You are here:
Developing Chemical Signatures for Categories of Household Consumer Products Using Suspect Screening Analysis
Citation:
Stanfield, Z., K. Favela, A. Yau, K. Phillips, A. Williams, K. Isaacs, AND J. Wambaugh. Developing Chemical Signatures for Categories of Household Consumer Products Using Suspect Screening Analysis. SOT, Salt Lake City, UT, March 10 - 14, 2024. https://doi.org/10.23645/epacomptox.25408714
Impact/Purpose:
N/A
Description:
Background and Purpose: Consumer products are a major source of chemical exposure and therefore potential risk. It is important to understand what chemicals are typically present in various types of products for risk evaluation and assessment of new products for uncommon chemical ingredients. The aim for this work was to identify basic “recipes” of five types of consumer products from non-targeted analysis (NTA) data. Methods: NTA using two-dimensional gas chromatography time-of-flight mass spectrometry (GCxGC-TOF/MS) was applied to samples from 160 selected consumer products across 5 categories (cotton clothing, fabric upholstery, shampoo, baby soap, and silicone kitchen utensils). Samples from products were extracted with dichloromethane (DCM). After addition of an internal standard, each extraction was analyzed via GCxGC-TOF/MS to obtain its mass spectra. The spectra were matched to the 2017 NIST database, and analytical standards were used to confirm a subset of the chemical identifications. Chemicals were annotated by structural classification via ClassyFire and reported or predicted functional uses. Sample replicates (2 distinct products of the same type) and duplicates (2 samples from the same individual product) were used to assess the variability in product composition at multiple levels of categorization. Lastly, chemical ingredient signatures were identified by selecting chemicals that occurred in ≥ 80% of samples in a product category. Signature chemicals were characterized by their observed sample concentration ranges, occurrence in a database of manufacturer-reported product ingredient information (EPA’s chemicals and product database, CPDat), and bioactivity in the form of predicted oral rat LD50 values. Results: 489 unique chemicals were tentatively identified (109 confirmed) across all products. The cotton clothing category contained the most unique chemicals (118) identifiable by GCxGC-TOF/MS while the silicone kitchen utensils category contained the least (30). A large majority (>85% on average per product category) of the identified chemicals were successfully assigned a structural classification and functional use, providing insights into the characteristics of these chemicals and their roles in products. Shampoo products showed the highest variability across samples in terms of chemical occurrence while cotton clothing products were the most uniform. Sample replicates and duplicates in all product categories clustered very distinctly, both in terms of chemical occurrence and estimated sample concentrations of constituent chemicals. Variability in sample replicates (a hint at consistency in the manufacturing process) was, on average, about twice as high as the variability in sample duplicates (consistency of the analytical method). Chemical signatures were obtained for each product category, with cotton clothing having the largest (29 chemicals) and baby soap having the smallest (5). A comparison of these signatures to a previous NTA pilot study (100 samples, 20 product categories with 5 samples each) showed good reproducibility, particularly for cotton clothing and fabric upholstery products, indicating the consistency and possible utility of these chemical signatures in assessing other products. Conclusions: This study provides a baseline set of chemical ingredients (that is, representative mixtures) across common types of consumer products, which will help in evaluating new and existing products. Separating constituent chemicals into typical and atypical might inform exposure assessment, in vitro bioactivity screening, and ultimately the risk related to using such products. This study is limited by the use of a single preparation/extraction technique and chemical analysis method (GC) such that other chemicals present may not have been detected. However, the analysis pipeline was designed to accommodate analysis of GCxGC-TOF/MS data from any set of products or substances (particularly substances...
URLs/Downloads:
DOI: Developing Chemical Signatures for Categories of Household Consumer Products Using Suspect Screening Analysis
POSTER.PDF (PDF, NA pp, 1734.876 KB, about PDF)