Scholarly article on topic 'Simulation of aerosol dynamics and deposition of combustible and electronic cigarette aerosols in the human respiratory tract'

Simulation of aerosol dynamics and deposition of combustible and electronic cigarette aerosols in the human respiratory tract Academic research paper on "Nano-technology"

CC BY
0
0
Share paper
Academic journal
Journal of Aerosol Science
OECD Field of science
Keywords
{Modeling / "Aerosol dynamics" / Deposition / "Cigarette aerosols" / "Human respiratory tract"}

Abstract of research paper on Nano-technology, author of scientific article — Lukas Pichelstorfer, Werner Hofmann, Renate Winkler-Heil, Caner U. Yurteri, John McAughey

Abstract The Aerosol Dynamics in Containments (ADiC) model describes the dynamic changes of inhaled cigarette smoke droplets during puffing, mouth-hold, and inspiration and expiration, considering coagulation, phase transition, conductive heat and diffusive/convective vapor transport, and dilution/mixing. The ADiC model has been implemented into a single path representation of the stochastic lung dosimetry model IDEAL to compute particulate phase deposition as well as vapor phase deposition in the airway generations of the human lung. In the present study, the ADiC model has been applied to the inhalation of combustible and electronic cigarette aerosols. Aerosol dynamics processes significantly influence the physical properties of particle and vapor phase in the human respiratory tract: (i) number reduction of inhaled aerosol particles is caused primarily by coagulation and less by deposition for both aerosols; (ii) hygroscopic growth rates are higher for e-cigarettes than for combustible cigarettes; (iii) the effect of particle growth on deposition leads to a lower total deposition in the case of cigarette smoke particles and a higher total deposition in the case of e-cigarette droplets relative to their initial size distributions; and, (iv) most of the nicotine is deposited by the vapor phase for both aerosols.

Academic research paper on topic "Simulation of aerosol dynamics and deposition of combustible and electronic cigarette aerosols in the human respiratory tract"

Journal of Aerosol Science I (I

Contents lists available at ScienceDirect

Journal of Aerosol Science

journal homepage: www.elsevier.com/locate/jaerosci

Simulation of aerosol dynamics and deposition of combustible and electronic cigarette aerosols in the human respiratory tract

Lukas Pichelstorfer a,n, Werner Hofmanna, Renate Winkler-Heila, Caner U. Yurterib, John McAugheyb

a Division of Physics and Biophysics, Department of Chemistry and Physics of Materials, University of Salzburg, Heilbrunner Str. 34, 5020 Salzburg, Austria

b British American Tobacco, Group Research & Development, Southampton, UK

ARTICLE INFO

ABSTRACT

Keywords:

Modeling

Aerosol dynamics

Deposition

Cigarette aerosols

Human respiratory tract

The Aerosol Dynamics in Containments (ADiC) model describes the dynamic changes of inhaled cigarette smoke droplets during puffing, mouth-hold, and inspiration and expiration, considering coagulation, phase transition, conductive heat and diffusive/convective vapor transport, and dilution/mixing. The ADiC model has been implemented into a single path representation of the stochastic lung dosimetry model IDEAL to compute particulate phase deposition as well as vapor phase deposition in the airway generations of the human lung. In the present study, the ADiC model has been applied to the inhalation of combustible and electronic cigarette aerosols. Aerosol dynamics processes significantly influence the physical properties of particle and vapor phase in the human respiratory tract: (i) number reduction of inhaled aerosol particles is caused primarily by coagulation and less by deposition for both aerosols; (ii) hygroscopic growth rates are higher for e-cigarettes than for combustible cigarettes; (iii) the effect of particle growth on deposition leads to a lower total deposition in the case of cigarette smoke particles and a higher total deposition in the case of e-cigarette droplets relative to their initial size distributions; and, (iv) most of the nicotine is deposited by the vapor phase for both aerosols.

© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY

license (http://creativecommons.org/licenses/by/4.0/).

