Ann. occup. Hyg., Vol. 47, No. 2, pp. 157-164, 2003
© 2003 British Occupational Hygiene Society
Published by Oxford University Press
Partitioning Theory for Respiratory Deposition of Semivolatile Aerosols
Department of Environmental Sciences and Engineering, University of North Carolina at Chapel Hill, CB 7431 Rosenau Hall, Chapel Hill, NC, 27599-7431, USA
Received 18 July 2002; in final form 7 November 2002
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONTHEORY DEVELOPMENTRESULTS AND DISCUSSIONCONCLUSIONSNOMENCLATUREAPPENDIX AAPPENDIX B: LUNG-AIR PARTITION...REFERENCES |
|---|
The objective of this work is to model the deposition of semivolatile aerosols in the lungs based on gas–particle and tissue–air partitioning theory. Semivolatile compounds exist in air as both particles and gases simultaneously. Mass distributes between the two phases according to a gas–particle partitioning ratio, Rpg = Kp(TSP). Particle deposition in the lungs is a function of aerodynamic diameter, whereas gas deposition is a function of tissue solubility, which is related to the air–lung partitioning ratio, Kla. Therefore, deposition to the lungs will vary with Rpg and Kla. These and other parameters determine a dimensionless deposition number, D, which indicates whether particles or gases are most responsible for deposition of semivolatile chemicals in the lung. The deposition number allows industrial hygienists to design effective air sampling strategies and control measures that will minimize risks associated with exposure to semivolatiles. Examples of deposition numbers for common semivolatile pollutants are provided, including alkanes, polycyclic aromatic hydrocarbons, pesticides and polychlorinated biphenyls.
Keywords: semivolatiles; exposure; deposition; lung; partitioning; Kp, Rpg; dose; metalworking fluid
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONTHEORY DEVELOPMENTRESULTS AND DISCUSSIONCONCLUSIONSNOMENCLATUREAPPENDIX AAPPENDIX B: LUNG-AIR PARTITION...REFERENCES |
|---|
Industrial hygienists often use exposure to calculate risk. Although exposure provides an indication of chemical concentrations measured outside the body, it addresses neither the chemical intake by the body nor chemical uptake by specific organs. Intake is the fraction of exposure that enters the human body, whereas uptake, or deposition, is the amount of material that enters and stays within the body (Crawford-Brown, 1997). Organizations concerned with occupational exposures such as the American Conference of Governmental Industrial Hygienists (ACGIH), International Standardization Organization (ISO) and the Comité Européen Normale (CEN) have recognized intake as an important determinant of health risk with guidelines and standards for inspirable particulate mass (ACGIH, 1985; ISO, 1991; CEN, 1992). These guidelines, in turn, spurred the development of new size selective samplers that help estimate human health risk more accurately than older, ‘total’ particulate samplers (Vincent et al., 1999). Recent outdoor particulate standards have also addressed the importance of intake; the US Environmental Protection Agency’s (EPA) PM2.5 standard is one example (EPA, 2002). However, intake does not guarantee uptake, as inhaled particles sometimes exit the body upon exhalation (e.g. exhaled cigarette smoke). Therefore, uptake is a better determinant of risk than either exposure or intake. The objective of this work is to predict the uptake of semivolatile compounds to the lungs.
Examples of common semivolatile exposures include: metalworking fluid mists, pesticides, polycyclic aromatic hydrocarbons (PAHs) in diesel exhaust, asphalt fumes, polychlorinated biphenyls (PCBs) and dioxins (CONCAWE, 1986; Lioy and Daisey , 1986; Woskie et al., 1988; Norseth et al., 1991; Zaebst et al., 1991; Hawthorne et al., 1996). These compounds usually exist as solids or liquids in bulk form; however, in air they exist simultaneously as particles and gases. The distribution of mass between the two phases is a strong function of vapor pressure. Compounds with high vapor pressures exist primarily in the gas phase whereas those with low vapor pressures exist primarily as particles. Estimates of human exposure to semivolatiles are complicated since both particle and gas phases must be measured simultaneously. Furthermore, conventional sampling methods for semivolatile aerosols are frequently biased; particle-phase exposures are often measured as gas-phase and vice versa (Volckens et al., 1999; Turpin et al., 2000). Therefore, a model that predicts the mode of deposition to the lung of semivolatile chemicals may provide industrial hygienists with important a priori information to reduce risk and limit exposure.
