioning between the gas phase and settled dust indoors. Atmospheric Environment 44, 3609‐3620, doi:10.1016/j.atmosenv.2010.06.029 (2010).
  2. Weschler, C. J., Salthammer, T. & Fromme, H. Partitioning of phthalates among the gas phase, airborne particles and settled dust in indoor environments. Atmospheric Environment 42, 1449‐ 1460, doi:10.1016/j.atmosenv.2007.11.014 (2008).
  3. Dodson, R. E., Camann, D. E., Morello‐Frosch, R., Brody, J. G. & Rudel, R. A. Semivolatile Organic Compounds in Homes: Strategies for Efficient and Systematic Exposure Measurement Based on Empirical and Theoretical Factors. Environmental Science & Technology 49, 1‐122, doi:10.1021/es502988r (2015).
  4. Meeker, J. D., Cooper, E. M., Stapleton, H. M. & Hauser, R. Urinary Metabolites of Organophosphate Flame Retardants: Temporal Variability and Correlations with House Dust Concentrations. Environmental Health Perspectives 121, 580‐585, doi:10.1289/ehp.1205907 (2013).
  5. Coakley, J. D., Harrad, S. J., Goosey, E., Ali, N., Dirtu, A. C., Van den Eede, N., Covaci, A., Douwes,

J. & t Mannetje, A. Concentrations of polybrominated diphenyl ethers in matched samples of indoor dust and breast milk in New Zealand. Environment International 59, 255‐261, doi:10.1016/j.envint.2013.06.020 (2013).

  1. Shin, M. Y., Lee, S., Kim, H. J., Lee, J. J., Choi, G., Choi, S., Kim, S., Kim, S. Y., Park, J., Moon, H. B., Choi, K. & Kim, S. Polybrominated Diphenyl Ethers in Maternal Serum, Breast Milk, Umbilical Cord Serum, and House Dust in a South Korean Birth Panel of Mother‐Neonate Pairs. International Journal of Environmental Research and Public Health 13, doi:10.3390/ijerph13080767 (20).
  2. Hammel, S. C., Hoffman, K., Lorenzo, A. M., Chen, A., Phillips, A. L., Butt, C. M., Sosa, J. A., Webster, T. F. & Stapleton, H. M. Associations between flame retardant applications in furniture foam, house dust levels, and residents' serum levels. Environment International 107, 181‐189, doi:10.1016/j.envint.20.07.015 (20).
  3. Hoffman, K., Garantziotis, S., Birnbaum, L. S. & Stapleton, H. M. Monitoring Indoor Exposure to Organophosphate Flame Retardants: Hand Wipes and House Dust. Environmental Health Perspectives 123, 160‐165, doi:10.1289/ehp.1408669 (2015).
  4. Watkins, D. J., McClean, M. D., Fraser, A. J., Weinberg, J., Stapleton, H. M. & Webster, T. F. Associations between PBDEs in office air, dust, and surface wipes. Environment International 59, 124‐132, doi:10.1016/j.envint.2013.06.001 (2013).
  5. Bamai , Y. A., Araki, A., Kawai, T., Tsuboi, T., Saito, I., Yoshioka, E., Cong, S. & Kishi, R. Exposure to phthalates in house dust and associated allergies in children aged 6‐12 years. Environment International 96, 16‐23, doi:10.1016/j.envint.16.08.025 (16).
  6. Johnson, P. I., Stapleton, H. M., Mukherjee, B., Hauser, R. & Meeker, J. D. Associations between brominated flame retardants in house dust and hormone levels in men. Science of the Total Environment 445, 177‐184, doi:10.1016/j.scitotenv.2012.12.017 (2013).
  7. Meeker, J. D., Johnson, P. I., Camann, D. & Hauser, R. Polybrominated diphenyl ether (PBDE) concentrations in house dust are related to hormone levels in men. Science of the Total Environment 407, 3425‐3429, doi:10.1016/j.scitotenv.2009.01.030 (2009).
  8. Philippat , C., Bennett, D. H., Krakowiak, P., Rose, M., Hwang, H. M. & Hertz‐Picciotto, I. Phthalate concentrations in house dust in relation to autism spectrum disorder and developmental delay in the CHildhood Autism Risks from Genetics and the Environment (CHARGE) study. Environmental Health 14, doi:10.1186/s12940‐015‐0024‐9 (2015).
  9. Ao , J. J., Yuan, T., Ma, Y. N., Gao, L., Ni, N. & Li, D. Identification, characteristics and human exposure assessments of triclosan, bisphenol‐A, and four commonly used organic UV filters in indoor dust collected from Shanghai, China. Chemosphere 184, 575‐583, doi:10.1016/j.chemosphere.2017.06.033 (2017).
