Influence of suspended particles on indoor semi-volatile organic compounds emission

Atmospheric Environment 79 (2013) 695e704

Contents lists available at ScienceDirect

Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv

Influence of suspended particles on indoor semi-volatile organic compounds emission Kang Hu, Qun Chen*, Jun-Hong Hao Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Engineering Mechanics, Tsinghua University, Beijing 100084, China

h i g h l i g h t s
The adsorption of suspended particles is considered for SVOCs transport in the air.
SVOCs diffusion in particles can be described by the lumped parameter method.
A model of SVOCs transport in the air with suspended particles is established.
The gas-phase DEHP concentration increases rapidly in the first few seconds.
Increasing ACH can effectively reduce the particle-phase concentration of SVOCs.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 March 2013 Received in revised form 6 June 2013 Accepted 5 July 2013

Semi-volatile organic compounds (SVOCs) have been attracting more and more attentions to many researchers in these years. Because SVOCs have a strong tendency for adsorption to suspended particles, we take the effect of suspended particles into account to study the transport mechanism of SVOCs in the air. We establish a mathematical model to describe the transport mechanism of SVOCs, and study the transport processes of both gas- and particle-phase di-2-ethylhexyl phthalate (DEHP) in Field and Laboratory Emission Cells (FLECs). The predictions by the proposed model not only fit well with the experimental data of previous studies, but also show that the gas-phase DEHP concentration increases rapidly in the first few seconds and increases slowly during the following 200 days due to different transport mechanisms in the two periods. Meanwhile, when the particle radiuses are of the order of micron and the air changes per hour (ACH) is large enough, the characteristic time for DEHP getting gas/ particle equilibrium is much longer than the residence time of a particle in the flow field, and thus there is no significant influence of suspended particles on the total concentration of DEHP in the air. Oppositely, the influence of particles on DEHP emission will be enhanced for a cycling air flow system with a small ACH, where increasing ACH will reduce the concentrations of particle-phase SVOCs. Besides, if the particle radiuses are of the order of nanometer, decreasing the particle radiuses will shorter the characteristic time for DEHP getting gas/particle equilibrium, and finally increase the particle-phase concentration of DEHP. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Indoor air Semi-volatile organic compounds Suspended particle Transport mechanism Adsorption

1. Introduction A man keeps in contact with indoor air for about 90% time per day (Robinson and Nelson, 1995), and hence human health is straightly impacted by indoor air quality (IAQ) (Jones, 1999). The factors influencing IAQ mainly include volatile inorganic compounds, e.g. ammonia, carbon monoxide, and sulfur dioxide,

* Corresponding author. Tel./fax: þ86 10 62781610. E-mail address: [email protected] (Q. Chen). 1352-2310/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atmosenv.2013.07.010

volatile organic compounds (VOCs), e.g. formaldehyde, acetone, and aromatic hydrocarbon, semi-volatile organic compounds (SVOCs), e.g. di-2-ethylhexyl phthalate, pyrene, and polycyclic aromatic hydrocarbons, and inhalable particles (Brooks et al., 1991). Among the studies of these indoor pollutions, it is merely 20 years since researchers began to study on indoor SVOCs pollution (Loock et al., 1993). According to the World Health Organization (WHO)’s classification for indoor organic compounds (WHO, 1989), SVOCs feature high boiling points and low saturated vapor pressures. They originate from many kinds of sources such as polyvinylchloride (PVC)