1. Introduction

Tobacco smoke is a dynamic mixture of particulate and vapor phase constituents, consisting of a multitude of chemical substances (Baker & Dixon, 2006; Roemer et al., 2012; St. Charles, McAughey, & Shepperd, 2013). Deposition of inhaled cigarette smoke particles depends on the aerosol properties of the particle phase, such as size distribution and particle concentration, as well as on the physical and chemical form of each compound. The primary compounds commonly found in cigarettes are tar, nicotine, and water (Roemer et al., 2012). Since inhalation of mainstream tobacco smoke is a recognized health risk as a result of chronic exposure to tobacco smoke toxicants, dosimetry of inhaled smoke aerosols in the human respiratory tract is an important component of any toxicological assessment (Asgharian, Price, Yurteri, Dickens, & McAughey, 2014; Pichelstorfer, Winkler-Heil, & Hofmann, 2013; Robinson & Yu, 2001). In recent times, electronic cigarettes are

* Corresponding author. Tel.: + 43 662.8044 5718; fax: + 43 662 8044 150. E-mail address: Lukas.Pichelstorfer@sbg.ac.at (L. Pichelstorfer).

http://dx.doi.org/10.1016/joaerosci.2016.01.017

0021-8502/© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

becoming increasingly popular because they are considered as a less harmful alternative to conventional smoking due to lower levels of combustion related toxicants. The primary compounds commonly considered for deposition modeling are nicotine, water and glycerol, propylene glycol or a combination of both (Fuoco, Buonanno, Stabile, & Vigo, 2014).

In the case of inhalation of inert particles commonly used in human inhalation studies, inhaled particle sizes remain constant during a complete breathing cycle, and deposition is the only mechanism reducing the particle concentration as coagulation based number concentration reduction is negligible due to its low number concentration. In case of inhalation of combustible and electronic cigarette aerosols, however, inhaled particles experience, besides deposition, coagulation, condensation of water vapor, evaporation of semi-volatile compounds and chemical reactions and deposition, which change particle diameters and concentration. Moreover, in addition to particle transport in the lungs, it is necessary to simulate the transport and deposition of the vapor produced by evaporation from the particle phase. For example, nicotine is deposited in the lungs in both particulate and vapor form.

Thus the objectives of the present study are: (i) to apply an aerosol dynamics model that considers the effects of coagulation, phase transition, conductive heat and diffusive vapor transport, and phase transition to the inhalation of combustible and electronic cigarette aerosols; and, (ii) to compare the dynamic changes of the inhaled aerosols and the resulting deposition patterns between cigarette smoke particles and e-cigarette droplets.

2. Materials and methods

2.1. ADiC aerosol dynamics model

2.1.1. Overview

The dynamic changes of inhaled cigarette smoke particles and electronic cigarette droplets during puffing, mouth-hold and within the lungs during inspiration and expiration are described by the recently developed aerosol dynamics model ADiC (Aerosol Dynamics in Containments) (Pichelstorfer & Hofmann, 2015), which considers coagulation, conductive heat transport, diffusive vapor transport, phase transition and chemical reactions in a confined space. In the present study, the confined space is a cylindrical human airway. For the simultaneous simulation of coagulation, phase transition, heat and vapor transport, dilution, mixing and deposition, an airway is subdivided into several length segments. The salient features of the different dynamic processes are briefly described below.

2.1.2. Coagulation

For sufficiently high particle concentrations, coagulation reduces particle concentration, increases particle diameters and changes the composition of the particles (Pichelstorfer et al., 2013). Two droplets, i.e. spherical particles, may collide due to their relative motion. Relative motions can be caused by thermal motion, gravitational settling, laminar shear, turbulent flow, electrical charge and inertial effects at airway bifurcations, with thermal motion being the dominant mode of coagulation (Pichelstorfer et al., 2013). In the present study thermal motion is the only mechanism considered to cause coagulation. As a result, a new bigger, spherical particle is formed, while the particle concentration is reduced by one particle.

