Air Partition Coefficients Research Articles

Data integration and knowledge transfer: application to the tissue: air partition coefficients

Data integration and knowledge transfer: application to the tissue: air partition coefficients

Bioaccumulation potential of air contaminants: Combining biological allometry, chemical equilibrium and mass-balances to predict accumulation of air pollutants in various mammals

In the present study we develop and test a uniform model intended for single compartment analysis in the context of human and environmental risk assessment of airborne contaminants. The new aspects of the model are the integration of biological allometry with fugacity-based mass-balance theory to describe exchange of contaminants with air. The developed model is applicable to various mammalian species and a range of chemicals, while requiring few and typically well-known input parameters, such as the adult mass and composition of the species, and the octanol–water and air–water partition coefficient of the chemical. Accumulation of organic chemicals is typically considered to be a function of the chemical affinity for lipid components in tissues. Here, we use a generic description of chemical affinity for neutral and polar lipids and proteins to estimate blood–air partition coefficients ( K ba) and tissue–air partition coefficients ( K ta) for various mammals. This provides a more accurate prediction of blood–air partition coefficients, as proteins make up a large fraction of total blood components. The results show that 68% of the modeled inhalation and exhalation rate constants are within a factor of 2.1 from independent empirical values for humans, rats and mice, and 87% of the predicted blood–air partition coefficients are within a factor of 5 from empirical data. At steady-state, the bioaccumulation potential of air pollutants is shown to be mainly a function of the tissue–air partition coefficient and the biotransformation capacity of the species and depends weakly on the ventilation rate and the cardiac output of mammals.

Comparison of the thermogravimetric analysis (TGA) and Franz cell methods to assess dermal diffusion of volatile chemicals

Recent work in our laboratory shows that it is possible to measure the minute weight changes of skin during exposure to chemical vapor using the thermogravimetric analysis (TGA) method. Further, the TGA weight curves can be analyzed with a two-compartment diffusion model, yielding two sets of diffusion and skin:air partition coefficients. In this study, the dermal diffusion of six chemical vapors ( n-butanol, cyclohexanone, ethyl benzene, n-hexane, styrene and m-xylene) were studied by the TGA method using neonatal pig skin. The penetration of the six substances was also studied using conventional Franz diffusion cells and neat, liquid chemical. A combined analysis of the data, along with previously published data on four additional substances ( n-butyl acetate, methanol, 2-propanol and toluene) suggest a good agreement between the TGA and the Franz cell methods with respect to diffusion coefficients. An interesting observation is that correction of the TGA-derived diffusion coefficients by the octanol:water partition coefficient improves the agreement with the Franz cell-derived coefficients. TGA-data for seven of the substances were further compared with dermal diffusion studies on chemical vapor reported by other investigators. The comparisons show good agreement with respect to permeability as well as skin:air partition coefficients. In conclusion, it appears that good estimates of diffusion coefficients as well as skin:air and permeability coefficients can be derived for volatile chemicals with the TGA technique.

A study of gas/particle partitioning of SVOCs in the tropical atmosphere of Southeast Asia

Gas- and particle-phase polycyclic aromatic hydrocarbons (PAHs) and polychlorinated biphenyls (PCBs) were collected at a tropical site in Southeast Asia over 12-h periods during November and December 2006 to determine their gas/particle distributions by analyzing integrated quartz filter and polyurethane foam samples. Gas/particle partitioning coefficients, K p, were calculated, and their relationship with the subcooled liquid vapor pressure p L o for both PAHs and PCBs was investigated. The regressions of log K p vs. log p L o for most of samples gave high correlations for both PAHs and PCBs and the slopes were statistically shallower than −1, but they were relatively steeper than those obtained in temperate zones of the Northern Hemisphere. By comparison, the particle-bound fraction of low molecular weight (LMW) PAHs was underestimated by both Junge-Pankow adsorption and K OA (octanol–air partition coefficient) absorption models, while the predicted values agree relatively better with those observed ones for high molecular weight (HMW) PAHs. In addition, the adsorption onto the soot phase (elemental carbon) predicted accurately the gas/particle partitioning of PAHs, especially for LMW compounds. On the other hand, the K OA absorption model using the measured organic matter fraction ( f OM) value fitted the PCB data much better than the adsorption model did, indicating the sorption of nonpolar compounds to aerosols might be dominated by absorption into organic matters in this area.