696

K. Hu et al. / Atmospheric Environment 79 (2013) 695e704

plastic products with plasticizers and house furnishings with fireretardants, and will cause serious harm to human health. For example, SVOCs are able to get into human blood through lung or skin and lead to several kinds of diseases, especially asthma and allergic (Bornehag and Nanberg, 2010; Jaakkola et al., 2004; Lyche et al., 2009; Kolarik et al., 2008). They will also affect the reproductive development and neurological development of human beings (Lottrup et al., 2006; Swan, 2006; Larsson et al., 2009; Testa et al., 2012). Recent years, a considerable number of investigations on SVOCs pollution status have been brought out. Xie et al. (2005), Teil et al. (2006), Schnelle-Kreis et al. (2007), Bessagnet et al. (2010), and Wang et al. (2010) have reviewed the SVOCs pollution situation affected by various factors in different areas, such as the US, Canada, some countries in Europe, and China. Weschler (2009) reviewed the indoor pollutants since the 1950s, indicated that the pollution of phthalate esters has increased and remains high these years. On the other hand, some other researchers focused on the transport mechanism of SVOCs in the air. Xu and Little (2006) extended the model that predicts the emission rate of VOCs to predict that of SVOCs from polymeric materials, which showed reasonable agreement between the experimental data and the model predictions. Clausen et al. (2004, 2007) studied the emission of di-2ethylhexyl phthalate (DEHP) from a PVC floor in a Field and Laboratory Emission Cell (FLEC) and a Chamber for Laboratory Investigations of Materials, Pollution, and Air Quality (CLIMPAQ), and indicated that the emission rate of DEHP was limited by gas-phase mass transport. Xu et al. (2009, 2012) did more chamber studies to elucidate the transport mechanisms of SVOCs in the air. They focused on the influence of surface adsorption on the variation of gas-phase DEHP concentration, and found that the strong partitioning of DEHP onto the stainless steel surface follows a simple linear relationship. Moreover, in order to overcome the difficulty for researchers to measure and estimate the concentrations of SVOCs promptly and accurately in experiments, Clausen et al. (2010, 2012) employed the numerical simulation to interpret the experimental results of emission of DEHP in a FLEC, and indicated that the steady-state concentrations of DEHP in the FLEC increased greatly with the increasing ambient temperature, because the adsorption to the chamber walls decreased greatly. Besides gas-phase state, it is well-recognized now that SVOCs are quite likely to be adsorbed by particulate matters, which results in a significant effect on indoor SVOCs pollution. For the influence mechanism of suspended particles in the transport of SVOCs, Xu and Little (2006) examined the effect of SVOCs partitioning onto airborne particles, and found that airborne particles may play an important role in SVOCs transport and inhalation exposure. Weschler and Nazaroff (2008) indicated that SVOCs with low vapor pressures (or large octanoleair partition coefficients) can partition meaningfully between the gas phase and airborne particles. When the particles are smaller, their sorption to SVOC will be stronger, and the perniciousness will be more serious. Liu et al. (2012, 2013) studied the dynamic interaction between SVOCs and organic particles, and indicated that the instantaneous equilibrium assumption is not reasonable for the less volatile species such as DEHP. Benning et al. (2013) used the emission chambers designed by Xu et al. (2012) to study the influence of ultra-fine particles on the emission of DEHP, and indicated that the total (gas plus particle) DEHP concentrations increased by a factor of 3e8 when particles were introduced to the chamber at concentrations of 100e245 mg m3. The experiments in the above studies focus mainly on the particles that are in the steady state. In other words, the SVOCs concentration has reached gas/particle equilibrium already in these studies. However, few papers focused on the influence of suspended particles on the transient emission mechanism of SVOCs in

the air. In this paper, we presented a transport model of indoor SVOCs both in gas- and particle-phases, and studied SVOCs’ emission in a FLEC using numerical simulations. Finally, we analyzed the influence of multiple factors, including particle number density, particle size, and air change rate, on the transport of SVOCs. 2. Transport mechanism of indoor SVOC As discussed above, indoor airborne SVOCs exist in two phases: the gas-phase SVOCs dispersing in the air, and the particle-phase SVOCs deposited in suspended particles. Convective mass transfer occurs among the gas-phase SVOCs in the air, while particle-phase SVOCs diffuse in the suspended particles and move with them together. Meanwhile, mass transfer of SVOCs occurs between the particles and the surrounding air. In the subsequent, we will discuss the issues in the following order: the diffusion mechanism of particle-phase SVOCs, gas-phase SVOCs, the mass transfer between the particle- and the gas-phase of SVOCs, and the boundary conditions. 2.1. Diffusion mechanism of SVOC in particles In order to study the transport mechanism of SVOCs in the air with suspended particles, it is necessary to study the diffusion mechanism of particle-phase SVOCs in particles at first. Liu et al. (2013) have studied on the mechanism of different kinds of real particles adsorbing SVOCs. As the main concern of this paper is the transport mechanism of SVOCs in the air with suspended particles, we adopt the simplest particle model for simplicity, i.e. a uniformly porous, adsorbing sphere without non-adsorbing inner core, as shown in Fig. 1. In spherical coordinates, the conservation of a SVOC during diffusion gives

! vCp v2 Cp 2 vCp ¼ Dp : þ r vr vt vr 2

(1)

The initial conditions and the boundary conditions are

 Cp ðr; tÞt¼0 ¼ 0;

(2)

 vCp ðr; tÞ ¼ 0; vr r¼0

(3)

Fig. 1. The radial diffusion model in particles.

K. Hu et al. / Atmospheric Environment 79 (2013) 695e704

Dp

 vCp ðr; tÞ vr 

" ¼ h r¼Rp

 Cp ðr; tÞ

697

# r¼Rp

Kpart

 Cg ;

(4)

where Cg is the concentration of a gas-phase SVOC around a particle, Cp is the concentration of particle-phase SVOC in the particle, Dp is the molecular diffusivity of SVOC in the particle, Rp is the particle radius, Kpart is the partition coefficient between the particle- and the gas-phases of SVOC, and h is the mass transfer coefficient between gas and particle, which can be calculated by (Li and Davis, 1996):

h ¼

Dg 1 þ Kn ; Rp 1 þ 1:71Kn þ 1:333Kn2

(5)

where Dg is the molecular diffusivity of SVOC in the air, and Kn is the Knudsen number, determined as follows:

rffiffiffiffiffiffiffiffiffi 3Dg l 8RT ;c ¼ ; Kn ¼ ;l ¼ pM Rp c

Fig. 2. The relation of [C*p(r* ¼ 0, Fop)/C*p(r* ¼ 1, Fop)  1], Bip/Kpart and Fop.

