α hbcd

According to the OECD guidelines (OECD 2004) , two different concentration levels of (I) 5 ng/µL and (II) 10 ng/µL of each of α-HBCD, β-HBCD, γ-HBCD and TBBP-A (Wellington Laboratories Inc., ON, Canada) were prepared in acetone. Based on the exposed surface area, a net dose of 500 ng/cm 2 (~7.8 µM/cm 2 ) and 1000 ng/cm 2 (~15.6 µM/cm 2 ) was applied to each of the investigated skin tissues using an appropriate volume (100 µL) of dosing solutions I and II, respectively. The applied doses fall within the range of potential human exposure to the studied BFRs via contact with indoor dust (Abdallah, et al. 2008a) . Moreover, they allow for measurement of expected low percentages (up to 0.01%) of the applied dose in various compartments of the exposure model.
To study the possible effect of the dosing vehicle on the percutaneous penetration of the tested chemicals, target BFRs were dissolved in 3 different dosing vehicles of: (A) acetone, (B) 30% acetone in water, and (C) 20% Tween 80 (Sigma-Aldrich, UK) in water at a concentration of 5 ng/µL. Preparation of the higher dosing level (i.e. 10 ng/µL) was not possible due to limited solubility of target BFRs in vehicles (B) and (C).

Solvents used during the extraction and analysis were all of analytical grade; n-hexane, acetone, methanol and dichloromethane were purchased from Merck (Darmstadt, Germany). Indoor dust SRM 2585 was purchased from NIST (Gaithersburg, MD, USA). Empty polypropylene filtration SPE cartridges (3 mL) were obtained from Sigma-Aldrich (Gillingham, UK). Silica gel (40 µm pore size) was purchased from J.T.Baker (London, UK). Anhydrous sodium sulfate (Na2SO4) and concentrated sulfuric acid (H2SO4, 98%) were purchased from Merck (Darmstadt, Germany). Standards of individual HBCDs (α-HBCD, β-HBCD, γ-HBCD), labelled 13 C HBCDs (α-, β-, γ-) and d18 γ-HBCD were purchased from Wellington Laboratories (Guelph, ON, Canada).
Glass fibre filters (GFFs, 12.5cm diameter, 1 µm pore size) were purchased from Whatman (Maidstone, UK). Florisil (60-100 mesh) and Silica gel (60Å, 60-100 mesh)
were purchased from Sigma-Aldrich (Dorset, UK). Oxygen-free nitrogen gas was purchased from BOC Gases (Manchester, UK).
For the CE-PBET model, analytical grade inorganic salts were obtained from Fisher Scientific (Loughborough, U.K) and organic components were purchased from Sigma-Aldrich (Dorset, UK). All glassware was cleaned by soaking for at least 12 h in a phosphate-free alkali solution, rinsed with water followed by distilled water and dried at 100°C for at least 12 h.

The dosing solutions for this study were prepared to mimic real-life exposure of a 12.3 kg toddler to indoor dust assuming a high dust ingestion scenario (200 mg dust day -1 ). 21 We used standard NIST SRM 2585 dust with known indicative values for all target compounds (Table SI-1). 240 mg of NIST SRM 2585 dust were extracted using pressurized liquid extraction (Dionex ASE 350, Sunnyvale, CA, USA) according to a previously reported method. 22 The extract was concentrated under a gentle stream of nitrogent, followed by solvent exchange to 2 mL of DMSO using 500 µL of toluene as a "keeper" to minimize analyte loss (D1). Another dosing solution (D2) containing 100 times the concentrations of target analytes in D1 was prepared by appropriate dilution of α-HBCD, β-HBCD, γ-HBCD, TCEP, TCIPP, and TDCIPP reference standards (Wellington Laboratories Inc., ON, Canada) with DMSO. D2 was used to mimic episodic high dose exposure which can be several orders of magnitude higher than average exposure scenarios. 23