|
THEORY DEVELOPMENT |
|---|
|
TOPABSTRACTINTRODUCTIONTHEORY DEVELOPMENTRESULTS AND DISCUSSIONCONCLUSIONSNOMENCLATUREAPPENDIX AAPPENDIX B: LUNG-AIR PARTITION...REFERENCES |
|---|
Semivolatile compounds are often characterized by the constant, Kp (m3/µg), which describes the partitioning of a compound between the gas and particle phases:
Kp = F/(A · TSP) (1)
where F is the airborne particle phase concentration (ng/m3), A is the gas phase concentration (ng/m3) and TSP is the amount of total suspended particulate matter (µg/m3). A dimensionless form for (1), which we denote as Rpg, is obtained by multiplying Kp by TSP:
Rpg = Kp·(TSP) = F/A (2)
Compounds with Rpg > 1 are present primarily in the particle phase, whereas predominantly gas phase semivolatiles have Rpg < 1. Knowledge of Kp and Rpg is important for deposition because inhaled particles deposit in the lungs through different mechanisms than do gases (Pankow, 2001).
Pankow developed equations to predict partitioning to both solid and liquid particles (Pankow, 1987, 1994). When particles are liquid-like, absorption dominates partitioning and Rpg is expressed as:
(3)
where fom is the fraction of particle mass made up of liquid organic material, R is the gas constant (8.2 x 10–5 m3 atm/K/mol), T is the temperature (K), Mom is the average molecular weight of the particle’s organic mass (g/mol), 
i is the activity coefficient that represents non-ideal interactions between the semivolatile compound and the particle matrix, Pl0 is the liquid vapor pressure of the semivolatile compound (atm), and the factor of 106 is used to convert micrograms to grams. Of all the terms in equation (2), the vapor pressure, Pl0, is the strongest determinant of Rpg. Equation (3) has been shown to predict Rpg accurately when all the constituent parameters are well known (Jang et al., 1997).
The partitioning of chemicals between air and tissue is described by a tissue–air partitioning ratio, Kta, where the subscript ‘t’ represents the particular tissue in question. Hence, the lung–air partitioning ratio, Kla, is expressed as:
Kla = L/A (4)
where L is the concentration of chemical in the lung tissue (ng/m3). Chemical solubility in the lung is usually estimated from linear or log-linear combinations of the product of tissue composition (lipid and water fractions) with compound-specific solubility parameters (octanol–air or oil–air, and water–air partition coefficients). The Rpg ratio is an important determinant for exposure, whereas Kla is an important determinant for deposition.
Gas deposition in the lungs
Gas deposition in the lungs, Dg (ng/m3), is expressed as:
Dg = fd,g·A (5)
where fd,g is the fraction of inhaled gas that deposits with each breath and A is the inhaled, gas-phase concentration (ng/m3). Two assumptions are necessary to incorporate Kla into the model.
1. The volume of inhaled air is equal to the lung tissue volume.
2. Lung surface is large enough for inhaled gas to reach air–tissue equilibrium during one breath.
Human, adult lungs weigh 
1000 g and consist of 98% water and 2% lipid portions, by mass (Crawford-Brown, 1997; Gerde and Scott, 2001). Using 
H2O = 1 g/cm3 we estimate the volume of the tissue in human lungs to be 1000 cm3. Therefore, the first assumption is reasonable for a worker engaging in light activity, whose average tidal volume is 1000 cm3 (Crawford-Brown, 1997). The second assumption is probably valid also, because the bronchial and alveolar regions of the lung have large surface to volume ratios.