(6)

where l is the mean free path of SVOC, c is the mean molecular speed of SVOC, R is the gas constant, T is the temperature, and M is the molecular weight of SVOC. By introducing such variables as dimensionless concentration, Biot number (dimensionless convective mass transfer coefficient), Fourier number (dimensionless time), and dimensionless radius:

Cp* ¼

Kpart Cg  Cp ; Kpart Cg

(7)

Bip ¼

Rp h ; Dp

(8)

Dp t Fop ¼ 2 ; Rp r* ¼

(9)

r ; Rp

(10)

vFop

¼

v2 Cp* vr *2

   Cp* r * ; Fop 

þ

  vCp* r * ; Fop    vr *

(11)

¼ 0;

(12)

¼ 0;

(17)

¼ r* ¼1

 Bip *  *  C r ; Fop  * : Kpart p r ¼1

Bip ; Fop ; r * : Kpart

where Cp(0) is the initial concentration of particle-phase SVOC in the particle. If Cp(0) ¼ 0, Eq. (17) is simplified as

 Cp ¼ Kpart Cg 1  exp 

3h t Rp Kpart

 :

(18)

(13)

(14)

s¼

It is clearly from Eqs. (11)e(14) that Cp* is a function of Bip/Kpart, Fop and r*:

Cp* ¼ f

where Vp is the volume of a single particle, Ap is the surface area of a single particle. If Cg is constant, the solution of Eq. (16) is

A characteristic time s, as shown in Eq. (19), is defined to describe the time scale that the gas- and particle-phase SVOCs approach equilibrium, i.e. the proportion (Cp  KpartCg)/ (Cp(0)  KpartCg) reaches exp(1) ¼ 36.8%.

r * ¼0

  vCp* r * ; Fop     vr *

(16)

*

2 vCp ; r * vr *

Fop ¼0

 dCp Ap h Cp ¼ Cg  ; dt Kpart Vp

  Cp ðtÞ  Kpart Cg Ap h 3h ¼ exp  t ¼ exp  t ; Rp Kpart Cp ð0Þ  Kpart Cg Vp Kpart

Eqs. (1)e(4) are simplified as

vCp*

In heat transfer, the lumped parameter method is often used to simplify the model when Bi < 0.1, in which the max excess temperature is less than 5%. In mass transfer, we can also take a similar simplification. From Fig. 2, we can find that the max excess concentration, i.e. [C*p(r* ¼ 0, Fop)/C*p(r* ¼ 1, Fop)  1], is less than 5% when Bip/Kpart < 0.1. Table 1 lists the physical properties (the molecular weights, the molecular diffusivities of the gas- and particlephases of SVOCs, and the partition coefficients between them) of some typical indoor SVOCs. For common indoor suspended particles, the diameter of which is usually less than 100 mm (Wang et al., 2010), the values of Bip/Kpart are all less than 0.1. Therefore, the concentrations of SVOCs, Cp, can be assumed uniform in particles by the lumped parameter method. That is, it is only determined by time, expressed as:



RKpart : 3h

Table 1 Physical properties of some typical indoor SVOCs (Barring et al., 2002; Strommen and Kamens, 1999). Name

M (g mol1)

Dg  106 (m2 s1)

Dp  1011 (m2 s1)

Kpart

DEHP Pyrene BDE-47 BDE-99

391 202 486 565

3.9 4.71 3.88 3.65

2.13 0.078 2.24 2.40

2.3 9.55 1.58 3.16

(15)

Solving Eq. (11) with the given initial and boundary conditions, Eqs. (12)e(14), yields the relation of [C*p(r* ¼ 0, Fop)/ C*p(r* ¼ 1,Fop)  1], Bip/Kpart and Fop, as shown in Fig. 2.

(19)

s (for PM2.5)    

1011 108 1010 1011

9h 2 min 38 min 13 h

698

K. Hu et al. / Atmospheric Environment 79 (2013) 695e704

Table 1 also gives the characteristic times for different SVOCs when the radius of the particle equals 2.5 mm. For these SVOCs, it takes several minutes, even several hours, to reach equilibrium, so the process is not instantaneous. Therefore, the mass transfer rate between gas- and particle-phases of SVOCs is limited, and should be considered as



_ ¼ hNp Ap Cg  m

Cp Kpart

;

(20)

_ represents the mass where Np is the particle number density, and m transfer rate between gas- and particle-phases of SVOCs per unit _ means that gas-phase volume of air. Here, the positive value of m SVOCs are translating into particle-phase, while the negative value is in contrast. 2.2. Transport mechanism of SVOC in the air For convective mass transfer of a SVOC in the air, the continuity equation of the air is

V$U ¼ 0;

(21)

the momentum equation is

vU 1 þ ðU$VÞU ¼  Vp þ nV2 U; r vt

(22)

and the convective mass transfer of the gas-phase SVOC is governed by

vCg _ þ ðU$VÞCg ¼ Dg V2 Cg  m: vt

(23)

where U is the velocity, r is the density, p is the pressure, and n is the kinematic viscosity. Meanwhile, the particle-phase SVOC is moving with suspended particles. Assuming that the local particle velocity, i.e. the local particle-phase SVOCs velocity, equals to the local air velocity, the mass transfer rate of the particle-phase SVOC is determined by

Fig. 3. The photo and the cross sectional view of a FLEC (Zhang and Niu, 2003).