  10. Hwang, H. M., Park, E. K., Young, T. M. & Hammock, B. D. Occurrence of endocrine‐disrupting chemicals in indoor dust. Science of the Total Environment 404, 26‐35, doi:10.1016/j.scitotenv.2008.05.031 (2008).
  11. Lankova , D., Svarcova, A., Kalachova, K., Lacina, O., Pulkrabova, J. & Hajslova, J. Multi‐analyte method for the analysis of various organohalogen compounds in house dust. Analytica Chimica Acta 854, 61‐69, doi:10.1016/j.aca.2014.11.007 (2015).
  12. Dodson, R. E., Perovich, L. J., Covaci, A., Van den Eede, N., Ionas, A. C., Dirtu, A. C., Brody, J. G. & Rudel, R. A. After the PBDE Phase‐Out: A Broad Suite of Flame Retardants in Repeat House Dust Samples from California. Environmental Science & Technology 46, 13056‐13066, doi:10.1021/es303879n (2012).
  13. Rudel , R. A., Camann, D. E., Spengler, J. D., Korn, L. R. & Brody, J. G. Phthalates, Alkylphenols, Pesticides, Polybrominated Diphenyl Ethers, and Other Endocrine‐Disrupting Compounds in Indoor Air and Dust. Environmental Science & Technology 37, 4543‐4553, doi:10.1021/es0264596 (2003).
  14. Krauss, M., Singer, H. & Hollender, J. LC-high resolution MS in environmental analysis: from target screening to the identification of unknowns. Analytical and Bioanalytical Chemistry 397, 943‐951, doi:10.1007/s00216‐010‐3608‐9 (2010).
  15. Moschet , C., Piazzoli, A., Singer, H. & Hollender, J. Alleviating the reference standard dilemma using a systematic exact mass suspect screening approach with liquid chromatography‐high resolution mass spectrometry. Anal Chem 85, 10312‐10320, doi:10.1021/ac4021598 (2013).
  16. Schymanski , E. L., Singer, H. P., Longree, P., Loos, M., Ruff, M., Stravs, M. A., Ripolles Vidal, C. & Hollender, J. Strategies to characterize polar organic contamination in wastewater: exploring the capability of high resolution mass spectrometry. Environ Sci Technol 48, 1811‐1818, doi:10.1021/es4044374 (2014).
  17. Gago‐Ferrero, P., Schymanski, E. L., Bletsou, A. A., Aalizadeh, R., Hollender, J. & Thomaidis, N. S. Extended Suspect and Non‐Target Strategies to Characterize Emerging Polar Organic Contaminants in Raw Wastewater with LC‐HRMS/MS. Environ Sci Technol 49, 12333‐12341, doi:10.1021/acs.est.5b03454 (2015).
  18. Gomez, M. J., Gomez‐Ramos, M. M., Malato, O., Mezcua, M. & Fernandez‐Alba, A. R. Rapid automated screening, identification and quantification of organic micro‐contaminants and their main transformation products in wastewater and river waters using liquid chromatography‐ quadrupole‐time‐of‐flight mass spectrometry with an accurate‐mass database. J Chromatogr A 1217, 7038‐7054, doi:10.1016/j.chroma.2010.08.070 (2010).
  19. Wang, M. & Helbling, D. E. A non‐target approach to identify disinfection byproducts of structurally similar sulfonamide antibiotics. Water Res 102, 241‐251, doi:10.1016/j.watres.2016.06.042 (2016).
  20. Hollender, J., Schymanski, E. L., Singer, H. P. & Ferguson, P. L. Nontarget Screening with High Resolution Mass Spectrometry in the Environment: Ready to Go? Environmental Science & Technology 51, 11505‐11512, doi:10.1021/acs.est.7b02184 (2017).
  21. Rager, J. E., Strynar, M. J., Liang, S., McMahen, R. L., Richard, A. M., Grulke, C. M., Wambaugh, J. F., Isaacs, K. K., Judson, R., Williams, A. J. & Sobus, J. R. Linking high resolution mass spectrometry data with exposure and toxicity forecasts to advance high‐throughput environmental monitoring. Environment International 88, 269‐280, doi:10.1016/j.envint.2015.12.008 (2016).
  22. Ouyang, X., Weiss, J. M., de Boer, J., Lamoree, M. H. & Leonards, P. E. G. Non‐target analysis of household dust and laundry dryer lint using comprehensive two‐dimensional liquid chromatography coupled with time‐of‐flight mass spectrometry. Chemosphere 166, 431‐4, doi: http ://dx.doi.org/10.1016/j.chemosphere.2016.09.107(2017).