2.1.3. Phase transition

The particle/vapor phase transition affects the diameter of the droplet, either increasing diameters by condensation or decreasing diameters by evaporation of volatile compounds, as well as the composition of the particles. The mole flux towards the droplet or away from the droplet depends on the vapor pressure difference between the droplet and the far field. Likewise, the heat flux depends on the temperature difference between the droplet and the far field (Pichelstorfer & Hofmann, 2015). Since the vapor pressure on the particle is a function of the droplet temperature, and the droplet temperature is a function of the mole flux, mass and heat flux are described by coupled differential equations.

2.1.4. Heat and vapor transport

Heat and vapor transport affect vapor concentration and the temperature in an airway volume. An airway wall is a heat reservoir and can be a sink or source of vapor concentration, depending on the concentration gradient between airborne and surface vapor concentration. The physical mechanisms describing heat and vapor transport are conductive heat transfer and diffusive vapor transport (Pichelstorfer & Hofmann, 2015). In the extrathoracic and the alveolar region, diffusive transport was simulated in a spherical volume, while a laminar flow field was applied in the conducting (non-alveolar) airways.

2.1.5. Deposition

Deposition of particles passing through an airway decreases the particle concentration and, because of the size-specificity of operating physical deposition mechanisms, also the particle diameter distribution. Size-specific deposition mechanisms considered in the present study were diffusion in the oral cavity during puffing and mouth-hold, and diffusion, impaction and sedimentation in bronchial and alveolar airways during inhalation and exhalation (Hofmann, 2011). Note that diffusion is the dominant deposition mechanism for cigarette smoke particles in the airways of the human lung, while

e-cigarette droplets are primarily deposited by inertial impaction in proximal bronchial airways and sedimentation in peripheral alveolated airways.

2.1.6. Dilution, mixing

Dilution and mixing in the oral cavity between the residual air in the mouth and the inhaled air volume during puffing and mouth-hold affects all particle and vapor parameters. During puffing, a constant volume flow (50 mL puff during a 2 s puffing time) was applied that feeds fresh smoke to the continuously aging aerosol in the oral cavity.

The composition of both cigarette aerosols and related model parameter values required for the aerosol dynamics simulations are listed in the Appendix.

2.3. IDEAL particle deposition model

Particle deposition in the airway generations of the human lung is computed with the IDEAL (Inhalation, Deposition and Exhalation of Aerosols in the Lungs) Monte Carlo code (Hofmann, 2011; Hofmann, Winkler-Heil, & Balashazy, 2006; Koblinger & Hofmann, 1990), which simulates the random walk of individual inhaled and exhaled particles through an asymmetric, stochastic airway model of the human lung (Koblinger & Hofmann, 1985, 1990). The behavior of inhaled particles is simulated by the action of individual particles inhaled at random times during the inhalation phase. Since coagulation requires information on the particle concentration in each airway length segment, instead of tracking individual particles, the random path of an elemental air volume (bolus) containing information on size distribution and concentration is simulated, thereby adding an Eulerian element to the initially Lagrangian random path model (Pichelstorfer et al., 2013). Because of the complexity of the model and the resulting extensive computational time, a single path version of the IDEAL airway geometry was used, i.e. average airway dimensions for each airway generation were derived for the particle and vapor transport in the lungs. However, average deposition fractions for each airway generation were based on the full stochastic deposition model.

d [nm]

d [nm]

Fig. 1. Temporal evolution of the number size distribution of inhaled cigarette smoke particles (panel A) and e-cigarette droplets (panel B) during puffing, mouth-hold (MH), inhalation and exhalation, based on the same initial size distribution.

3. Results

For the deposition calculations, cigarette smoke particles were characterized by a number size distribution with a median diameter of 163 nm and a GSD of 1.44 and a particle concentration of 1.54 x 109 cm(Ingebrethsen, Alderman, & Ademe, 2011). Since Fuoco et al. (2014) found that particle size distributions and number concentrations of e-cigarette droplets are similar to those for combustible cigarette aerosols, the same size distributions and particle concentrations were adopted for both cigarette aerosols. Indeed, spectral transmission measurements of e-cigarette yielded particle size distributions that are more comparable to those of combustible cigarettes than had been suggested by measurements that required high levels of dilution (Ingebrethsen, Cole, & Alderman, 2012).