K. Hu et al. / Atmospheric Environment 79 (2013) 695e704 Table 2 The geometrical structures of the FLEC (Clausen et al., 2004). Diameter Height Area of the entrance Area of the export

150 mm 18 mm 4.71  104 m2 6.36  105 m2

Internal surface area Area of test piece Volume Internal surface area/volume

    v Np Vp Cp _ þ ðU$VÞ Np Vp Cp ¼ m: vt

F ¼  lim

Dt/0

0.0179 m2 0.0177 m2 35 ml 505 m2 m3

(24)

Due to the fact that the particle-phase SVOC is unable to diffuse directly between each two separate particles by itself, there is no diffusion term in Eq. (24).

2.3. Boundary conditions For most boundary surfaces, it can be modeled that a reversible equilibrium of SVOC’s concentration exists between the exposed interior material surface and the air immediately adjacent to the surface, which can be expressed as:

Cg;surf ¼

Csurf ; Ksurf

(25)

where Cg,surf is the gas-phase concentration of SVOC adjacent to the surface, Csurf is the interior surface concentration of SVOC, and Ksurf is the partition coefficient of SVOC between the gas-phase and the interior surface concentration of SVOC, depending on the kind of SVOC, the material of boundary surface, and the environment temperature. When Cg,surf varies, the surface will adsorb or desorb a certain amount of SVOC to maintain the balance due to the reversible equilibrium, which can be expressed as (Clausen et al., 2010):

699

Csurf ðt þ DtÞ  Csurf ðtÞ Dt

Ksurf Cg;surf ðt þ DtÞ  Ksurf Cg;surf ðtÞ ; Dt Dt/0  dCg;surf vCg  ¼ Ksurf ¼ Ksurf dt vt surf ¼  lim

(26)

where F represents the mass flux of gas-phase SVOC through the boundary surface. Here, the positive value of F means that the SVOC are adsorbed by the surface, while the negative value is in contrast. In practice, the SVOC has a strong tendency for adsorption to most material surfaces. The magnitude of the SVOCs adsorbed or desorbed is as great as that of the gas-phase SVOCs adjacent to the surfaces, so that it should be considered. In this case, the boundary condition should be

  vCg  vCg  Dg ¼ F ¼ Ksurf ; vn surf vt surf

(27)

where n is the normal direction of the boundary surface. For some organic materials (such as PVC flooring), their Ksurf is much greater than that of other materials (such as stainless steel surface). In the former case, the amount of SVOCs adsorbed/desorbed by the material surface in a short time is far less than that stored in the material, so that the concentration of SVOC in the material can be assumed unchanged. In this case, the boundary condition is simplified to the Dirichlet boundary condition, expressed as

Cg;surf ¼ const:

(28)

For particle-phase SVOCs, the deposition of the suspended particles from the air to the surface is ignored, so that there is no mass flow of particle-phase SVOCs through the surfaces. The boundary condition is expressed as

Fig. 4. The simplified model of the FLEC used for simulations.

700

 vCp  ¼ 0: vn surf

K. Hu et al. / Atmospheric Environment 79 (2013) 695e704

(29)

Solving the Eqs. (21)e(24) together with the given initial and boundary conditions shown above will give the velocity field of the air and the concentration distributions of the gas- and the particlephase SVOCs in the air. 3. DEHP emission in FLECs

stainless steel that is positioned upon the test piece, as shown in Fig. 3. The geometrical structures of the FLEC are listed in Table 2. Because of the symmetry of the flow field of FLEC, it can be simplified to an axisymmetric field during mesh construction, as shown in Fig. 4. The axis of rotation is on the left side, the inlet edge of the fresh air is on the right side, the outlet edge of the air is at the top, and a vinyl flooring as a SVOCs pollution source is at the bottom, where the boundary condition given by Eq. (28) was adopted. The upper edge is the stainless steel surface, considering the boundary condition given by Eq. (27).

3.1. Equipment structure 3.2. Transport in the air without suspended particle FLEC is a kind of widely used experimental instrument employed for studying the emission of indoor air pollutants (Wolkoff, 1991, 1996). It is a micro-emission cell made of polished

Fig. 5. Comparison of numerical results of the relative concentrations at the FLEC outlet with experimental results (a) Ksurf ¼ 7500 m (Clausen et al., 2004), (b) Ksurf ¼ 1000 m (Clausen et al., 2007), (c) Ksurf ¼ 100 m (Clausen et al., 2012). (Qin ¼ 450 ml min1).