  23. Ionas , A. C., Ballesteros Gómez, A., Leonards, P. E. G. & Covaci, A. Identification strategies for flame retardants employing time‐of‐flight mass spectrometric detectors along with spectral and spectra‐less databases. Journal of Mass Spectrometry 50, 1031‐10, doi:10.1002/jms.3618 (2015).
  24. Peng, H., Saunders, D. M. V., Sun, J., Jones, P. D., Wong, C. K. C., Liu, H. & Giesy, J. P. Mutagenic Azo Dyes, Rather Than Flame Retardants, Are the Predominant Brominated Compounds in House Dust. Environmental Science & Technology 50, 12669‐12677, doi:10.1021/acs.est.6b054 (2016).
  25. Megson , D., Robson, M., Jobst, K. J., Helm, P. A. & Reiner, E. J. Determination of Halogenated Flame Retardants Using Gas Chromatography with Atmospheric Pressure Chemical Ionization (APCI) and a High‐Resolution Quadrupole Time‐of‐Flight Mass Spectrometer (HRqTOFMS). Analytical Chemistry 88, 116‐11411, doi:10.1021/acs.analchem.6b01550 (2016).
  26. Hilton, D. C., Jones, R. S. & Sjodin, A. A method for rapid, non‐targeted screening for environmental contaminants in household dust. J Chromatogr A 1217, 6851‐6856, doi:10.1016/j.chroma.2010.08.039 (2010).
  27. Roberts, J. W., Budd, W. T., Ruby, M. G., Bond, A. E., Lewis, R. G., Wiener, R. W. & Camann, D. E. Development and field testing of a high volume sampler for pesticides and toxics in dust. Journal of exposure analysis and environmental epidemiology 1, 143‐155 (1991).
  28. Ma, W.‐L., Subedi, B. & Kannan, K. The Occurrence of Bisphenol A, Phthalates, Parabens and Other Environmental Phenolic Compounds in House Dust: A Review. Current Organic Chemistry 18, 2182‐2199 (2014).
  29. Mercier, F., Glorennec, P., Thomas, O. & Le Bot, B. Organic contamination of settled house dust, a review for exposure assessment purposes. Environ Sci Technol 45, 6716‐6727, doi:10.1021/es200925h (2011).
  30. Stapleton, H. M., Klosterhaus, S., Eagle, S., Fuh, J., Meeker, J. D., Blum, A. & Webster, T. F. Detection of Organophosphate Flame Retardants in Furniture Foam and U.S. House Dust. Environmental Science & Technology 43, 7490‐7495, doi:10.1021/es9014019 (2009).
  31. Papachlimitzou , A., Barber, J. L., Losada, S., Bersuder, P. & Law, R. J. A review of the analysis of novel brominated flame retardants. J Chromatogr A 1219, 15‐28, doi:10.1016/j.chroma.2011.11.029 (2012).
  32. Negreira , N., Rodríguez, I., Rubí, E. & Cela, R. Determination of selected UV filters in indoor dust by matrix solid‐phase dispersion and gas chromatography-tandem mass spectrometry. Journal of Chromatography A 1216, 5895‐5902, doi: http ://dx.doi.org/10.1016/j.chroma.2009.06.020 (2009).
  33. Weschler, C. J. Roles of the human occupant in indoor chemistry. Indoor Air 26, 6‐24, doi:10.1111/ina.12185 (2016).
  34. Goldsmith, M. R., Grulke, C. M., Brooks, R. D., Transue, T. R., Tan, Y. M., Frame, A., Egeghy, P. P., Edwards, R., Chang, D. T., Tornero‐Velez, R., Isaacs, K., Wang, A., Johnson, J., Holm, K., Reich, M., Mitchell, J., Vallero, D. A., Phillips, L., Phillips, M., Wambaugh, J. F., Judson, R. S., Buckley, T. J. & Dary, C. C. Development of a consumer product ingredient database for chemical exposure screening and prioritization. Food Chem Toxicol 65, 269‐279, doi:10.1016/j.fct.2013.12.029 (2014).
  35. Moschet , C., Lew, B. M., Hasenbein, S., Anumol, T. & Young, T. M. LC‐ and GC‐QTOF‐MS as Complementary Tools for a Comprehensive Micropollutant Analysis in Aquatic Systems. Environmental Science & Technology 51, 1553‐1561, doi:10.1021/acs.est.6b05352 (2017).