Quantitative Structure–Activity Relationship (QSAR) Modeling of Human Blood : Air Partitioning with Proper Statistical Methods and Validation

Blood: air partition coefficient (BApc) is important in assessing toxicokinetics of chemicals. Since very few experimental data are available, quantitative structure-activity relationship (QSAR) models based on calculated molecular descriptors can be useful in estimating BApc. Since all descriptors used in the analysis are computed strictly from structure, they can be applied to any chemical, real or hypothetical. In this article, we report models for BApc estimation using four methods, including stepwise ordinary least-squares regression, which is commonly used in QSAR studies but often results in an inflated 'naïve' q2, over-representing the predictive ability of the model. The models developed using proper statistical techniques had q2 values of 0.825 and 0.841, and may be used to reliably predict BApc values for new compounds that are structurally similar to those upon which the models are based. The models developed using improper techniques had associated q2 values, as computed using naïve methods, of 0.920 and 0.934, severely overstating their actual quality.

Octanol–air partition coefficients of polybrominated biphenyls

The octanol–air partition coefficients ( K OA) for PBB15, PBB26, PBB31, PBB49, PBB103 and PBB153 were determined as a function of temperature using a gas chromatographic retention time technique with 1,1,1-trichloro-2,2-bis (4-chlorophenyl) ethane (p,p′-DDT) as a reference substance. The internal energies of phase change from octanol to air ( Δ OA U) were calculated for the six compounds and were in the range from 74 to 116 kJ mol −1. Simple regression equations of log K OA versus relative retention times (RRTs) on gas chromatography (GC), and log K OA versus molecular connectivity indexes (MCI) were obtained, for which the correlation coefficients ( r 2) were greater than 0.985 at 283.15 K and 298.15 K. Thus the K OA values of the remaining PBBs can be predicted by using their RRTs and MCI according to these relationships.

Gas- and particle-phase distribution of polycyclic aromatic hydrocarbons in two-stroke, 50-cm 3 moped emissions

Gas- and particle-phase polycyclic aromatic hydrocarbons (PAHs) concentrations evaluated in the exhaust of 10 two-stroke, 50-cm 3 mopeds belonging to three different levels of emission legislation (EURO-0, EURO-1 and EURO-2) were used to assess the prevalent mechanism driving the gas/particle partitioning of PAHs in moped exhaust. Sampling was performed on a dynamometer bench both during the “cold-start” and the “hot” phases of the ECE-47 driving cycle. Gas and particulate phase PAHs were collected on polyurethane foam (PUF) plugs and 47-mm Pallflex T60A20 filters, respectively, under isokinetic conditions by using sampling probes inserted into the dilution tunnel of a Constant Volume Sampling – Critical Flow Venturi (CVS–CFV) system. The results show that semi-volatile PAHs were predominantly partitioned to the particle phase. The soluble organic fraction (SOF) of the collected particulates ranged between 72 and 98%. Measured total suspended particulate matter normalized partition coefficients ( K p) were predicted within a factor of 3–5 by assuming absorption into the organic fraction according to a model developed by Harner and Bidleman [Harner, T., Bidleman, T.F., 1998. Octanol–air partition coefficient for describing particle/gas partitioning of aromatic compounds in urban air. Environmental Science & Technology 32, 1494–1502.]. This suggests that the gas/particle partitioning in moped exhaust is mainly driven by the high fraction of organic matter of the emitted particles and that absorption could be the main partitioning mechanism of PAHs.

Semivolatile organic compounds in indoor environments

Semivolatile organic compounds (SVOCs) are ubiquitous in indoor environments, redistributing from their original sources to all indoor surfaces. Exposures resulting from their indoor presence contribute to detectable body burdens of diverse SVOCs, including pesticides, plasticizers, and flame retardants. This paper critically examines equilibrium partitioning of SVOCs among indoor compartments. It proceeds to evaluate kinetic constraints on sorptive partitioning to organic matter on fixed surfaces and airborne particles. Analyses indicate that equilibrium partitioning is achieved faster for particles than for typical indoor surfaces; indeed, for a strongly sorbing SVOC and a thick sorptive reservoir, equilibrium partitioning is never achieved. Mass-balance considerations are used to develop physical-science-based models that connect source- and sink-rates to airborne concentrations for commonly encountered situations, such as the application of a pesticide or the emission of a plasticizer or flame retardant from its host material. Calculations suggest that many SVOCs have long indoor persistence, even after the primary source is removed. If the only removal mechanism is ventilation, moderately sorbing compounds ( K oa > 10 10) may persist indoors for hundreds to thousands of hours, while strongly sorbing compounds ( K oa > 10 12) may persist for years. The paper concludes by applying the newly developed framework to explore exposure pathways of building occupants to indoor SVOCs. Accumulation of SVOCs as a consequence of direct air-to-human transport is shown to be potentially large, with a maximum indoor-air processing rate of 10–20 m 3/h for SVOC uptake by human skin, hair and clothing. Levels on human skin calculated with a simple model of direct air-to-skin transfer agree remarkably well with levels measured in dermal hand wipes for SVOCs possessing a wide range of octanol–air partition coefficients.