In order to test and verify the model and the numerical simulation method, we consider an SVOC transport process without suspended particle at first. Using the experimental data of Clausen et al. (2004, 2007, 2012) as references, the flooring at the bottom of the FLEC is assumed to contain DEHP only, and Cg,surf is set to 0.8 mg m3. The fresh air inflowing is assumed to be free of DEHP. For the initial conditions, the concentration of DEHP is assumed to be zero in the whole field at the beginning. Figs. 5(a)e(c) give the comparison of numerical and experimental results for different Ksurf where the vertical axes are the relative concentration of SVOCs at the outlet of the FLEC, defined as the concentration of SVOCs normalized by Cg,surf (the same below). The difference between the Ksurf in each situation is mainly caused by the temperature difference of the environments. As is shown, the results of the simulations are close to those from the experiments, which shows that the mass transfer model and the solution method in this paper are reliable. According to the results above, we can find a phenomenon that the outlet DEHP concentration increases rapidly in the first few seconds. (A similar phenomenon has been found in Xu et al. (2012)’s and Liu et al. (2013)’s studies.) In order to give a further analysis (taking the phenomenon in Fig. 5(a) for example), Fig. 6 gives the outlet DEHP concentration variation and the emission rate of the vinyl flooring in the first 5 s. In addition, Fig. 7 shows the velocity distribution and the concentration distributions of DEHP at nine different times. As shown in Figs. 6 and 7, the concentration of DEHP at the outlet of the FLEC increases rapidly in the first 2 s, during which the DEHP near the pollution source surface spreads to the main flow field of the FLEC rapidly. In this period, the concentration does not show a significant growth near the upper surface because of the adsorption of the stainless steel surface. Then, during the following 200 days, the concentration of DEHP at the outlet increases slowly,

Fig. 6. The relative concentration variation at the outlet of the FLEC and the emission rate at the vinyl flooring versus time during the first 5 s, gained from the numerical simulation.

K. Hu et al. / Atmospheric Environment 79 (2013) 695e704

701

Fig. 7. The numerical results: velocity field (a), relative concentration fields of DEHP at (b) 0.5 s, (c) 1.0 s, (d) 1.5 s, (e) 2.0 s, (f) 100.0 s, (g) 1 day, (h) 10 days, (i) 100 days, (j) 200 days.

during which the concentration near the stainless steel surface increases under the constraints between the convective mass transfer of DEHP itself and the adsorption of the upper surface, reaching a steady-state value at last. Therefore, the increasing process of DEHP concentration is divided into two main stages: (1) the rapid increasing process, depending mainly on the emission of DEHP from the pollution source and the convective mass transfer of DEHP itself, (2) the slow increasing process, depending mainly on the constraints between the convective mass transfer of DEHP and the adsorption of the stainless steel surface. Fig. 8 gives the outlet DEHP concentrations variation and the emission rates of the vinyl flooring at different air flow rates. As shown, the air flow rate influences the increasing of the concentration of DEHP. A higher air flow rate leads to a rapider increasing of concentration in the first 2 s, a decline of steady-state concentration, and a higher emission rate of DEHP. It is because a higher air flow rate strengthens the convective mass transfer of DEHP that the emission rate becomes higher and the concentration increases rapider at the beginning. Meanwhile, a higher air flow rate reduces the period of time that a certain mass of air flowing through the FLEC, declines the increment of its containing DEHP during the period, and finally lowers the steady-state concentration as a macroscopic phenomenon. In addition, Fig. 9 gives the outlet DEHP concentration variation in the first 6 s where the scale of the FLEC is doubled or halved, and the air inflow velocity is maintained. It is shown that the scale of the FLEC has an effect on the increase of the concentration of DEHP

as well. A larger size of the FLEC leads to a slower speed of the increase of the concentration at the beginning of the process, which is mainly because of the longer time that DEHP spreading to the whole field from the pollution source in this case. 3.3. Transport in the air with suspended particles For the case of the air with suspended particles present, the particle number density Np is assumed constant in the whole flow field for simplicity. Besides, the DEHP concentration of the inflowing air is assumed to be zero. When the air flow rate Qin is 450 ml min1, Fig. 10 gives the total concentration variation of DEHP at the outlet of the FLEC for different particle number densities and different diameter of particles. The vertical axis is the sum of relative concentration of the gas- and the particle-phase DEHP, normalized by Cg,surf. It is needed to mention that the particle-phase concentration of DEHP is different from the true concentration, Cp, of DEHP in particles. The former is the total quantity of DEHP per unit volume of air, expressed as

Cpphase ¼ Np Vp Cp ¼

TSP

rpart

Cp ;

(30)

where TSP is total suspended particulate in the flow domain, rpart is the density of particles, assumed to be 2500 kg m3. Fig. 10 shows that the total concentration of gas- and particlephase SVOC is not straightly impacted by the particle number

702

K. Hu et al. / Atmospheric Environment 79 (2013) 695e704

Fig. 8. The relative concentrations at the outlet of the FLEC and the emission rates at the vinyl flooring with different air flow rates.