  36. NIST. NIST Standard Reference Database 1A, < http://www.nist.gov/srd/nist1a.cfm> (2016).
  37. Schymanski , E. L., Jeon, J., Gulde, R., Fenner, K., Ruff, M., Singer, H. P. & Hollender, J. Identifying small molecules via high resolution mass spectrometry: communicating confidence. Environ Sci Technol 48, 2097‐2098, doi:10.1021/es5002105 (2014).
  38. Ruttkies , C., Schymanski, E. L., Wolf, S., Hollender, J. & Neumann, S. MetFrag relaunched: incorporating strategies beyond in silico fragmentation. J Cheminform 8, 3, doi:10.1186/s13321‐ 016‐0115‐9 (2016).
  39. NORMAN Network. NORMAN Suspect List Exchange, < http://www.norman‐ network.com/?q=node/236> (2017).
  40. Allen, F., Greiner, R. & Wishart, D. Competitive fragmentation modeling of ESI‐MS/MS spectra for putative metabolite identification. Metabolomics 11, 98‐110, doi:10.1007/s11306‐014‐0676‐ 4 (2015).
  41. Hornung, R. W. & Reed, L. D. Estimation of Average Concentration in the Presence of Nondetectable Values. Applied Occupational and Environmental Hygiene 5, 46‐51, doi:10.1080/1047322X.1990.10389587 (1990).
  42. Shin, H. M., McKone, T. E., Nishioka, M. G., Fallin, M. D., Croen, L. A., Hertz‐Picciotto, I., Newschaffer, C. J. & Bennett, D. H. Determining source strength of semivolatile organic compounds using measured concentrations in indoor dust. Indoor Air 24, 260‐271, doi:10.1111/ina.12070 (2014).
  43. Mercier, F., Gilles, E., Saramito, G., Glorennec, P. & Le Bot, B. A multi‐residue method for the simultaneous analysis in indoor dust of several classes of semi‐volatile organic compounds by pressurized liquid extraction and gas chromatography/tandem mass spectrometry. Journal of Chromatography A 1336, 101‐111, doi: http ://dx.doi.org/10.1016/j.chroma.2014.02.004(2014).
  44. Canosa , P., Perez‐Palacios, D., Garrido‐Lopez, A., Tena, M. T., Rodriguez, I., Rubi, E. & Cela, R. Pressurized liquid extraction with in‐cell clean‐up followed by gas chromatography‐tandem mass spectrometry for the selective determination of parabens and triclosan in indoor dust. J Chromatogr A 1161, 105‐112, doi:10.1016/j.chroma.2007.05.089 (2007).
  45. Fraser, A. J., Webster, T. F., Watkins, D. J., Strynar, M. J., Kato, K., Calafat, A. M., Vieira, V. M. & McClean, M. D. Polyfluorinated compounds in dust from homes, offices, and vehicles as predictors of concentrations in office workers' serum. Environ Int , 128‐136, doi:10.1016/j.envint.2013.08.012 (2013).
  46. Liao, C., Liu, F., Guo, Y., Moon, H. B., Nakata, H., Wu, Q. & Kannan, K. Occurrence of eight bisphenol analogues in indoor dust from the United States and several Asian countries: implications for human exposure. Environ Sci Technol 46, 9138‐9145, doi:10.1021/es302004w (2012).
  47. Alam, M. S., West, C. E., Scarlett, A. G., Rowland, S. J. & Harrison, R. M. Application of 2D‐GCMS reveals many industrial chemicals in airborne particulate matter. Atmospheric Environment 65, 101‐111, doi: https ://doi.org/10.1016/j.atmosenv.2012.10.014 (2013).
  48. Loos, M. & Singer, H. Nontargeted homologue series extraction from hyphenated high resolution mass spectrometry data. Journal of Cheminformatics 9, 12, doi:10.1186/s13321‐017‐ 0197‐z (2017).
  49. Ferguson, P. L., Iden, C. R. & Brownawell, B. J. Analysis of nonylphenol and nonylphenol ethoxylates in environmental samples by mixed‐mode high‐performance liquid chromatography-electrospray mass spectrometry. Journal of Chromatography A 938, 79‐91, doi: https ://doi.org/10.1016/S0021 ‐9673(01)01091‐3 (2001).
  50. Lara‐Martín, P. A., González‐Mazo, E. & Brownawell, B. J. Multi‐residue method for the analysis of synthetic surfactants and their degradation metabolites in aquatic systems by liquid chromatography-time‐of‐flight‐mass spectrometry. Journal of Chromatography A 1218, 4799‐ 4807, doi: http ://dx.doi.org/10.1016/j.chroma.2011.02.031(2011).