The following inhalation conditions were assumed for cigarette smoke particles: 2 s puff inhalation (continuous inhalation of 50 cm3 into oral cavity), 1 s mouth-hold, 2.5 s inhalation into the lungs with a 750 cm3 tidal volume (inhalation of a 50 cm3 bolus from the mouth during the first 0.167 s, followed by aerosol-free air in the remaining 2.366 s) (Ingebrethsen et al., 2011), and 2.5 s exhalation. Current information on breathing conditions for e-cigarettes suggest that inhalation parameters do not deviate significantly from those for conventional cigarettes, although considerable subjective differences have been reported (Evans & Hoffman, 2014; Norton, June, & O'Connor, 2014). Thus for modeling purposes, the above detailed breathing pattern for combustible cigarettes was also assumed for the inhalation of e-cigarette droplets.

The temporal evolution of the number size distribution of inhaled cigarette smoke particles (panel A) and e-cigarette droplets (panel B) after puffing, mouth-hold, inhalation and exhalation, based on the same initial size distribution, is plotted in Fig. 1. As can be seen on both panels, most of the inhaled particles are already removed during the puffing phase and the mouth-hold period due to coagulation. Since thermal motion is the dominant coagulation mechanism, the majority of particles are eliminated in the smallest size fractions. Further removal of particles in the mouth and in the airways of the

condensed deposition fraction [%] vapor deposition fraction [%]

21 0,16 -

£ с о

о. ф

тз тз <ю со

тз с

lung generation

condensed deposition fraction [%] vapor deposition fraction [%]

0,12 -

0,08 -

0,04 -

10 20 lung generation

Fig. 2. Particle and vapor phase nicotine deposition fractions of cigarette smoke particles (panel A) and of e-cigarette droplets (panel B) as a function of airway generation numbers, normalized to the total mass inhaled for each product.

lung is caused by size-selective deposition, i.e. deposition by Brownian motion increases with decreasing particle diameter. In summary, the shift to larger particle diameters during a full breathing cycle is caused by coagulation, primarily in the mouth, and by hygroscopic growth, i.e. growth rates increase with rising particle diameter (Winkler-Heil, Ferron, & Hofmann, 2014), and size-selective deposition in the airways of the lung. Since hygroscopic growth rates of e-cigarette droplets are potentially significantly higher than those for cigarette smoke particles, their shift to larger particle diameters at the end of inhalation and exhalation is much more pronounced. The small peak at about 40 nm in the e-cigarette size distribution is caused by evaporation in the mouth during puffing and mouth-hold. Furthermore, a split of the main peak starting in the alveolar region can be observed, where nicotine is already almost completely removed. Both effects can be explained by the Kelvin effect, which describes the vapor pressure above a curved surface. Thus, the aerosol particles formed by the electronic cigarette will never be in equilibrium with their surrounding as they consist of volatile substances. While the particles at the lower end of the particle size distribution start to shrink by evaporation of glycerol and water, the bigger particles are growing at their expense. In contrast, combustion type cigarettes are considered to contain a semi-soluble and non-volatile tar fraction which has a stabilizing effect.

Particle and vapor phase nicotine deposition fractions of cigarette smoke particles (panel A) and of e-cigarette droplets (panel B) in a single breath as a function of airway generation number are plotted in Fig. 2. Note that the deposition fractions are normalized to the total inhaled nicotine mass for each product and thus represent relative values, while the absolute values are higher by about a factor of 3 for the conventional cigarettes because of the higher amount of inhaled nicotine, i.e. 7.2% vs. 2.5% (see the Appendix). For both aerosols about 99% is deposited by the vapor phase, which emphasizes the necessity of aerosol dynamics considerations for the correct determination of nicotine deposition in the lung. In contrast to the deposition of nicotine, the model suggests that 37.9% of the glycerol, initially present in the electronically generated aerosol, is deposited in the condensed phase, while only 2% are deposited in vapor form. While the maximum of particle deposition occurs at about airway generation 11 for both aerosols, vapor phase deposition peaks at about generation 12 for cigarette smoke particles and at about generation 15 for e-cigarette droplets. The nicotine vapor deposition patterns result from the nicotine vapor production through evaporation from the condensed phase, which increases with depth of penetration into the lungs, and from the subsequent, almost immediate deposition on the airway walls by diffusion due to their high diffusion coefficient. Moreover, it is important to note that vapor phase deposition strongly depends on the interaction of nicotine with the aerosol, such as chemical reactions, which may shift vapor phase deposition to more peripheral airway generations.