density and the diameter of particles. Fig. 11 gives the particlephase concentration variation of DEHP. It is shown that increasing the particle number density and decreasing of the diameter of particles both lead to a higher concentration of particle-phase SVOCs. However, the magnitude of the particle-phase SVOCs is much less than that of gas-phase SVOCs, so that we cannot find a significant influence of the factors in Fig. 10. In the above cases, the mass flow rate of air, Qin, is 450 ml min1, i.e. the air changes per hour1 (ACH) is 771 h1, which means that the average residence time for particles in the FLEC is only 4.67 s. Nevertheless, the time scales that the concentration of gas- and particle-phase SVOCs reaching equilibrium (discussed in 2.1) is much longer than 4.67 s, so the particle-phase concentration of SVOCs only increases by 6.7  104 in such short time. Meanwhile, for the three cases shown in Fig. 11, smaller particle has a higher mass transfer coefficient h (shown by Eqs. (5) and (6)) and a shorter characteristic time s (calculated by Eq. (19)), which means that the particle will adsorb more DEHP during the residence time. Besides, higher TSP means more particles involved in the adsorption, so the concentration of particle-phase DEHP will be higher in this case. After a sufficiently long time (about 200 days), the concentration of gas-phase DEHP is approximated as Cg,surf in the whole field. In this case, for an arbitrary particle, substituting Eq. (18) into Eq. (30) yields the particle-phase concentration of DEHP at the outlet, expressed as

1 Air changes per hour is the ratio of the mass flow rate of fresh air by the volume of a room.

Cpphase ¼

TSP

rpart

 Kpart Cg;surf 1  exp 

3h Dt Rp Kpart

 :

(31)

where Dt is the average residence time for particles in the FLEC. Solving Eq. (31) gives the theoretical solutions of the particlephase concentrations of DEHP at the outlet for the three situations shown in Fig. 11, 6.54  104, 1.31  103 and 4.26  105, which are extremely closed to the numerical results. Therefore, numerical and theoretical analysis both indicate that the existence of particles does increase the concentration of DEHP, but the influence is not significant because of such a short time that a particle goes through the FLEC.

Fig. 9. The relative concentrations at the outlet of the FLEC for different scales of FLEC.

K. Hu et al. / Atmospheric Environment 79 (2013) 695e704

703

Table 3 The comparison between experimental and predicted results for Benning et al. (2013)’s experiment. Case Flow rate, Residence Effluent Average Experimental Predicted Q (mL min1) Time TSP Cg-surf Cp-phase Cp-phaseb Dta (min) (mg m3) (mg m3) (mg m3) (mg m3)

Fig. 10. The total relative concentrations at the outlet of the FLEC for different particle number densities and different diameter of particles. (Qin ¼ 450 ml min1).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 a b

110 123 201 195 317 776 860 800 675 1870 1390 1840 2610 3420 3140 2330 3380 4200

18.2 16.3 9.95 10.3 6.31 2.58 2.33 2.50 2.96 1.07 1.44 1.09 0.766 0.585 0.637 0.858 0.592 0.476

150 150 150 150 134 195 235 197 199 100 131 154 177 111 114 121 131 148

0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65 0.65

2.9 3.0 3.4 3.2 4.8 4.6 4.5 3.9 3.6 2.4 3.1 3.4 3.3 2.5 2.1 2.3 1.4 1.3

3.12 3.12 3.12 3.12 2.79 4.00 4.79 4.03 4.11 1.73 2.48 2.68 2.66 1.44 1.55 1.91 1.71 1.69

Dt ¼ Vchamber/Q, where Vchamber ¼ 2 L. Estimated by Eq. (31) with the data from Benning et al. (2013): Kpart/

rpart ¼ 0.032 m3 mg1, s ¼ RpKpart/3h ¼ 0.6 min.

Fig. 11. The particle-phase concentration variation of DEHP for different particle number densities and different diameter of particles.

In reality, ACH of an indoor environment may be much less than the cases above, so that the influence of particles on the emission of SVOC becomes significant. For an airtight, air-conditioned room, the air conditioner will combine the circulate air from the room with a proper ratio of fresh air, then blow it back to the room, and the ACH is usually between the order of magnitude of 0.1 h1 w 1 h1. In order to study the cases more similar to this kind of rooms by using the model of FLEC while keeping the same air flow rate, Fig. 12 gives the total (gas plus particle) DEHP concentration variation with three different ACHs, 7.71 h1, 3.86 h1 and

0.77 h1. In these cases, the DEHP concentration of the inflowing air, including gas- and particle-phases, were set to 99.0%/99.5%/ 99.9% that of the outflowing air. As shown in Fig. 12, the concentrations of the particle-phase SVOCs will reach 56.8%/12.8%/6.5% of the gas-phase SVOCs after 500 days. Meanwhile, solving Eqs. (17) and (30) theoretically gives the concentrations of the particle-phase DEHP, Cp-phase, at the outlet for the three situations above, 57.4%, 12.7% and 6.5%, which are extremely closed to the numerical results. From both numerical and theoretical results, it is clear that for a constant air flow rate, a higher ACH means more fresh air and less circulate air blowing into the equipment, and thus lowers the concentration of particle-phase DEHP. Therefore, increasing ACH has a much more significant effect on reducing the particle-phase SVOCs than on reducing the gasphase SVOCs. Another factor that will enhance the influence of particles on DEHP emission is a shorter characteristic time of particles. Just like the experiments of Benning et al. (2013), 45 nm average diameter ammonium sulfate particles were used, for which the characteristic time s is equal to 0.6 min. Meanwhile, the average residence time for them in the chamber, which is in the order of magnitude of minutes, is much longer than s. In this case, a particle will adsorb more DEHP during the residence time, so that the influence of particles becomes significant. Additionally, a comparison between Benning et al. (2013)’s experimental results and the predicted results by Eq. (31) is shown in Table 3. The predictions are extremely close to the experimental results, which show that the model of the particle-concerned system in this paper is reliable. 4. Conclusions

Fig. 12. The total relative concentrations at the outlet of the FLEC for different ACHs.