  51. Corada ‐Fernandez, C., Lara‐Martin, P. A., Candela, L. & Gonzalez‐Mazo, E. Tracking sewage derived contamination in riverine settings by analysis of synthetic surfactants. Journal of Environmental Monitoring 13, 2010‐2017, doi:10.1039/C1EM10150A (2011).
  52. Myers, A. L., Jobst, K. J., Mabury, S. A. & Reiner, E. J. Using mass defect plots as a discovery tool to identify novel fluoropolymer thermal decomposition products. Journal of Mass Spectrometry 49, 291‐296, doi:10.1002/jms.3340 (2014).
  53. De Silva, A. O., Allard, C. N., Spencer, C., Webster, G. M. & Shoeib, M. Phosphorus‐Containing Fluorinated Organics: Polyfluoroalkyl Phosphoric Acid Diesters (diPAPs), Perfluorophosphonates (PFPAs), and Perfluorophosphinates (PFPIAs) in Residential Indoor Dust. Environmental Science & Technology 46, 12575‐12582, doi:10.1021/es303172p (2012).
  54. Portolés , T., Mol, J. G. J., Sancho, J. V. & Hernández, F. Use of electron ionization and atmospheric pressure chemical ionization in gas chromatography coupled to time‐of‐flight mass spectrometry for screening and identification of organic pollutants in waters. Journal of Chromatography A 1339, 145‐153, doi: http ://dx.doi.org/10.1016/j.chroma.2014.03.001(2014).
  55. Schymanski , E. L., Singer, H. P., Slobodnik, J., Ipolyi, I. M., Oswald, P., Krauss, M., Schulze, T., Haglund, P., Letzel, T., Grosse, S., Thomaidis, N. S., Bletsou, A., Zwiener, C., Ibanez, M., Portoles, T., de Boer, R., Reid, M. J., Onghena, M., Kunkel, U., Schulz, W., Guillon, A., Noyon, N., Leroy, G., Bados, P., Bogialli, S., Stipanicev, D., Rostkowski, P. & Hollender, J. Non‐target screening with high‐resolution mass spectrometry: critical review using a collaborative trial on water analysis. Anal Bioanal Chem 407, 6237‐6255, doi:10.1007/s00216‐015‐8681‐7 (2015).
  56. Moschet , C., Anumol, T., Lew, B. M., Bennett, D. H. & Young, T. M. Household dust as a repository of chemical accumulation: New insights from a comprehensive high‐resolution mass spectrometry study. Environmental Science & Technology In press (2018).
  57. Guo, W. H., Park, J. S., Wang, Y. Z., Gardner, S., Baek, C., Petreas, M. & Hooper, K. High polybrominated diphenyl ether levels in California house cats: House dust a primary source? Environ Toxicol Chem 31, 301‐306, doi:10.1002/etc.1700 (2012).
  58. Shen, B., Whitehead, T. P., McNeel, S., Brown, F. R., Dhaliwal, J., Das, R., Israel, L., Park, J. S. & Petreas, M. High Levels of Polybrominated Diphenyl Ethers in Vacuum Cleaner Dust from California Fire Stations. Environmental Science & Technology 49, 4988‐4994, doi:10.1021/es505463g (2015).
  59. Weschler, C. J., Langer, S., Fischer, A., Beko, G., Toftum, J. & Clausen, G. Squalene and Cholesterol in Dust from Danish Homes and Daycare Centers. Environmental Science & Technology 45, 3872‐3879, doi:10.1021/es103894r (2011).
  60. Alfonso‐Garrido, J., Bennett, D. H., Parthasarathy, S., Moschet, C., Young, T. M. & McKone, T. E. Exposure Assessment For Air‐To‐Skin Uptake of Semivolatile Organic Compounds (SVOCs) Indoors. Environmental Science & Technology, doi:10.1021/acs.est.8b05123 (2018).
  61. Shin, H. M., Ernstoff, A., Arnot, J. A., Wetmore, B. A., Csiszar, S. A., Fantke, P., Zhang, X. M., McKone, T. E., Jolliet, O. & Bennett, D. H. Risk‐Based High‐Throughput Chemical Screening and Prioritization using Exposure Models and in Vitro Bioactivity Assays. Environmental Science & Technology 49, 60‐6771, doi:10.1021/acs.est.5b00498 (2015).
  62. Mitro , S. D., Dodson, R. E., Singla, V., Adarnkiewicz, G., Elmi, A. F., Tilly, M. K. & Zota, A. R. Consumer Product Chemicals in Indoor Dust: A Quantitative Meta‐analysis of US Studies. Environmental Science & Technology 50, 10661‐10672, doi:10.1021/acs.est.6b02023 (2016).