To emphasize the necessity of the application of an aerosol dynamics model, deposition of both the 3R4F reference cigarette (Roemer et al., 2012) and the electronic cigarette in the lung were simulated in the same lung geometry without considering coagulation, heat and vapor transport and phase transition, and compared to the present aerosol dynamics calculations (Table 1). The results for number concentrations, median number and mass diameters, and deposition fractions of non-volatiles and nicotine are summarized in Table 1. Note that deposition fractions refer to deposition in a single breath, i.e. any differences in the number of puffs between the two types of cigarettes are not considered. Within the lung the number reduction is higher if coagulation is not considered. This is due to the fact that coagulation reduces the particle number concentration mainly during puffing and mouth-hold by removing primarily small particles due to their higher diffusivity.

Median mass and number diameters remain almost constant if coagulation and phase transition is neglected. Furthermore, deposition fractions of nicotine and non-volatile components are the same as volatiles do not leave the condensed phase and, therefore, particle deposition is the only mechanism to reduce these aerosol components. Note that nicotine deposition calculated that way is considerably lower than suggested by experimental data. For example, Armitage, Dixon, Frost, Mariner, and Sinclair (2004) found complete removal of nicotine from the exhaled aerosol while only 35% of the nonvolatile phase is retained in the lung in terms of mass for comparable breathing parameters.

When aerosol dynamics are considered, total particle number and total aerosol surface area are reduced. Small particles that constitute a considerable number and surface area of the inhaled aerosol are removed by coagulation, shifting the median particle diameter towards larger diameters. Due to the Kelvin term, these bigger particles have a higher potential to grow by water condensation. While combustion type cigarette aerosols feature moderate growth rates of the median

Table 1

Simulation of various aerosol parameters with and without considering aerosol dynamics for combustible and electronic cigarette aerosols. Relative parameter changes (in percent) refer to the difference between inhaled and exhaled parameter values and deposition fractions represent deposition in the lung. Note that the nicotine deposition fractions presented in this table are normalized to the total mass inhaled for each product.

Aerosol parameter No dynamics 3R4F (dynamics) e-cigarettes (dynamics)

Number concentration - 38.5% - 7.9% ( - 74.5%)a -19.8% ( - 90.6%)

Median number diameter 6.0% + 68% ( +146%) + 242% ( + 395%)

Median mass diameter 4.0% + 79% ( + 268%) +466% ( + 892%)

Deposition (non-volatiles)b 29.1% 22.7% 37.9%

Deposition (nicotine) 29.1% ~ 100% -100%

Note that glycerol has a very low vapor pressure that is considered in the aerosol dynamics scenario. However, deposition of vapor phase glycerol is almost negligible (-2% of total glycerol). a Numbers in parenthesis include combined relative changes (mouth and lung). b Non-volatiles are represented by glycerol in the case of electronic cigarettes.

diameters, electronic cigarettes have the potential for high growth rates. This is due to complete solubility of glycerol and the lower molecular mass compared to cigarette smoke tar. Based on data from Li and Hopke (1993), 60% solubility and a molecular mass of 300 kg mol~1 was assumed for modeling purposes (see Appendix for further information). As a result, hygroscopic growth and coagulation of combustion type aerosols shifts the size distribution toward the deposition minimum of the U-shaped total deposition curve of the lung, thus reducing total deposition, i.e. 22.9% of total inhaled non-volatiles when dynamic processes are considered vs. 29.1% when dynamic processes are neglected, while the growth of electronic cigarette aerosols shifts the size distribution beyond the deposition minimum to the ascending part of the deposition curve, thereby increasing total deposition, i.e. 37.9% of glycerol is deposited in the condensed phase. Thus, different deposition mechanisms are important for these two types of cigarettes: for the combustion type cigarette diffusion is the dominant mode of deposition, while deposition of the electronic cigarette droplets is driven primarily by inertial effects in the bronchial airways and gravitational effects in the alveolar region.

