Experiment and Numerical Investigation of Inhalable Particles...
Experiment and Numerical Investigation of Inhalable Particles...
Experiment and Numerical Investigation of Inhalable Particles...
a r t i c l e i n f o a b s t r a c t
Article history: After the outbreak of COVID-19, the indoor environment has become particularly important in closed
Received 24 April 2022 spaces, being a common concern in environmental science and public health, and of great significance
Revised 25 June 2022 for the building environment. To improve the indoor air quality and control the spread of viruses, the
Accepted 8 July 2022
analysis of inhalable particles in indoor environments is critical. In this research, we study standards
Available online 15 July 2022
focused on inhalable particles and indoor environmental quality, as well as analyzing the movement
and diffusion of indoor particles. Based on our analysis, we conduct an experimental study to determine
Keywords:
the distribution of indoor inhalable particles of different sizes before and after diffusion under the con-
Inhalable particles
Aerosol transmission
ditions of underfloor air distribution. Furthermore, the mathematical modeling method is adopted to
Particle experiment simulate the indoor flow field, particle trajectories, and pollutant dispersion process. The k-e two-
Mathematical modeling method equation model is applied as the turbulence model in the numerical simulation, while the Lagrangian dis-
Underfloor air distribution (UFAD) system crete phase model is adopted to trace the motion of particles and analyze the distribution characteristics
Building environment of indoor particles. The results demonstrate that fine particles (i.e., those with size less than 0.5 lm) have
a significant impact on the indoor particle concentration, while coarse particles (i.e., with size above
2.5 lm) have a greater influence on the total mass concentration of indoor particles. Small-sized particles
can easily follow the airflow and diffuse to upper parts of the room. Overall, the effects of indoor particles
on indoor air quality, including the potential threat of aerosol transmission of respiratory infectious dis-
eases, are non-negligible. Application of the presented research can contribute to improving the health-
related aspects of the building environment.
Ó 2022 Elsevier B.V. All rights reserved.
https://doi.org/10.1016/j.enbuild.2022.112309
0378-7788/Ó 2022 Elsevier B.V. All rights reserved.
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
ing, have obvious advantages in terms of energy saving and air work are summarized in Section 5. Our conclusions are drawn
quality improvement [4]. in Section 6.
Study of the distribution and movement characteristics of
indoor particles is of great significance, allowing us to improve
2. Theory and research development
the indoor air quality and promote the sustainable and healthy
development of indoor environments. To date, scholars have car-
2.1. Transmission of aerosol particles
ried out much research in this area, but there still remain areas
to be explored. First, almost no complete or detailed theoretical
The classification of particulate matter (see Fig. 1) mainly
system and method comparison of indoor particles has been pre-
includes: Dust (particle size>30 lm), which can settle quite
sented in the existing literature, and experimental investigations
quickly from the air; thoracic particles (PM10, GB/3095-1996),
and relevant research results are also very limited. Second, exper-
which refer to suspended particulate matter that can be inhaled
imental studies on the diffusion of indoor particles have mainly
through the mouth and nose; coarse fraction (aerodynamic diam-
focused on pollution sources at different time points [5–8], while
eter 2.5–10 lm), fine particles (PM2.5, 2.5 lm), which can pene-
few have considered the movement and diffusion of particles
trate into the lungs; and ultrafine particles (UFPs, 0.1 lm).
under a specific indoor flow field, especially with respect to the
The description of particulate matter in Fig. 1 above is from
air supply outlet of underfloor air distribution system, which is
research in Science (2005) [9], which reflects the particle size rang-
set on the floor and can easily cause an increase of particle concen-
ing from gas molecules, viruses, and bacteria to pollen and hair.
tration when the air supply flow is directly sent into rooms from
According to the comparison of particles with different sizes,
below. Furthermore, few studies have combined experimental
PM10, PM2.5, PM0.1, and so on, reflect particle clusters in the air.