The adsorption of suspended particles is taken into account to study the transport mechanism of SVOCs in the air. By analyzing the SVOCs diffusion processes in particles with the lumped parameter method, a mathematical model is established to describe the transport process of SVOCs in the air with suspended particles. Comparisons between numerical predictions and experimental results show the reliability of the newly proposed mathematical model and the corresponding numerical method.

704

K. Hu et al. / Atmospheric Environment 79 (2013) 695e704

For the situation without suspended particles, the outlet DEHP concentration increases rapidly in the first few seconds, but then increases slowly during the following 200 days. It is because of the different transport mechanisms in the above two periods: in the former period, it depends mainly on the emission of DEHP from the pollution source and the convective mass transfer of DEHP itself, while in the latter period, it depends mainly on the constraints between the convective mass transfer of DEHP and the adsorption of the stainless steel surface. Moreover, analysis of the influence of different air flow rates and different FLEC sizes on the mass transfer of gas-phase DEHP shows that (1) a higher air flow rate leads to a rapider increasing of the concentration in the first 2 s, a decline of the steady-state concentration, and a higher emission rate of DEHP; (2) a larger size of the FLEC leads to a slower speed of the concentration increment at the beginning of the process. For the situation with suspended particles, if the particle radiuses are of the order of microns and the ACH is large enough, the residence time of a particle is much shorter than the characteristic time for DEHP to get gas/particle equilibrium, and thus particles may not sharpen the SVOC pollution significantly. However, for a cycling system with lower ACH similar to a real room, where the equivalent residence time of particles becomes longer, the influence of particles on SVOC pollution becomes significant. A lower ACH leads to a higher particle-phase SVOC. Therefore, increasing ACH of a room is an effective way to reduce the concentrations of particle-phase SVOCs. On the other hand, if the particles are smaller, i.e. the characteristic time is shorter, SVOCs will approach gas/particle equilibrium more easily in the air, and the influence of particles on SVOC pollution will become more significant as well. Acknowledgments The present work is supported by the National Natural Science Foundation of China (Grant No. 51136002). References Barring, H., Bucheli, T.D., Broman, D., Gustasson, O., 2002. Soot-water distribution coefficients for polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans and polybrominated diphenylethers determined with the soot cosolvency-column method. Chemosphere 49, 515e523. Benning, J.L., Liu, Z., Tiwari, A., Little, J.C., Marr, L.C., 2013. Characterizing gas-particle interactions of phthalate plasticizer emitted from vinyl flooring. Environmental Science & Technology 47, 2696e2703. Bessagnet, B., Seigneur, C., Menut, L., 2010. Impact of dry deposition of semi-volatile organic compounds on secondary organic aerosols. Atmospheric Environment 44, 1781e1787. Bornehag, C.G., Nanberg, E., 2010. Phthalate exposure and asthma in children. International Journal of Andrology 33, 333e345. Brooks, B.O., Utter, G.M., Debroy, J.A., Schimke, R.D., 1991. Indoor air pollution: an edifice complex. Journal of Toxicology e Clinical Toxicology 29, 315e374. Clausen, P.A., Hansen, V., Gunnarsen, L., Afshari, A., Wolkoff, P., 2004. Emission of di2-ethylhexyl phthalate from PVC flooring into air and uptake in dust: emission and sorption experiments in FLEC and CLIMPAQ. Environmental Science & Technology 38, 2531e2537. Clausen, P.A., Liu, Z., Kofoed-Sorensen, V., Little, J., Wolkoff, P., 2012. Influence of temperature on the emission of di-(2-ethylhexyl)phthalate (DEHP) from PVC flooring in the emission cell FLEC. Environmental Science & Technology 46, 909e915. Clausen, P.A., Liu, Z., Xu, Y., Kofoed-Sorensen, V., Little, J.C., 2010. Influence of air flow rate on emission of DEHP from vinyl flooring in the emission cell FLEC: measurements and CFD simulation. Atmospheric Environment 44, 2760e2766. Clausen, P.A., Xu, Y., Kofoed-Sorensen, V., Little, J.C., Wolkoff, P., 2007. The influence of humidity on the emission of di-(2-ethylhexyl) phthalate (DEHP) from vinyl flooring in the emission cell “FLEC”. Atmospheric Environment 41, 3217e3224.