An integrated QSPR–PBPK modelling approach for in vitro–in vivo extrapolation of pharmacokinetics in rats†

In vitro data on metabolism and partitioning may be integrated within physiologically-based pharmacokinetic (PBPK) models to provide simulations of the kinetics and bioaccumulation of chemicals in intact organisms. Quantitative structure–property relationship (QSPR) modelling of available in vitro data may be performed to predict metabolism rates and partition coefficients (PCs) for developing in vivo PBPK models. The objective of the present study was to develop an integrated QSPR–PBPK modelling approach for the conduct of in vitro to in vivo extrapolation. For this purpose, data on rat blood:air (P b) and fat:air (P f) PCs, as well as intrinsic metabolic clearance (CLint) obtained using rat liver slices for some C5–C10 volatile organic compounds (VOCs) were compiled from the literature. Multilinear additive QSPR models for P f, P b and CLint were developed based on the number and nature of molecular fragments in these VOCs (CH3, CH2, CH, C, C=C, H, benzene ring and H in benzene ring structure). The mean estimated/experimental (est/exp) ratios (±SD; range) were 1.0 (±0.04; 0.93 − 1.06) for log P f, 1.08 (±0.26; 0.70 − 1.62) for log P b, and 1.07 (± 0.21; 0.80 − 1.44) for CLint. By accounting for the difference in the content of neutral lipids in fat and other tissues, the liver : air and muscle : air PCs of the compounds investigated in this study, with the excerption of n-decane, were adequately predicted from P f. Integrating the QSPRs for P f, P b and CLint within a rat PBPK model, simulations of inhalation pharmacokinetics of several VOCs were generated on the basis of molecular structure, for a given exposure scenario. The integrated QSPR–PBPK model developed in this study is a potentially useful tool for predicting in vivo kinetics and bioaccumulation of chemicals in rats under poor data situations. †Presented at the 13th International Workshop on QSARs in the Environmental Sciences (QSAR 2008), 8–12 June 2008, Syracuse, USA.

Prediction of the PCB pollution in the soils of Bursa, an industrial city in Turkey

The purpose of this study was to predict the equilibrium soil concentrations and distributions of Σ 41Polychlorinated Biphenyl (PCB) congeners from the air data measured at 4 sites (urban, suburban, residential, and industrial) in Bursa between 2004 and 2005. The soil–air partition coefficient ( K SA) was used for the predictions. Predicted soil concentrations ranged from 25 pg/g to 690 pg/g, while the atmospheric concentrations fluctuated between 35 pg/m 3 and 1112 pg/m 3. The soil concentrations were found to be in agreement with the lower range of European and global values, and below the regulatory limits. The urban and residential sites yielded higher PCB concentrations and a higher degree of heavier congeners than the sites characterized as suburban. The urban and residential sites were dominated by hexa-chlorinated biphenyls (CBs), followed by tri-CBs, while the industrial site was dominated by tri-CBs, followed by tetra-CBs. PCB congeners of 153, 180, 138, 118, and 101 were higher in all locations. The effect of the temperature changes on the concentrations was found to be insignificant, as the correlations ranged between 0.0009 and 0.22 for all sites. The K SA approach yielded reasonable results, and it is recommended to use this approach when there is limited data about the pollution levels of a certain media such as soil by using the measured levels of the other media such as air.