analysis with numerical simulation and explored the two methods
Researchers from Tongji University have added the parts to which
together, in order to extend the research findings. Therefore, there
particles with different sizes can reach in the human body (e.g.,
exists an urgent need to research the distribution characteristics of
pharynx, bronchus, and lung lobe). The toxicity of particles is clo-
indoor inhalable particles, as well as the potential threat to the
sely related to their morphology, particle size, composition, and
indoor environment posed by the air supply mode, the building
properties. The surfaces of particles can adsorb various harmful
design, and the indoor air quality.
pollutants in the air, allowing these toxic substances to more
Hence, in this paper, we consider the background of the
quickly react and dissolve in the human body.
increased demand for indoor environment quality and building
Aerosol particles with an aerodynamic diameter of less than
ventilation improvement due to the COVID-19 pandemic outbreak.
10 lm are called inhalable particles [10]. In recent years, many
Based on the premise of the UFAD system, we conduct measure-
scholars have found that inhalable particles, such as aerosol parti-
ment experiments on the indoor particle concentration, as well
cles, which are rich in a large number of bacteria and viruses, can
as using the FLUENT software to carry out numerical simulation
cause various health problems, such as respiratory and nervous
of the indoor particle distribution and movement trajectories, the
diseases, where the incidence rate of carcinogenesis and deformi-
results of which we compare with the experimental results. One
ties may be increased [11–13].
of the main purposes of this study is to analyze the characteristics
The viruses are special particles, which may also be called bio-
and motion trajectory of indoor particles, in order to determine
logical aerosols. When patients cough or sneeze, they release a
suitable ways in which we can improve indoor air quality. Another
large number of viruses. The virus spreads through aerosols and
of our aims is determining whether underfloor air supply systems
droplets. Aerosol transmission refers to droplets that are mixed
can more efficiently settle or discharge particles to the outdoors,
in the air to form aerosols, leading to infection after subsequent
compared with traditional air supply modes, in order to create a
inhalation. The main transmission routes of COVID-19 are respira-
cleaner and safer indoor environment, thereby reducing the infec-
tory droplets and contact transmission, while aerosol and fecal
tion rate of viruses such as COVID-19. Furthermore, a large number
transmission routes remain to be further clarified. (Fig. 2)
of pathogenic bacteria, viruses, and other micro-organisms will
The World Health Organization has also pointed out that aero-
breed without air circulation, and people who stay indoors will
sols usually persist for a long time in a closed environment, while,
also release waste gas and carbon dioxide, causing the indoor air
in a natural environment, under the action of wind and water,
quality to decline and potentially become harmful to human
aerosol particles easily to combine with other media, accelerating
health. Thus, in this study, we also consider how to alleviate the
the degradation of the virus nucleic acid. According to the preven-
harm of aerosols in building environments through creating a
tion and treatment of influenza virus, if indoor ventilation is main-
hygienic, healthy, and comfortable green indoor air environment.
tained, the possibility of aerosol infection can be greatly reduced;
This paper is organized as follows. Section 1 presents the
hence, more ventilation and maintaining the circulation of indoor
introduction. Section 2 reviews the theoretical background, and
air are conducive to reducing aerosol infection.
then discusses the distribution and motion mechanism of indoor
Thus, it is of great significance to study the distribution charac-
particles. The experimental study, the mathematical modeling
teristics and movement of indoor inhalable particles, in order to
method, and how they are combined to contribute to the research
determine more effective methods for the control of indoor parti-
findings are clarified in Section 3. Section 4 details the experi-
cles and to improve building environments.