  63. Allen, J. G., McClean, M. D., Stapleton, H. M. & Webster, T. F. Critical factors in assessing exposure to PBDEs via house dust. Environ. Int. 34, 1085‐1091, doi:10.1016/j.envint.2008.03.006 (2008).
  64. Julien, R., Adamkiewicz, G., Levy, J. I., Bennett, D., Nishioka, M. & Spengler, J. D. Pesticide loadings of select organophosphate and pyrethroid pesticides in urban public housing. Journal of Exposure Science and Environmental Epidemiology 18, 167‐174, doi:10.1038/sj.jes.7500576 (2008).
  65. Vojta , P. J., Friedman, W., Marker, D. A., Clickner, R., Rogers, J. W., Viet, S. M., Muilenberg, M. L., Thorne, P. S., Arbes, S. J. & Zeldin, D. C. First national survey of lead and allergens in housing: Survey design and methods for the allergen and endotoxin components. Environ. Health Perspect. 110, 527‐532 (2002).
  66. Wu, N., Herrmann, T., Paepke, O., Tickner, J., Hale, R., Harvey, E., La Guardia, M., McClean, M. D. & Webster, T. F. Human exposure to PBDEs: Associations of PBDE body burdens with food consumption and house dust concentrations. Environ. Sci. Technol. 41, 1584‐1589, doi:10.1021/es0620282 (2007).
  67. Wei, W., Boumier, J., Wyart, G., Ramalho, O. & Mandin, C. Cleaning practices and cleaning products in nurseries and schools: to what extent can they impact indoor air quality? Indoor Air, doi:10.1111/ina.12236 (2015).
  68. Morawska , L., He, C., Johnson, G., Guo, H., Uhde, E. & Ayoko, G. Ultrafine Particles in Indoor Air of a School: Possible Role of Secondary Organic Aerosols. Environ. Sci. Technol. 43, 9103‐9109, doi:10.1021/es902471a (2009).
  69. Nazaroff , W. W. & Weschler, C. J. Cleaning products and air fresheners: exposure to primary and secondary air pollutants. Atmos. Environ. 38, 21‐2865, doi:10.1016/j.atmosenv.2004.02.040 (2004).
  70. Singer, B. C., Coleman, B. K., Destaillats, H., Hodgson, A. T., Lunden, M. M., Weschler, C. J. & Nazaroff, W. W. Indoor secondary pollutants from cleaning product and air freshener use in the presence of ozone. Atmos. Environ. 40, 6696‐6710, doi:10.1016/j.atmosenv.2006.06.005 (2006).
  71. ACI. Consumer Product Ingredient Safety. (American Cleaning Institute, Washington, DC, 2010).
  72. USEPA. Exposure and Fate Assessment Screening Tool (E‐FAST): Version 2.0, documentation manual. ( U.S. Environmental Protection Agency, Springfield, VA, 2007).
  73. USEPA. Consumer exposure model (CEM) DRAFT user guide. ( U.S. Environmental Progection Agency, Washington, DC, 2015).
  74. Shin, H.‐M., McKone, T. E. & Bennett, D. H. Volatilization of Low Vapor Pressure ‐ Volatile Organic Compounds (LVP‐VOCs) during Three Cleaning Products‐Associated Activities: Potential Contributions to Ozone Formation. Chemosphere 153, 130‐137, doi:10.1016/j.chemosphere.2016.02.131 (2016).
  75. ten Berge, W. A simple dermal absorption model: Derivation and application. Chemosphere 75, 1440‐1445, doi:10.1016/j.chemosphere.2009.02.043 (2009).
  76. USEPA. Chemical and Product Categories (CPCat) database. ( U.S. Environmental Protection Agency, Washington, DC, 2016).
  77. USEPA. Estimation Program Interface SuiteTM for Microsoft Window, v 4.11. ( U.S. Environmental Protection Agency, Washington, DC, USA, 2016).
  78. Howard, C. & Corsi, R. L. Volatilization of chemicals from drinking water to indoor air: The role of residential washing machines. J. Air Waste Manag. Assoc. 48, 907‐914 (1998).
  79. Howard‐Reed, C., Corsi, R. L. & Moya, J. Mass transfer of volatile organic compounds from drinking water to indoor air: The role of residential dishwashers. Environ. Sci. Technol. 33, 2266‐ 2272, doi:10.1021/es981354h (1999).
  80. Shepherd, J. L., Corsi, R. L. & Kemp, J. Chloroform in indoor air and wastewater: The role of residential washing machines. J. Air Waste Manag. Assoc. 46, 631‐642 (1996).