4. Discussion and conclusions

Differences in initial size distributions, i.e. e-cigarettes may have a significant fraction of very small particles (Fuoco et al., 2014), are almost removed after the mouth-hold phase. Thus median diameters and number concentrations are similar after the mouth-hold phase for both cigarette aerosols. As a result, potential differences in inhaled size distributions will hardly affect deposition. Potential differences in breathing parameters (Evans & Hoffman, 2014; Norton et al., 2014), e.g. the slightly longer puff duration of e-cigarettes, will lead to a slightly lower flow rate, and hence to slightly higher deposition fractions by diffusion, while a higher puff volume (Norton et al., 2014) will lead to a higher flow rate, and hence to a slightly lower deposition fractions by diffusion. Thus, on average, potential differences in puff topography will not change significantly the deposition results presented in this study.

The present calculations have demonstrated that aerosol dynamics significantly influences particle and vapor phase deposition patterns of inhaled cigarette smoke particles and e-cigarette droplets within the human respiratory tract. The most significant results can be summarized as

• Number reduction of inhaled cigarette aerosols is caused primarily by coagulation ( > 95%) and to a lesser extent (about 5%) by deposition for both combustible and electronic cigarette aerosols.

• Electronic cigarette droplets have potentially much higher hygroscopic growth rates than cigarette smoke particles. As a result, lung deposition of e-cigarettes is higher than that of combustible cigarettes for the specific assumptions made in this modeling study.

• Deposition of the non-volatile fraction of the combustion type cigarette is 40% lower than that of the condensed phase of glycerol of the electronic cigarette (see Table 1).

• For the composition of the products simulated in the present study (see the Appendix), nicotine deposition for combustible cigarettes is higher by about a factor of 3 than that for e-cigarettes because of the higher amount of initially inhaled nicotine. For different formulations, however, the effects of coagulation, hygroscopic growth, evaporation, and chemical interactions may lead to higher or lower deposition fractions for each product.

• About 99% of the nicotine is deposited by the vapor phase for both cigarette aerosols, while only a minute fraction is deposited by the particle phase. This observation is consistent with the experimental results of Armitage et al. (2004) who found that practically all nicotine is deposited in the respiratory tract, while the retention of non-volatile solanesol was significantly lower.

In conclusion, if aerosol dynamics processes are not considered for the inhalation of cigarette aerosols, this (i) would lead to an overestimation of cigarette smoke particle deposition and an underestimation of electronic cigarette droplet deposition in the lung, and (ii) and would grossly underestimate nicotine deposition by neglecting vapor phase deposition, which dominates nicotine deposition. This convincingly demonstrates that aerosol dynamics processes must be included into cigarette aerosol deposition modeling to correctly predict deposition patterns in the lung for both combustible and electronic cigarette aerosols. It should be noted, however, that the results presented here are based on specific assumptions and thus may not be generalized.

Acknowledgments

This research was funded in part by British American Tobacco (Investments) Limited, Southampton, UK (Grant no. A10012363SV).

Appendix

Composition and main model parameter values for cigarette smoke particles and electronic cigarette droplets used in the present study.