mental study, in which the distribution and diffusion characteris-
tics of indoor inhalable particles and their impact on indoor air
quality are analyzed. An experiment on the unsteady diffusion 2.2. Distribution characteristics of particles
process of mosquito-repellent incense is carried out using aerosol
spectrometer monitoring before and after implementation of the Many scholars have focused on the diffusion characteristics of
air supply. Next, the results of the mathematical methods are pre- particles. Fine [14] has studied the characteristics of particles pro-
sented, where the realizable two-equation model and discrete duced by candle combustion, and pointed out the type of particles
phase model are selected to simulate and track the indoor flow depended on the combustion mode of the candle. Lee et al. [15]
field and particle trajectory. Finally, a comparison between the have studied the impact of different cooking methods on the
experimental and numerical simulation results is presented, indoor environment, especially on PM2.5 and PM10 concentration
which are checked to verify the rationality and correctness of and particle distribution. Wang et al. [16] have used a high-
simulated results. The main findings and directions for future resolution electrostatic-scanning particle spectrometer to study
2
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
Fig. 2. Infected persons can spread diseases through a variety of aerosol forms, including close contact and inhalation by susceptible persons (upper part); airborne aerosols
inhaled by susceptible persons (medium); or, after sedimentation, spreading through touching the mucosa and spreading on surfaces used by susceptible persons (lower
part). Source: U.S. Centers for Disease Control and Prevention.
the distribution and attenuation of sub-micron aerosols. They 2.3. Motion characteristics of indoor inhalable particles
showed that the aerosol produced by a 15-minute mosquito-
repellent incense recovered to the initial level after 1–2 h. Li The distribution and movement of indoor particles in a venti-
et al. [17] carried out real-time online monitoring on the size and lated room are affected by several factors, mainly including parti-
composition of single aerosol particles. Their results showed that cles characteristics, room structures, indoor heat sources, the
the particle range of fresh cigarette smoke is wider than that of location of indoor particles, airflow distribution, and so on. The
aging smoke. Zhou et al. [18] have conducted an experiment, influence of airflow distribution includes the ventilation form, ven-
which showed that the PM2.5 concentration outdoors can influence tilation rate, and layout of air inlet and outlet. The motion charac-
the indoor PM2.5 concentration. Ahmadzadeh et al. [19] have ana- teristics of indoor particles are mainly affected by gravity
lyzed the characteristics, distribution, and transmission of sedimentation, inertial collision, particle diffusion, particle coagu-
COVID-19 virus particles. lation, particle attachment, and rebound. Yao and Li have con-
3
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
ducted an experiment on the sedimentation characteristics of Different air-conditioning and ventilation methods lead to dif-
indoor inhalable particles with an underfloor air distribution sys- ferent air distribution forms, where the movement and distribution
tem, and proved that the settlement of large particles is mainly of particles under the various flows are different. According to the
determined by gravity, while the settlement of small particles is experimental results of Zhao and Li et al. [20,28], the sedimenta-
determined by the Brownian diffusion force [20]. Jin et al. [21] tion and concentration of particles are mainly affected by ventila-
explained the flow and diffusion of suspended particles from out- tion conditions. The sedimentation of large particles is mainly
doors to indoors, and showed that there are differences in the determined by gravity, while the sedimentation of small particles
steady-state particle density at different locations in a room under is determined by the Brownian diffusion force. Meanwhile, increas-
different ventilation conditions. Chang et al. have focused on sim- ing the air exchange rate can reduce the concentration of inhalable
ulation of the thermal stratification height with underfloor air dis- particles when using an underfloor air distribution system. There-
tribution using EnergyPlus, and obtained influence factors for the fore, ventilation provides an effective measure to dilute and control
thermal stratification height, providing reference for building the indoor particle concentration.
design [22,23] Table 1. This section discusses the mechanisms associated with indoor
particles, involving the basic concepts, their distribution and
2.4. Ventilation and its influence on indoor particles movement, and indoor ventilation laying a theoretical foundation
for the experimental study and mathematical models for the
Underfloor air distribution (UFAD), as part of the design of an numerical simulation of the indoor inhalable particle distribution
HVAC system, is used to provide ventilation and space conditioning and movement.
in buildings. Differing from traditional ceiling air-conditioning sys-
tems, it has many distinct characteristics, such as using the air ple-
3. Research methods
num beneath a raised floor to provide conditioned air through
diffusers directly to the occupied zone. It has been applied in Nor-
3.1. Experimental study
dic countries since as early as the 17th century. Due to its potential
advantages of better indoor thermal comfort and indoor air quality,
Experimental measurement is an important and reliable means
ventilation efficiency, layout flexibility, lower life-cycle cost, and
to study the distribution and movement rules of indoor particles.