  81. Singer, B. C., Destaillats, H., Hodgson, A. T. & Nazaroff, W. W. Cleaning products and air fresheners: emissions and resulting concentrations of glycol ethers and terpenoids. Indoor Air 16, 179‐191, doi:10.1111/j.1600‐0668.2005.00414.x (2006).
  82. Battelle. Measurement and Characterization of Aerosols Generated from a Consumer Spray Product. (Battelle, Richland, WA, 1999).
  83. Bennett, D. H. & Furtaw, E. J. Fugacity‐based indoor residential pesticide fate model. Environ. Sci. Technol. 38, 2142‐2152, doi:10.1021/es034287m (2004).
  84. Shin, H.‐M., McKone, T. E. & Bennett, D. H. Intake Fraction for the Indoor Environment: A Tool for Prioritizing Indoor Chemical Sources. Environ. Sci. Technol. 46, 10063‐10072, doi:10.1021/es3018286 (2012).
  85. McKone , T. E. (Livermore, CA, 1993).
  86. Clark, B., Henry, J. G. & Mackay, D. Fugacity analysis and model of organic‐chemical fate in a sewate‐treatment plant. Environ. Sci. Technol. 29, 1488‐1494, doi:10.1021/es00006a009 (1995).
  87. Isaacs, K. K., Glen, W. G., Egeghy, P., Goldsmith, M.‐R., Smith, L., Vallero, D., Brooks, R., Grulke,

C. M. & Özkaynak, H. SHEDS‐HT: an integrated probabilistic exposure model for prioritizing exposures to chemicals with near‐field and dietary sources. Environmental Science & Technology 48, 12750−12759, doi:10.1021/es502513w (2014).

  1. Beko , G., Weschler, C. J., Langer, S., Callesen, M., Toftum, J. & Clausen, G. Children's Phthalate Intakes and Resultant Cumulative Exposures Estimated from Urine Compared with Estimates from Dust Ingestion, Inhalation and Dermal Absorption in Their Homes and Daycare Centers. Plos One 8, doi:ARTN e62442 10.1371/journal.pone.0062442 (2013).
  2. Koch, H. M., Drexler, H. & Angerer, J. An estimation of the daily intake of di(2‐ ethylhexyl)phthalate (DEHP) and other phthalates in the general population. International Journal of Hygiene and Environmental Health 206, 77‐83, doi:Doi 10.1078/1438‐4639‐00205 (2003).
  3. USEPA. Exposure Factors Handbook. (U.S. Environmental Protection Agency, Washington, DC, 2011).
  4. Weschler, C. J. & Nazaroff, W. W. SVOC exposure indoors: fresh look at dermal pathways. Indoor Air 22, 356‐377, doi:10.1111/j.1600‐0668.2012.00772.x (2012).
  5. Roberts, E. M., English, P. B., Grether, J. K., Windharn, G. C., Somberg, L. & Wolff, C. Maternal residence near agricultural pesticide applications and autism spectrum disorders among children in the California Central Valley. Environ. Health Perspect. 115, 1482‐1489, doi:10.1289/ehp.10168 (2007).
  6. Schmidt, R. J., Tancredi, D. J., Ozonoff, S., Hansen, R. L., Hartiala, J., Allayee, H., Schmidt, L. C., Tassone, F. & Hertz‐Picciotto, I. Maternal periconceptional folic acid intake and risk of autism spectrum disorders and developmental delay in the CHARGE (CHildhood Autism Risks from Genetics and Environment) case‐control study. Am. J. Clin. Nutr. 96, 80‐89, doi:10.3945/ajcn.110.004416 (2012).
  7. Harrington, R. A., Lee, L.‐C., Crum, R. M., Zimmerman, A. W. & Hertz‐Picciotto, I. Prenatal SSRI Use and Offspring With Autism Spectrum Disorder or Developmental Delay. Pediatrics 133, E1241‐E1248, doi:10.1542/peds.2013‐3406 (2014).
  8. Rosenberg, R. E., Law, J. K., Yenokyan, G., McGready, J., Kaufmann, W. E. & Law, P. A. Characteristics and Concordance of Autism Spectrum Disorders Among 277 Twin Pairs. Arch. Pediatr. Adolesc. Med. 163, 907‐914 (2009).
  9. Grandjean, P. & Landrigan, P. J. Developmental neurotoxicity of industrial chemicals. Lancet 368, 2167‐2178, doi:10.1016/S0140‐6736(06)69665‐7 (2006).
  10. Arndt, T. L., Stodgell, C. J. & Rodier, P. M. The teratology of autism. Int. J. Dev. Neurosci. 23, 189‐ 199, doi:10.1016/j.ijdevneu.2004.11.001 (2005).