Given these assumptions, we use the lung–air partitioning ratio, Kla, to approximate the uptake of gas-phase compounds by the lungs. We define fd,g from Kla as:
(6)
where fd,g represents the fraction of mass that transfers from the gas phase to the tissue phase (lungs) at equilibrium. Substituting (6) into (5):
(7)
Several authors have developed expressions to estimate lung–air partition coefficients (Payne and Kenney, 2002). We chose the model developed by Abraham and Weathersby (1994) and Abraham et al. (1994), which produced the best estimates for human air–lung partitioning ratios as reviewed by Payne and Kenney (2002). With PCBs and DDT, for which no data exists for Abraham’s model, we used the model of Poulin and Krishnan (1996), as adapted by Payne and Kenney (2002) to predict Kla. Details of the model parameters are provided as supplementary material. Further partitioning from the lungs into the blood could be estimated by multiplying Kla by a lung tissue–blood partitioning coefficient; however, such an application is beyond the scope of this work.
Particle deposition in the lungs
Particle deposition in the lungs, Dp (ng/m3), is expressed as the product of the particle deposition fraction, fd,p, and the particle concentration, F (ng/m3):
Dp = fd,p·F (8)
However, the particle deposition fraction varies with particle size:
(9)
where M(dp) is the particle frequency distribution by mass, and Ei is the deposition fraction for particles of size i that penetrate into the lung. Ei is a complicated function of several deposition mechanisms that include impaction, interception, sedimentation, diffusion, and electric force.
We can also develop an empirical expression for the dependence of Ei on dp. Particle deposition curves, such as those provided by Schlesinger (1995) are readily interpolated with a simple algebraic expression. Figure 1 shows an interpolation of Stahlhofen’s model (1989), which agrees well with the data of Schlesinger. The interpolation equation estimates deposition to the lungs for particles between 0.01 and 10 µm in diameter:
|
(10) The product of the interpolation equation and the particle size distribution gives the deposition fraction, fd,p:
(11)
where Fi is the concentration of particles in size range i. This calculation is easily performed with the use of a spreadsheet program.
Dividing (8) by (7), we obtain a dimensionless ratio of particle deposition in the lung (Dp) to gas deposition in the lung (Dg):
(12)
The concentration dependence of equation (12) is removed with the use of equation (1), noting that F/A = Rpg:
(13)
Equation (13) determines whether particles or gases govern deposition to the lungs as a result of exposure to semivolatiles. Particles dominate deposition when D > 1, whereas gases dominate deposition when D < 1.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONTHEORY DEVELOPMENTRESULTS AND DISCUSSIONCONCLUSIONSNOMENCLATUREAPPENDIX AAPPENDIX B: LUNG-AIR PARTITION...REFERENCES |
|---|
Values of Rpg and D for a variety of semivolatile chemicals are listed in Table 1; details of the calculations are given in the supplementary material. In general, exposure correlates well with deposition, as seen in Table 1. Hence, for these chemicals, deposition follows exposure (i.e. Rpg and D are of similar rank).
|
For many pesticides, however, there is no connection between exposure and deposition. With DDT, for example, exposures occur primarily from particles (Rpg = 2.7), whereas the estimated deposition to the lung is dominated by gases (D = 0.44). Hence, gas-phase DDT molecules deposit in the lung with higher efficiency than do DDT molecules absorbed on particles. In general, gas-phase semivolatiles deposit in the lungs with greater efficiency than particle-phase, owing to the higher diffusion rates of gases than particles and to the lipid and aqueous portions of lung tissue, which can accommodate both hydrophobic and hydrophilic compounds, respectively. As a result, many compounds that exist predominantly in the particle phase (Rpg > 1) will transfer more mass in the lungs through gas-phase deposition (D < 1).