energy saving, it has been increasingly favored by designers and
By using different instruments to measure the concentration and
widely used in the past two decades [4,24–27]. As people put for-
spatial distribution of indoor particles, we can evaluate the indoor
ward higher requirements for thermal comfort and the air quality
air quality. At present, cumulative sampling devices and continu-
of building environments, research focusing on UFAD systems has
ous monitors are mainly used in the research of atmospheric inhal-
been ongoing.
able particles. Current experimental technologies can monitor the
overall indoor particle concentration and distribution, as well as
Table 1 the movement of individual particles.
Comparison between displacement of Brownian diffusion and gravitational sedimen- In this experiment, we combine the flow and diffusion charac-
tation distance of particles in standard condition. teristics of particles with an indoor airflow field, focusing on the
Particle size dp (lm) DxD ðmÞ xG ðmÞ DxD =xG distribution of indoor inhalable particles and exploring potential
indoor air quality problems occurring when using an underfloor
0.00037* 6103 2.4109 2.5106
0.01 2.6104 6.6108 3900 air distribution system. The GRIMM Model 1.108 aerosol spec-
0.1 3.0105 8.6107 35 trometer was used to monitor the concentration changes for
1.0 5.9106 3.5105 0.17 inhalable particles with and without indoor air supply by the
10 1.7106 3.0103 5.7104 underfloor air distribution system, and a mosquito-repellent
*
Equal to the diameter of an air molecule. incense experiment was conducted to analyze the diffusion of par-
ticles and impact on indoor air quality, thus providing experimen-
tal data that can be used to verify the accuracy of the numerical
results.
lighting 20%, person 100%, computer equipment 100%, and enclo- Table 2
sure structures 60%. Thus, the total heat gain of the working area Technical parameters of Grimm Model 1.108 aerosol spectrometer.
5
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
Table 3
Comparison of the most recent research methods and conclusions.
Table 4
Comparison of different particle phase models.
Model Method Effect of particles Interphase sliding UCS Particle transport properties
Single particle dynamic model Discrete system Neglect Yes Lagrange None
Sliding model Continuous medium Neglect Yes Euler Yes
Non sliding model Continuous medium Partial Consideration None Euler Yes (diffusion equilibrium)
Particle-trajectory model Discrete system Consideration Yes Lagrange None (orbit determination)Yes (stochastic)
Pseudo-fluid model Continuous medium Consideration Yes Euler Yes
synergistic and complementary relationship and can be used indoor particle concentration, and the larger particles were
together to enhance the research findings. reduced more significantly. This was mainly caused by the parti-
cles being carried by the flow pattern in the wind direction of
the underfloor air supply. When the underfloor air distribution
4. Research results was started, the lighter particles may be suspended in a short time,
while the large particles can easily to settle or be taken away with
4.1. Experimental results the exhaust air, due to their high gravity and settling speed. During
continuous ventilation, the particle concentration decreased grad-
4.1.1. Particle size distribution of indoor inhalable particles ually, as evidenced by a reduction in suspension rate. When the
A comparison between the steady-state concentration and ini- indoor particle concentration reached a stable state, the settlement
tial concentration of inhalable particulate matter in the under- and discharge of particles were larger than their suspension value.
floor air distribution room is shown in Fig. 4. The results showed The 0.3–10 lm particulate matter was divided into four levels
that fine particles (0.3–2.5 lm) had a great effect on the total par- for statistical analysis (0.3–0.5 lm, 0.5–1 lm, 1–2.5 lm, and
ticle concentration, where the highest concentration appeared at 2.5–10 lm), and we calculated the ratio of the steady-state value
0.3 lm (8.91104/L); coarse particles (2.5–10 lm) also had a great to the initial value for each indoor particle concentration. The
effect on the total particle mass concentration, where the highest results are listed in Tables 5 and 6. It can be further seen, from
concentration appeared at 3 lm (8.283 lg/m3). It was found that Table 5, that the indoor pollution of 0.3–0.5 lm particles was
the use of underfloor air distribution significantly reduced the 152,080/L, accounting for over 90% of the total particles when
6
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
Fig. 4. Comparison between stable and initial concentrations of particles under UFAD system.