  11. Daniels, J. L. Autism and the environment. Environ. Health Perspect. 114, A396 (2006).
  12. Miodovnik , A. Environmental Neurotoxicants and Developing Brain. Mount Sinai Journal of Medicine: A Journal of Translational and Personalized Medicine 78, 58‐77, doi:10.1002/msj.20237 (2011).
  13. Chess, S. Follow‐up Report on Autism in Congenital Rubella. J. Autism Child. Schizophr. 7, 69‐81, doi:10.1007/bf01531116 (1977).
  14. Rodier , P. M., Bryson, S. E. & Welch, J. P. Minor malformations and physical measurements in autism: Data from Nova Scotia. Teratology 55, 319‐325, doi:10.1002/(sici)1096‐ 9926(199705)55:5<319::aid‐tera4>3.0.co;2‐u (1997).
  15. Landrigan , P. J., Lambertini, L. & Birnbaum, L. S. A Research Strategy to Discover the Environmental Causes of Autism and Neurodevelopmental Disabilities. Environ. Health Perspect. 120, A258‐A260, doi:10.1289/ehp.1104285 (2012). 118 CDC. Vol. 63 1‐24 (MMWR, 2014).
  16. Sparrow, S. S., Balla, D. A. & Cicchetti, D. V. Vineland Adaptive Behavior Scales. (1984).
  17. Moore, S. J., Turnpenny, P., Quinn, A., Glover, S., Lloyd, D. J., Montgomery, T. & Dean, J. C. S. A clinical study of 57 children with fetal anticonvulsant syndromes. J. Med. Genet. 37, 489‐497, doi:10.1136/jmg.37.7.489 (2000).
  18. Andersen, S. L., Laurberg, P., Wu, C. S. & Olsen, J. Attention deficit hyperactivity disorder and autism spectrum disorder in children born to mothers with thyroid dysfunction: a Danish nationwide cohort study. BJOG‐an International Journal of Obstetrics and Gynaecology , 1365‐1374, doi:10.1111/1471‐0528.12681 (2014).
  19. Khan, A., Harney, J. W., Zavacki, A. M. & Sajdel‐Sulkowska, E. M. Disrupted brain thyroid hormone homeostasis and altered thyroid hormone‐dependent brain gene expression in autism spectrum disorders. J. Physiol. Pharmacol. 65, 257‐272 (2014).
  20. Auyeung, B., Baron‐Cohen, S., Ashwin, E., Knickmeyer, R., Taylor, K., Hackett, G. & Hines, M. Fetal Testosterone Predicts Sexually Differentiated Childhood Behavior in Girls and in Boys. Psychol. Sci. 20, 144‐148, doi:10.1111/j.1467‐9280.2009.02279.x (2009).
  21. Auyeung, B., Baron‐Cohen, S., Ashwin, E., Knickmeyer, R., Taylor, K. & Hackett, G. Fetal testosterone and autistic traits. Br. J. Psychol. 100, 1‐22, doi:10.1348/000712608X311731 (2009).


Journal Articles on this Report : 3 Displayed | Download in RIS Format

Other project views: All 11 publications 3 publications in selected types All 3 journal articles
Type Citation Project Document Sources
Journal Article Alfonso‐Garrido J, Bennett DH, Parthasarathy S, Moschet C, Young TM, McKone TE. Exposure Assessment For Air‐To‐Skin Uptake of Semivolatile Organic Compounds (SVOCs) Indoors. Environmental Science & Technology 2019;53(3):1608‐1616 R835641 (Final)
  • Abstract from PubMed
  • Full-text: Environmental Science & Technology - Full Text HTML and PDF Download
    Exit
  • Journal Article Moschet C, Anumol T, Lew BM, Bennett DH Young TM. Household dust as a repository of chemical accumulation: New insights from a comprehensive high‐resolution mass spectrometry study. Environmental Science & Technology 2018;52(5):2878‐2887. R835641 (Final)
  • Abstract from PubMed
  • Full-text: Environmental Science & Technology - Full Text and PDF Download
    Exit
  • Journal Article Shin H-M, McKone TE, Bennett DH. Model framework for integrating multiple exposure pathways to chemicals in household cleaning products. Indoor Air 2017;27(4):829-839. R835641 (2016)
    R835641 (2017)
    R835641 (Final)
  • Abstract from PubMed
  • Abstract: Wiley-Abstract
    Exit
  • Progress and Final Reports:

    Original Abstract
  • 2015 Progress Report
  • 2016 Progress Report
  • 2017 Progress Report