Figure 2 represents a typical exposure profile taken from field measurements of metalworking fluid mists, which are semivolatile (Volckens et al., 2000). Metalworking fluids are used as cutting lubricants in the metal-machining industry. These fluids, usually composed of straight-chained alkanes, are liberally applied to the cutting interface to cool and lubricate the metal and to wash away excess chips. An aerosol is created whenever the cutting tool strikes the workpiece, thus generating an inhalation hazard for nearby workers. The alkanes shown in Fig. 2 span the semivolatile range; dodecane, C12, is present primarily in the gas phase, while docosane, C24, is present primarily in the particle phase. Gas-phase alkanes clearly dominate the exposure in Fig. 2 as particles make up less than one-tenth of the inhaled concentration.
|
Figure 3 depicts the estimated deposition profile when equations (7) and (8) are applied to the gas and particle exposures shown in Fig. 2. Data for the calculations are provided in the appendix. The data in Fig. 3 indicate that control of gas-phase exposures is important in the metal-machining industry, as gas-phase deposition dwarfs that of particles. However, to date, most efforts in metal machining plants focus on control of metalworking fluid mist particles and little attention is paid to gases (DHHS, 1998).
|
Model limitations
The simplified model presented here has several limitations. First, the model does not account for enhanced deposition resulting from humidification of hygroscopic particles upon inhalation. Second, the model does not account for the dynamic uptake of reactive gases into the lung; only equilibrium partitioning is considered. Third, the model does not account for the elevation of compound vapor pressures associated with the temperature gradient between ambient air and body-heated air in the lungs. This effect would probably increase the gas-phase deposition of some chemicals. Fourth, the model assumes that mass transfer from lung tissue into the blood is fast enough to prevent saturation at the lung surface. This last assumption is likely to be valid at relatively low concentrations (i.e. <10 000 µg/m3). Finally, although the model predicts uptake, it does not account for metabolism and dose; deposited particles and gases may exert different effects upon the tissue over time.
Estimates of D are approximate; uncertainty is inherent with Rpg, Kla and the particle size distribution. Estimates of Rpg and Kla vary by as much as an order of magnitude (Mackay et al., 1992; Payne and Kenney, 2002). Uncertainty in the inhaled particle size distribution can vary the estimate for D by at most a factor of 5, as seen in Fig. 1.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONTHEORY DEVELOPMENTRESULTS AND DISCUSSIONCONCLUSIONSNOMENCLATUREAPPENDIX AAPPENDIX B: LUNG-AIR PARTITION...REFERENCES |
|---|
A model was developed to estimate the deposition of semivolatile chemicals to the lungs using dimensionless partitioning coefficients. A dimensionless deposition number, D, indicates whether particles or gases are more responsible for deposition. The deposition number can provide a priori information on the relative risks associated with exposure to particles and gases of semivolatile chemicals. Furthermore, D, can guide engineers towards more efficient control strategies to prevent emissions of either particles or gases in the workplace. Future work will address hygroscopic particle growth and vapor pressure elevation within the lung.
Acknowledgements—This work was supported in part by grant T32ES07018 from the National Institute for Environmental Health Sciences. The authors are grateful for the insightful comments of the reviewers.
|
NOMENCLATURE |
|---|
|
TOPABSTRACTINTRODUCTIONTHEORY DEVELOPMENTRESULTS AND DISCUSSIONCONCLUSIONSNOMENCLATUREAPPENDIX AAPPENDIX B: LUNG-AIR PARTITION...REFERENCES |
|---|
A = gas phase concentration (ng/m3)
D = deposition number, dimensionless, D = Dp/Dg
Dp = particle deposition rate to the lungs (ng/m3)
Dg = gas deposition rate to the lungs (ng/m3)
dp = aerodynamic particle diameter (µm)
Ei = deposition efficiency of particles of size i in the lung
fd,g = fraction of inhaled gas that deposits in the lung
fd,p = fraction of inhaled particles that deposit in the lung
F = particle phase concentration (ng/m3)
Fi = particle phase concentration for particle size i (ng/m3)
fom = fraction of particle mass made up of liquid organic material
Kla = lung–air partitioning coefficient, dimensionless
Kp = gas–particle partitioning constant (m3/µg)
Kta = tissue–air partitioning coefficient, dimensionless
L = lung-phase concentration (ng/m3)
M(dp) = particle frequency distribution by mass
Rpg = gas–particle partitioning ratio, dimensionless, Rpg = Kp(TSP)
Pl0 = compound vapor pressure (atm)
R = gas constant, 8.2 x 10–5 m3 atm/K/mol
T = temperature (K)
H2O = water density (g/cm3)
|
APPENDIX A |
|---|
|
TOPABSTRACTINTRODUCTIONTHEORY DEVELOPMENTRESULTS AND DISCUSSIONCONCLUSIONSNOMENCLATUREAPPENDIX AAPPENDIX B: LUNG-AIR PARTITION...REFERENCES |
|---|
Table A1 gives the data for the calculations in Table 1.