Some inferences on the mechanism of atmospheric gas/particle partitioning of polycyclic aromatic hydrocarbons (PAH) at Zaragoza (Spain)

Gas/particle partitioning of pollutants is an important mechanism determining atmospheric processing and its impact to environmental and human health. In this paper, the gas–particle partitioning of polycyclic aromatic hydrocarbons (PAH) has been studied with the aim of determining the main mechanism of PAH partitioning in Zaragoza (Spain) aerosols. To reach this goal, the ambient concentrations of PAH (gas and particle phase) collected in this city for one year period (2003–2004) have been analyzed. The partitioning between the particle and gas phases was studied according to three different models: the Junge adsorption model, the absorption into the organic matter model using the octanol–air ( K OA) partition coefficient and the absorption into the organic matter plus the adsorption onto the soot carbon model using the soot–air ( K SA) partition coefficients. Experimental gas/particle partition coefficients ( K P) correlated well with the subcooled liquid vapour pressures ( P L 0 ) of PAH but with slopes higher than the expected value of −1. Experimental K p values were well fit to the modelled ones when, in addition to absorption into organic matter, adsorption onto the soot carbon was considered. It could be concluded that the main partition mechanism in Zaragoza aerosols was explained by adsorption onto the soot carbon. However, K p modelled values were affected by the different thermodynamic parameters related to soot types. The influence of the organic matter and elemental carbon fractions on the K p modelling was also studied. The different particle characteristics, local factors, the presence of non-exchangeable fraction and non-equilibrium were considered like main keys to explain deviations of the experimental K p values from predictions according to models.

Field derived accumulation and release kinetics of DDTs in plants

Vegetation plays an important role in influencing the air/surface exchange of semivolatile organic compounds (SOCs). In order to predict the capability of different plant species to capture chemicals from the air, plant–air partition coefficients and kinetic accumulation parameters must be defined. In this study, potted plants of three different species were transferred to the vicinity of a source point for DDT, namely a contaminated area around a former production plant in Italy. Leaves were constantly sampled in order to follow the uptake from air over time. Later, the potted plants were transported to a location characterized by background diffuse air concentrations for the release phase. Coupling the experimental results with a two-compartment accumulation model it was possible to derive the kinetics parameters and the plant–air partition coefficient KPA for p,p′-DDT. The logKPA (on a mass/volume basis) ranged between 1.7 and 2.2 for the different species. The uncertainties related to the different phenomena involved in a field uptake/release experiment are discussed.

Novel Chamber to Measure Equilibrium Soil–Air Partitioning Coefficients of Low-Volatility Organic Chemicals under Conditions of Varying Temperature and Soil Moisture

The need to determine soil-air partitioning coefficients (K(SA)) of low-volatility organic chemicals as a measure of their distribution in the soil surface after release into the environment resulted in the development of a novel chamber system, which has been filed for patent. A major advantage of this pseudo-static system is that sufficient time can be factored into the experiment to ensure that the system has achieved equilibrium. In a highly precise method, the air is collected in adsorption tubes and subsequently liberated in a thermodesorption system for the quantitation of the adsorbed compound. The precision of the method is great enough that even the effects of temperature and soil moisture on the soil-air partitioning of very low-volatility compounds can be quantified. Because of analytical detection limits, quantitation of these influences has not been possible to date. Functionality of the setup was illustrated by measurements on the fungicide fenpropimorph. K(SA) values of fenpropimorph displayed a negative relationship with temperature and soil moisture. The type of application (spraying or incorporation) and the use of formulated compounds was shown to have a major impact on the measured K(SA) values. Comparison with calculations using an estimation method revealed that the use of experimentally determined K(SA) values will facilitate a more adequate consideration of volatilization in recent model approaches.

Are highly lipophilic volatile compounds expected to bioaccumulate with repeated exposures?

With non-volatile compounds, high lipophilicity (i.e., fat:blood partition coefficients, Pf, in the range of several hundred to a thousand or higher) typically leads to concerns for bioaccumulation. To evaluate the extent to which highly cleared, lipophilic vapors are expected to accumulate in blood and tissues, we conducted pharmacokinetic (PK) analysis, using both a generic physiologically based (PBPK) model for inhalation of volatile compounds (VCs) and a more detailed PBPK model specifically developed for a highly lipophilic volatile (decamethylcyclopentasiloxane, D 5). The generic PBPK model for inhalation of VCs in humans showed that highly metabolized, lipophilic compounds, with a low blood:air partition coefficient (Pb), do not accumulate in blood or systemic tissues with repeat exposures although a period of days to weeks may be required for fat to reach periodic steady state. VCs with higher Pb (in the hundreds) and lower hepatic extraction accumulate in blood on repeat exposures. The more detailed PBPK model for D 5 also showed that this lipophilc VC does not accumulate in blood and predictions of the increases in D 5 in fat with repeat exposures in rats agreed with experiments. In general, the major characteristic favoring accumulation of VCs in blood and systemic tissues is poor whole-body clearance, not lipophilicty. The term bioaccumulation should be used to refer to cases where repeat exposures lead to increases in VC blood (or central compartment) concentration. Based on this definition, highly cleared VCs, such as D 5, would not be considered to bioaccumulate on repeat exposures.