Table 5
Particle concentration in different particle size range.
Particle range(lm) Steady state value of air supply n Initial value n0 n/n0
No./L Cumulative frequencies No./L Cumulative frequencies
0.3–0.5 94,609 94,609 142,080 152,080 0.6221
0.5–1 3733 98,342 7731 159,811 0.4828
1–2.5 360 98,702 910 160,721 0.3955
2.5–10 62 98,762 209 160,930 0.2968
7
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
Table 6
Particle mass concentration in different particle size ranges.
Particle range(lm) Steady state value of air supply m Initial value m0 m/m0
lg/m 3
Cumulative mass concentration lg/m 3
Cumulative mass concentration
0.3–0.5 8.442 8.442 14.849 14.849 0.5685
0.5–1 3.900 12.342 8.053 22.902 0.4842
1–2.5 2.823 15.165 10.705 33.607 0.2637
2.5–10 4.295 19.460 45.755 79.362 0.0939
the office was not ventilated during working hours. Once inhaled, mosquito coil incense can produce the carcinogens benzopyrene,
these particles can stay in bronchial terminals and even enter the CO, and other harmful substances during its combustion, causing
alveoli. The smaller the particle size, the larger the surface area deterioration of indoor air quality. Hence, after the indoor particle
of the particles and the higher their physical and chemical activity. concentration reached a stable state in the condition of the under-
It is easier to adsorb some harmful heavy metals and organic sub- floor air distribution, solid mosquito coil incense was used in a dif-
stances, which can be seriously harmful to human health. In the fusion experiment, in order to explore the effect of diffusion of
case of underfloor air distribution, due to the dilution and exhaust- inhalable particles on indoor air quality.
ing of fresh air, 0.3–0.5 lm particles were reduced to 94,609/L Fig. 5 shows the comparison of indoor particle concentrations
which was 62.21% of the initial concentration, thus greatly reduc- with different particle sizes before and after burning the
ing the threat of fine particles to human health. mosquito-repellent incense. It can be seen, from the figure, that
It can be seen, from Table 6, that when there was no ventilation, the particle concentration and particle mass concentration of fine
the particle mass concentration of the indoor particle sections of particles below 0.5 lm obviously increased, while the concentra-
size 0.3–0.5 lm, 0.5–1 lm, 1–2.5 lm, and 2.5–10 lm were tion of coarse particles above 2.5 lm had little change after burn-
14.849 lg/m3, 8.053 lg/m3, 10.705 lg/m3, and 45.755 lg/m3, ing the mosquito-repellent incense. The peak position of the
respectively. The mass concentration of indoor PM2.5 and PM10 particle size distribution remained unchanged before and after
was 33.607 lg/m3 and 79.362 lg/m3, respectively. When the burning the mosquito-repellent incense, and the distribution of
underfloor air distribution was kept stable, the particle mass con- particle number concentration showed an unimodal pattern with
centrations of the above four particle sections were reduced to peak position at 0.3 lm (from 58,258/L before burning to 81,652/
8.442 lg/m3, 3.900 lg/m3, 2.823 lg/m3, and 4.295 lg/m3, respec- L after burning), while the distribution of particle mass concentra-
tively; at this time, the mass concentration of the indoor PM2.5 and tion showed a bimodal pattern with peak positions at 0.3 lm and
PM10 was 15.165 lg/m3 and 19.460 lg/m3, respectively. Compar- 3 lm (0.3 lm particles rose from 2.039 lg/m3 before burning up to
ing the indoor PM2.5 and PM10 before and after underfloor air dis- 3.616 lg/m3 after burning; while the particle mass concentration
tribution, their concentrations decreased by 54.88% and 75.48%, of 3 lm particles remained almost unchanged). Based on statistical
respectively. Small-sized particles have good follow-up perfor- calculations, the mosquito-repellent incense accounted for about
mance, which can be sent to the upper part of the room through 22.6% of the total indoor particles and the associated particle mass
the action of the thermal plume and, finally, discharged from the concentration accounted for about 20.9% of the total particle mass
exhaust outlet. For large-size particles, due to their large settling concentration. Thus, it can be inferred that the particle concentra-
speed, the upward flow from the underfloor air supply was not suf- tion may significantly rise when there are indoor dust sources.