L. Pichelstorfer et al. / Journal of Aerosol Science I (l

Nomenclature

D diffusion coefficient

kv thermal conductivity

L latent heat of vaporization

Mm molecular mass

pv vapor pressure

p density

Mmwater 18.016x10-3 [kg/mol] 1 Pwater 1049.572 - 0.1763T [kg/m3] 2

ln(pv,water) 77.34491296 - 7235 424651 - 8.2 x log(T) + 0.0057113T[Pa] 3

Dwater.air 1.9545 x 10-4l1pîî[m2/s]

kvwater - 6.7194 x 10-3+ 7.4857 x 10-3T [W/m K]

Lwater 3.14566 x 106 - 2361.64T[J/kg]

Nicotine

Mm,nicotine 162.26x10-3 [kg/mol] Pnicotine 1010 [kg/m3]

log(pvnicotine/7.5006 x 10-3) 172.8-9492-60.6log(T) + 0.0248T[Pa] 2

Dnicotine,air 6.5x10- 6[m2/s]3

Lnicotine 2.302585 x 10 - [9492. - 60.623tj2585 + 0.0248T^ [J/kg] 5

Glycerol

Mm,glycerol92.09x10-3 [kg/mol] p 1261 [kg/m3]

Pv [Pa] 8.932 x 10-2

Dglycerol,air1.5x10 - 5 [m2 /s]

Lglycerol 9.957x105 [J/kg]

e-cigarette (generic formulation)

Glycerol 88.5% Water 9.0% Nicotine 2.5%

It is further assumed that compositions of solution and aerosol particles are identical. 3R4F reference cigarette Average composition of condensed material 4

Tar 81.7% Nicotine 7.2% Water 11.1%

1 Wukalowitsch (1958).

2 Boldridge and Kelly (1988).

3 Eatough et al. (1989).

4 Roemer et al. (2012).

Water solubility of dried mainstream cigarette smoke 5

Soluble fraction 60% Mmtar 0.3 [kg/mol] 6

Particle number distribution 7

Concentration 1.54x109[ 1/cm3] dmedian 1.63 nm8 Ogsd 1.44

Note: Tar is the only non-volatile compound.

Only 9% of the particle phase nicotine is assumed to be present in its volatile form.9

References

Armitage, A. K., Dixon, M., Frost, B. E., Mariner, D. C., & Sinclair, N. M. (2004). The effect of inhalation volume and breath-hold duration on the retention of nicotine and solanesol in the human respiratory tract and on subsequent plasma nicotine concentrations during cigarette smoking. Beiträge zur Tabakforschung International, 21, 240-249.

Asgharian, B., Price, O. T., Yurteri, C. U., Dickens, C., & McAughey, J. (2014). Component-specific, cigarette particle deposition modeling in the human respiratory tract. Inhalation Toxicology, 26, 36-47.

Baker, R. R., & Dixon, M. (2006). The retention of tobacco smoke constituents in the human respiratory tract. Inhalation Toxicology, 18, 255-294.

Boldridge, D.W., & Kelly, S.P. (1988). Nicotine vapor pressure at moderate temperatures. R&DM Report no. 94, RJR, obtained from <http://legacy.library.ucsf. edu/tid/vmy13d00;jsessionid =4524EEB0D05C7AB347B468826A462B2E).

Eatough, D. J., Benner, C. L., Bayona, J. M., Richards, G., Lamb, J. D., Lee, M. L.,... Hansen, L. D. (1989). Chemical composition of environmental tobacco smoke. 1. Gas-phase acids and bases. Environmental Science and Technology, 23, 679-687.

Evans, S. E., & Hoffman, A. C. (2014). Electronic cigarettes: abuse liability, topography and subjective effects. Tobacco Control, 23, 23-29.

Fuoco, F. C., Buonanno, G., Stabile, L., & Vigo, P. (2014). Influential parameters on particle concentration and size distribution in the mainstream of e-cigarettes. Environmental Pollution, 184, 523-529.

Hofmann, W. (2011). Modelling inhaled particle deposition in the human lung - a review. Journal of Aerosol Science, 42, 693-724.

Hofmann, W., Winkler-Heil, R., & Balashazy, I. (2006). The effect of morphological variability on surface deposition densities of inhaled particles in human bronchial and acinar airways. Inhalation Toxicology, 18, 809-819.

Ingebrethsen, B. J., Alderman, S. L., & Ademe, B. (2011). Coagulation of mainstream cigarette smoke in the mouth during puffing and inhalation. Aerosol Science and Technology, 45, 1422-1428.