|
|
APPENDIX B: LUNG–AIR PARTITION MODELS |
|---|
|
TOPABSTRACTINTRODUCTIONTHEORY DEVELOPMENTRESULTS AND DISCUSSIONCONCLUSIONSNOMENCLATUREAPPENDIX AAPPENDIX B: LUNG-AIR PARTITION...REFERENCES |
|---|
The Abraham equation is a semi-empirical expression that uses a linear combination of the products of several solvatochromatic parameters with tissue-specific regression coefficients (Abraham and Weathersby et al., 1994). For gas-phase partitioning to the lungs, log Kla is expressed as:
(14)
where R2 is the solute excess molar refraction, 
2H is the solute dipolarity/polarizability, 
2H is the solute hydrogen-bond acidity, ß2H is the solute hydrogen-bond basicity, and L16 is the solute hexadecane–air partition coefficient. A list of compound-specific parameters are provided in Table A2 and in Abraham et al. (1994).
|
The model of Poulin and Krishnan (1996) is based on lipid and water volume fractions of lung tissue, and compound solubility in saline (water) and oil (1-octanol). Payne and Kenney (2002) adapted this model during their review of tissue–air partitioning models and express Kla as:
Kla = 0.756Kwa + 0.0057Koa (15)
where Kwa is the dimensionless water–air partitioning coefficient (the inverse of the dimensionless Henry’s law coefficient) and Koa is the dimensionless 1-octanol–air partition coefficient.
|
FOOTNOTES |
|---|
* Author to whom correspondence should be addressed. Tel: 919-966-7337; fax: 919-966-7911; e-mail: volckens{at}unc.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONTHEORY DEVELOPMENTRESULTS AND DISCUSSIONCONCLUSIONSNOMENCLATUREAPPENDIX AAPPENDIX B: LUNG-AIR PARTITION...REFERENCES |
|---|
Abraham MH, Weathersby PK. (1994) Hydrogen bonding. 30. Solubility of gases and vapors in biological liquids and tissues. J Pharm Sci; 83(10): 1450–6.
Abraham MH, Chadha HS, Whiting GS, Mitchell RC. (1994) Hydrogen bonding. 32. An analysis of water–octanol and water alkane partitioning and the Dlog P parameter of Seiler. J Pharm Sci; 83(8): 1085–100.
ACGIH. (1985) Particle size selective sampling in the workplace. Cincinnati, OH: American Conference of Governmental Industrial Hygienists.
CEN. (1992) Size fraction definitions for measurement of airborne particles in the workplace. Brussels, European Standardization Committee. EN 481.
CONCAWE. (1986) Health aspects of worker exposure to oil mists. Brussels 86/89.
Crawford-Brown DJ. (1997) Theoretical and mathematical foundations of human health risk analysis: biophysical theory of environmental health science. Boston, MA: Kluwer. ISBN 079239898X.
CRC. (1997) Handbook of physical properties of organic chemicals. Boca Raton, FL: CRC Press. ISBN 1566702275.
DHHS, CDC. (1998) Criteria for a recommended standard: occupational exposure to metalworking fluids. Cincinnati, OH: National Institute for Occupational Safety and Health. pp. 98–102.
EPA. (2002) Air quality critera for particulate matter. Research Triangle Park: US Environmental Protection Agency. EPA/600/P-99/002aC.
Gerde P, Scott BR. (2001) A model for absorption of low-volatile toxicants by the airway mucosa. Inhal Toxicol; 13: 903–29.[CrossRef][Web of Science][Medline]
Green CE, Abraham MH, Acree WE, De Fina KM, Sharp TL. (2000) Solvation descriptors for pesticides from the solubility of solids: diuron as an example. Pest Manage Sci; 56: 1043–53.[CrossRef]
Hawthorne SB, Miller DJ, Louie PKK, Butler RD, Mayer GG. (1996) Vapor-phase and particulate-phase pesticides amd PCB concentrations in eastern North Dakota air samples. J Environ Qual; 25: 594–600.