PAHs in Stockholm window films: Evaluation of the utility of window film content as indicator of PAHs in urban air

The thin organic film that builds up on the exterior surface of windows has been proposed as a ubiquitously available passive sampler for semi-volatile organic contaminants (SOCs) in urban air. Readily available school windows were sampled in Stockholm city centre and suburban locations in both winter and summer season to evaluate the putative usefulness of this matrix for assessing the integrated load of urban air pollution by polycyclic aromatic hydrocarbons (PAHs). The window-area normalised concentrations indicated more PAH contamination in the winter than in the summer in both the city centre and suburban locations, with highest concentrations in the city centre in the winter (∑PAH 43 451–467 ng m −2). However, normalising the PAH load to the amount of fatty window film, as measured by extractable organic matter (EOM), gave a more homogeneous picture with the EOM-normalised PAH load being inseparable both between summer and winter and between city centre and suburban locations. To evaluate the possibility of quantitatively employing urban window films as a means to provide predicted environmental concentrations of PAHs in air (PEC air), window film–air partition coefficients of PAHs were estimated using a set of coupled linear free energy relationships and physico-chemical properties of PAHs. Assuming dynamic equilibria between PAHs in air and dissolved in the window film, the obtained PEC air from the window films were consistently overestimating the urban vapour-phase PAH concentrations by factors 4–135. This discrepancy is quantitatively consistent with a strong and overwhelming association with black carbon aerosol particles accumulated in the window film. For SOCs that have a lower tendency to associate with black carbon, bulk window film concentrations may work better than for PAHs to estimate their vapour-phase concentrations in urban air.

Partitioning characteristics of PCBs in urban surface films

Urban areas are characterized by impervious surfaces on which a thin surface film layer forms. The film is an environmental compartment whereby semi-volatile organic compounds partition between the organic phase of the film and the ambient air. This study investigates the partitioning of polychlorinated biphenyls (PCBs) between surface films and ambient air. The film–air partition coefficient ( K fa) was calculated through laboratory experiments using ambient film samples, and then plotted against the log of the octanol–air partition coefficient ( K oa). The resultant log–log relationship is analogous to that between K oa and K p (gas–particle partition coefficient) where the latter has been measured for many compound classes, and thus, K fa can be used in models that consider the impact of the surface film on chemical dynamics. Ethylene vinyl acetate (EVA) was investigated as a surrogate film and EVA–air partition coefficients ( K EVA–air) values were measured. The relationship between log K fa and log K EVA–air is stronger than that between log K fa and log K oa, which leads to the preliminary conclusion that EVA may be a more suitable surrogate for ambient surface films than octanol. Film–air partition quotients ( Q fa) were derived based on field measurements of PCBs in the ambient air and concentrations in surface films. Q fa depends not only on the characteristics of the organic phase, but also on external factors, including but not limited to, temperature and ambient air concentrations. Finally, differences in PCB partitioning between film collected at a rural site versus an urban site highlights the importance of the composition of the organic bulk phase in determining chemical sorption into and subsequent volatilization out of the surface films.

Foliar limonene uptake scales positively with leaf lipid content: “non‐emitting” species absorb and release monoterpenes

Monoterpenes synthesized and released by emitting vegetation can be taken up by neighboring non-emitting plants, but the uptake capacity of non-emitting species has not been studied extensively. We investigated the foliar uptake potential of the hydrophobic monoterpene limonene in 13 species of contrasting leaf structure and lipid content to determine the structural and chemical controls of monoterpene uptake. Leaf dry mass per unit area (M(A,D)) varied 6.5-fold, dry to fresh mass ratio (D(F)) 2.7-fold, lipid content per dry mass (L(M)) 2.5-fold and per unit area (L(A)) 4.6-fold across the studied species. Average foliar limonene uptake rate (U(A)) from air at saturating limonene partial pressures varied from 0.9 to 6 nmol m(-2) s(-1), and limonene leaf to air partition coefficient (K(FA), ratio of limonene content per dry mass to limonene partial pressure) from 0.7 to 6.8 micromol kg(-1) Pa(-1). U(A) and K(FA) scaled positively with leaf lipid content, and were independent of D(F), indicating that variation in leaf lipid content was the primary determinant of species differences in monoterpene uptake rate and K(FA). Mass-based limonene uptake rates further suggested that thinner leaves with greater surface area per unit dry mass have higher uptake rates. In addition, limonene lipid to air partition coefficient (K(LA)=K(FA)/L(M)) varied 19-fold, indicating large differences in limonene uptake capacity at common leaf lipid content. We suggest that the significant uptake of hydrophobic monoterpenes when monoterpene ambient air concentration is high and release when the concentration is low should be included in large-scale monoterpene emission models.