Ingebrethsen, B. J., Cole, S. K., & Alderman, S. L. (2012). Electronic cigarette aerosol particle size distribution measurements. Inhalation Toxicology, 24, 976-984.

Koblinger, L., & Hofmann, W. (1985). Analysis of human lung morphometric data for stochastic aerosol deposition calculations. Physics in Medicine and Biology, 30, 541-556.

Koblinger, L., & Hofmann, W. (1990). Monte Carlo modelling of aerosol deposition in human lungs. Part I: Simulation of particle transport in a stochastic lung structure. Journal of Aerosol Science, 21, 661-674.

Lauterbach, J. H., Bao, M., Joza, P. L., & Rickert, W. S. (2010). Free-base nicotine in tobacco products. Part I. Determination of free-base nicotine in the particulate phase of mainstream cigarette smoke and the relevance of these findings to product design parameters. Regulatory Toxicology and Pharmacology, 58, 45-63.

Li, W., & Hopke, P. K. (1993). Initial size distributions and hygroscopicity of indoor combustion aerosol particles. Aerosol Science and Technology, 19, 305-316.

Norton, K. J., June, K. M., & O'Connor, R. J. (2014). Initial puffing behaviors and subjective responses differ between an electronic nicotine delivery system and traditional cigarettes. Tobacco Induced Diseases, 12, 17-24.

Pichelstorfer, L., & Hofmann, W. (2015). Modeling aerosol dynamics of cigarette smoke in a denuder tube. Journal of Aerosol Science, 88, 72-89.

Pichelstorfer, L., Winkler-Heil, R., & Hofmann, W. (2013). Lagrangian/Eulerian model of coagulation and deposition of inhaled particles in the human lung. Journal of Aerosol Science, 64, 125-142.

Robinson, R. J., & Yu, C. P. (2001). Deposition of cigarette smoke particles in the human respiratory tract. Aerosol Science and Technology, 34, 202-215.

Roemer, E., Schramke, H., Weiler, H., Buettner, A., Kausche, S., Weber, S.,... Wittke, S. (2012). Mainstream smoke chemistry and in vitro and in vivo toxicity of the reference cigarettes 3R4F and 2R4F. Beiträge zur Tabakforschung International, 25, 316-335.

St. Charles, F. K., McAughey, J., & Shepperd, C.J. (2013). Methodologies for the quantitative estimation of toxicant dose to cigarette smokers using physical, chemical and bioanalytical data. Inhalation Toxicology, 25, 383-397.

Weast, R. C. (1972). Handbook of chemistry and physics (53rd ed.). Cleveland, OH: The Chemical Rubber CompanyD-118.

Winkler-Heil, R., Ferron, G., & Hofmann, W. (2014). Calculation of hygroscopic particle deposition in the human lung. Inhalation Toxicology, 26, 193-206.

Wukalowitsch, M. (1958). Thermodynamische Eigenschaften des Wassers und des Wasserdampfes. Berlin: VEB Verlag Technik.

5 Li and Hopke (1993).

6 Calculated on the basis of 99.25% RH (mean of the relative humidity interval), obtained from Eq. (3) in Li and Hopke (1993).

7 Ingebrethsen, Alderman, and Ademe (2011).

8 Mass corrected: diameter times 1.186.

9 The simulation results in a water fraction of 84% (in the case of combustible cigarette aerosols) and 97% (in the case of electronically generated aerosols) within the lung. In the absence of a chemistry model describing reactions amongst acids and bases, we made an assumption on the pH value of particles within the lung. Due to the high water fraction we assumed that the pH value of the particles is similar to that of water in equilibrium, i.e. 7. This value is significantly different from experimentally determined smoke pH (see e.g. Lauterbach, Bao, Joza, & Rickert, 2010). Note, however, that in these experiments water mass fractions are of the order of the reactive substances, while the present simulations suggest a water to reactive substance ratio of about 10 in lung generation 7 and already 25 in lung generation 10. Applying the chemical equilibrium constant given by Weast (1972) of 1.047 x 108 mol-1 results in roughly 9% of nicotine present in its volatile, unprotonated form.