Jaakkola, J.J.K., Parise, H., Kislitsin, V., Lebedeva, N.I., Spengler, J.D., 2004. Asthma, wheezing, and allergies in Russian schoolchildren in relation to new surface materials in the home. American Journal of Public Health 94, 560e562. Jones, A.P., 1999. Indoor air quality and health. Atmospheric Environment 33, 4535e 4564. Kolarik, B., Naydenov, K., Larsson, M., Bornehag, C.G., Sundell, J., 2008. The association between phthalates in dust and allergic diseases among Bulgarian children. Environmental Health Perspectives 116, 98e103. Larsson, M., Weiss, B., Janson, S., Sundell, J., Bornehag, C.-G., 2009. Associations between indoor environmental factors and parental-reported autistic spectrum disorders in children 6-8 years of age. Neurotoxicology 30, 822e831. Li, W.G., Davis, E.J., 1996. Aerosol evaporation in the transition regime. Aerosol Science and Technology 25, 11e21. Liu, C., Morrison, G.C., Zhang, Y., 2012. Role of aerosols in enhancing SVOC flux between air and indoor surfaces and its influence on exposure. Atmospheric Environment 55, 347e356. Liu, C., Shi, S., Weschler, C., Zhao, B., Zhang, Y., 2013. Analysis of the dynamic interaction between SVOCs and airborne particles. Aerosol Science and Technology 47, 125e136. Loock, F., Lampe, T., Bahadir, M., 1993. Examination of the emission of (semi-) volatile organic compounds from polymer materials used as interior trim materials in automobiles. Fresenius’ Journal of Analytical Chemistry 347, 280e285. Lottrup, G., Andersson, A.M., Leffers, H., Mortensen, G.K., Toppari, J., Skakkebaeek, N.E., Main, K.M., 2006. Possible impact of phthalates on infant reproductive health. International Journal of Andrology 29, 172e180. Lyche, J.L., Gutleb, A.C., Bergman, A., Eriksen, G.S., Murk, A.J., Ropstad, E., Saunders, M., Skaare, J.U., 2009. Reproductive and developmental toxicity of phthalates. Journal of Toxicology and Environmental Health Part B: Critical Reviews 12, 225e249. Robinson, J., Nelson, W., 1995. National Human Activity Pattern Survey Data Base. USEPA, Research Triangle Park, NC. Schnelle-Kreis, J., Sklorz, M., Orasche, J., Stolzel, M., Peters, A., Zimmermann, R., 2007. Semi volatile organic compounds in ambient PM(2.5). Seasonal trends and daily resolved source contributions. Environmental Science & Technology 41, 3821e3828. Strommen, M.R., Kamens, R.M., 1999. Simulation of semivolatile organic compound microtransport at different time scales in airborne diesel soot particles. Environmental Science & Technology 33, 1738e1746. Swan, S.H., 2006. Prenatal phthalate exposure and anogenital distance in male infants. Environmental Health Perspectives 114, A88eA89. Teil, M.J., Blanchard, M., Chevreuil, M., 2006. Atmospheric fate of phthalate esters in an urban area (Paris-France). Science of the Total Environment 354, 212e223. Testa, C., Nuti, F., Hayek, J., De Felice, C., Chelli, M., Rovero, P., Latini, G., Papini, A.M., 2012. Di-(2-ethylhexyl) phthalate and autism spectrum disorders. Asn Neuro 4, 223e229. Wang, L., Zhao, B., Liu, C., Lin, H., Yang, X., Zhang, Y., 2010. Indoor SVOC pollution in China: a review. Chinese Science Bulletin 55, 1469e1478. Weschler, C.J., 2009. Changes in indoor pollutants since the 1950s. Atmospheric Environment 43, 153e169. Weschler, C.J., Nazaroff, W.W., 2008. Semivolatile organic compounds in indoor environments. Atmospheric Environment 42, 9018e9040. WHO, 1989. Indoor Air Quality: Organic Pollutants. Report on a WHO Meeting. In: EURO Reports and Studies, pp. 1e70. Wolkoff, P., 1991. Field and Laboratory Emission Cell: FLEC, pp. 160e165. IAQ 91 Healthy Buildings. Wolkoff, P., 1996. An emission cell for measurement of volatile organic compounds emitted from building materials for indoor use e the field and laboratory emission cell FLEC. Gefahrstoffe Reinhaltung Der Luft 56, 151e157. Xie, Z.Y., Ebinghaus, R., Temme, C., Caba, A., Ruck, W., 2005. Atmospheric concentrations and air-sea exchanges of phthalates in the North Sea (German Bight). Atmospheric Environment 39, 3209e3219. Xu, Y., Hubal, E.A.C., Clausen, P.A., Little, J.C., 2009. Predicting residential exposure to phthalate plasticizer emitted from vinyl flooring: a mechanistic analysis. Environmental Science & Technology 43, 2374e2380. Xu, Y., Little, J.C., 2006. Predicting emissions of SVOCs from polymeric materials and their interaction with airborne particles. Environmental Science & Technology 40, 456e461. Xu, Y., Liu, Z., Park, J., Clausen, P.A., Benning, J.L., Little, J.C., 2012. Measuring and predicting the emission rate of phthalate plasticizer from vinyl flooring in a specially-designed chamber. Environmental Science & Technology 46, 12534e 12541. Zhang, L.Z., Niu, J.L., 2003. Laminar fluid flow and mass transfer in a standard field and laboratory emission cell. International Journal of Heat and Mass Transfer 46, 91e100.