QSPR analysis of air-to-blood distribution of volatile organic compounds

Quantitative structure property relationship (QSPR) models for the prediction of human blood:air partition coefficient (log K blood) of volatile organic compounds (VOCs) has been developed based on the linear heuristic method (HM) and non-linear radial basis function neural networks (RBFNNs). Molecular descriptors that are calculated from the structures alone were used to represent the characteristics of the compounds. HM was used both to pre-select the whole descriptor sets and to build the linear model. RBFNN was performed to obtain more accurate models. Both the linear and the non-linear models can give very satisfactory prediction results: the correlation coefficient R was 0.964 and 0.979, and the root-mean-square (RMS) error was 0.3303 and 0.2542 for the whole data set, respectively. The prediction result of the non-linear model is better than that obtained by the linear model. In addition, this paper provides an effective method for predicting log K blood from its structures and gives some insight into the structural features related to the solubility of VOCs in human blood.

Determination of the Octanol−Air Partition Coefficients (KOA) of Fluorotelomer Alcohols

Octanol−air partition coefficients (KOA) for five fluorotelomer alcohols (4:2 FTOH, 6:2 FTOH, 8:2 FTOH, 10:2 FTOH, and for the first time 12:2 FTOH) were determined over a temperature range from (5 to 40) °C using a modified generator column method. Octanol-saturated air at a fixed temperature was blown over a column filled with glass spheres which were coated with an FTOH solution. FTOHs partitioning to the gas phase were collected on a trap filled with XAD-2 and eluted with acetone afterward. The concentrations of FTOH were measured by GC-MS using positive chemical ionization. The KOA values were calculated as the ratio of the FTOH concentrations in the octanol to the FTOH concentrations in the air. Values of log KOA at 25 °C ranged from 4.5 (4:2 FTOH) to 6.2 (12:2 FTOH). The calculated internal energies of phase change (ΔUOA) ranged from 42.8 kJ·mol−1 (12:2 FTOH) to 61.1 kJ·mol−1 (4:2 FTOH). The combined uncertainty of the method applied ranged from 8.3 % (4:2 FTOH) to 10.5 % (10:2 FTOH). The expanded uncertainty (P = 95 %) ranged from 18.7 % (4:2 FTOH) to 38.7 % (10:2 FTOH), respectively.

Physiologically Based Modeling of the Inhalation Pharmacokinetics of Ethylbenzene in B6C3F1 Mice*

A physiologically based pharmacokinetic (PBPK) model was developed for inhaled ethylbenzene (EB) in B6C3F1 mice. The mouse physiological parameters were obtained from the literature, but the blood:air and tissue:air partition coefficients were determined by vial equilibration technique. The maximal velocity for hepatic metabolism (Vmax) obtained from a previously published rat study was increased by a factor of ∼3 to account for enzyme induction during repeated exposures. The Michaelis affinity constant (Km) for hepatic metabolism of EB, obtained from a previously published rat PBPK modeling study, was kept unchanged during single and repeated exposure scenarios. Hepatic metabolism alone could not adequately describe the clearance of EB from mouse blood. Additional metabolism was assumed to be localized in the lung. The parameters for pulmonary metabolism were obtained by optimization of PBPK model fits to kinetic data collected following exposures to 75–1000 ppm. The PBPK model successfully predicted all available blood and tissue concentration data in mice exposed to 75 or 750 ppm EB. Overall, the results indicate that the rate of EB clearance is markedly higher in B6C3F1 mice than rats or humans and exceeds the hepatic metabolism capacity. Available biochemical evidence is consistent with a significant role for pulmonary metabolism; however, the extent to which the extrahepatic metabolism is localized in the lung is unclear. Overall, the PBPK model developed for the mouse adequately simulated the blood and tissue kinetics of EB by accounting for its high rate of clearance.