ISO. (1991) Air quality—particle size fraction definitions for health-related sampling. Geneva: International Organization for Standardization. CD 7708.
Jang M, Kamens RM, Leach K, Strommen MR. (1997) A thermodynamic approach using group contribution methods to model the partitioning of semi-volatile organic compounds on atmospheric particulate matter. Environ Sci Technol; 31: 2805–11.[CrossRef]
Lioy PJ, Daisey JM. (1986) Airborne toxic elements and organic substances. Environ Sci Technol; 20: 8–14.
Mackay D, Shiu WY, Ma KC. (1992) Illustrated handbook of physical-chemical properties and environmental fate for organic chemicals. Boca Raton, FL: Lewis.
Norseth T, Waage J, Dale I. (1991) Acute effects and exposure to organic compounds in road maintenance workers exposed to asphalt. Am J Ind Med; 20: 737–44.[Web of Science][Medline]
Pankow JF. (1987) Review and comparative analysis of the theories on partitioning between the gas and aerosol particulate phases in the atmosphere. Atmos Environ; 21: 2275–83.[CrossRef]
Pankow JF. (1994) An absorption model of the gas/particle partitioning of organic compounds in the atmosphere. Atmos Environ; 28: 185–8.[CrossRef]
Pankow JF. (2001) A consideration of the role of gas/particle partitioning in the deposition of nicotine and other tobacco smoke compounds in the respiratory tract. Chem Res Toxicol; 14: 1465–81.[CrossRef][Medline]
Payne MP, Kenny LC. (2002) Comparison of models for the estimation of biological partition coefficients. J Toxicol Environ Health; 65A: 897–931.
Poulin P, Krishnan K. (1996) A tissue-based composition algorithm for predicting tissue:air partition coefficients of organic chemicals. Toxicol Appl Pharmacol; 136: 126–30.[CrossRef][Web of Science][Medline]
Schlesinger RB. (1995) Deposition and clearance of inhaled particles. In McClellan RO, Henderson RF, editors. Concepts in Inhalation, 3rd edn. Washington, DC: Taylor & Francis.
Schwarzenbach RP, Gschwend PM, Imboden DM. (1997) Environmental organic chemistry. New York: John Wiley & Sons. ISBN 0471839418.
Stahlhofen W, Rudolf G, James AC. (1989) Intercomparison of experimental regional aerosol deposition data. J Aerosol Med; 2: 285–308.
Turpin BJ, Saxena P, Andrews E. (2000) Measuring and simulating particulate organics in the atmosphere: problems and prospects. Atmos Environ; 34: 2983–3013.[CrossRef]
Vincent JH, Ramachandran G, Thomassen Y, Keeler GJ. (1999) Application of recent advances in aerosol sampling science towards the development of improved sampling devices: the way ahead. J Environ Monit; 1: 285–92.[Medline]
Volckens J, Boundy M, Leith D, Hands D. (1999) Oil mist concentration: a comparison of sampling methods. Am Ind Hyg Assoc J; 60: 684–9.
Volckens J, Boundy M, Leith D. (2000) Mist concentration measurements II: Laboratory and field evaluations. Appl Occup Environ Hyg; 15: 370–9.[Medline]
Woskie SR, Smith TJ, Hammond SK, Schenker MB, Garshick E, Speizer FE. (1988) Estimation of the diesel exhaust exposures of railroad workers: I. Current exposures. Am J Ind Med; 13(3): 381–94.
Zaebst DD, Clapp DE, Blade LM et al. (1991) Quantitative determination of trucking industry workers’ exposures to diesel exhaust particles. AIHA J; 52(12): 529–41.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. W. Kim and P. C. Raynor A New Semivolatile Aerosol Dichotomous Sampler Ann. Hyg., April 1, 2009; 53(3): 239 - 248. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||