ficient to suspend them at the height of the experimental sampling From the perspective of human health, it is particularly urgent
point, such that most of them settled to the ground. Therefore, the and important to control indoor air pollution and reduce indoor
air quality in the indoor breathing area was improved when using dust.
underfloor air distribution.
According to indoor air quality standard of China (GB/T 18883- 4.2. Mathematical modeling results
2002), the maximum daily mean value of PM10 is 150 lg/m3, while
the second-level standard of American ambient air quality stipu- 4.2.1. Models of particle analysis
lates that the maximum daily average of PM2.5 is 65 lg/m3. The
indoor particle concentrations before and after ventilation did (1) Gas-phase governing equation
not exceed either standard. This is related to the fact that the At present, the High Reynolds number k-e model, corrected by
indoor doors and windows were closed during the experimental Launder and Spalding, is the most widely used for air flow calcula-
period, the persons in the room were not walking around, and tions in air-conditioned rooms. The k-e model is a typical two-
the indoor printers, computers, and other equipment were not equation model. Two parameters, addressing the turbulent flow
running. energy (k) and turbulent dissipation rate (e), are introduced in
In this paper, the particle concentration distribution under the the calculation process. The ability of the k- e Model to predict
UFAD system was studied experimentally and the distribution strong separated flows, flows with large curvature, and strong
characteristics of 0.3–10 lm (i.e., inhalable) particles were ana- pressure gradient flows is weak, due to the need for correction of
lyzed. The results showed that fine particles—especially those with the turbulent viscosity and laminar velocity fluctuation. In this
size below 0.5 lm—had the largest effect on total indoor particle light, the improved realizable k-e Model is better than the standard
concentration, the coarse particles (>2.5 lm) had the greatest con- one for determining rotating flows, boundary layer flows with
tribution to the total indoor particle concentration, and underfloor strong inverse pressure gradient, flow separation, and secondary
air distribution had a significant effect on reducing the indoor par- flows. This model has better performance and can accurately pre-
ticle concentration. dict the diffusion of plane and circular jets. Thus, in this paper,
the core region of gas-phase turbulence was simulated using the
4.1.2. Particle distribution before and after diffusion experiment k-e Model, combined with the Boussinesq density hypothesis to
As an experimental object, the smoke from mosquito-repellent calculate the influence of buoyancy on the flow field. The wall
incense can be used as a source of indoor inhalable particles. Solid function method is used in the wall area; that is, the physical quan-
8
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
Fig. 5. Distribution of indoor particle concentration before and after mosquito-repellent incense burning.
tity at the wall is connected with the variables to be solved in the where leff is the effective viscosity coefficient of turbulence
turbulent core area by using a semi-empirical formula, instead of (leff ¼ l þ lt ).
solving in the wall area. The basic equations are as follows. Energy equation:
Continuity equation:
@ qT @ q uj T @ @T k lt
@ q @ ðqui Þ þ ¼ þ ð3Þ
þ ¼0 ð1Þ @t @xj @xj @xj cp rT
@t @xi
whererT = 0.9–1.0.
Momentum equation:
Turbulent kinetic energy equation (k equation):
@ qui @ q ui uj @p @ @ui @uj
þ ¼ þ leff þ @ ðqkÞ @ ðqkui Þ @ lt @k
@t @xj @xi @xj @xj @xi þ ¼ lþ qe ð4Þ
@t @xi @xj rk @xj
þ qbðT T1Þg i ð2Þ
Turbulent energy dissipation rate equation(e equation):
9
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
@ ðqeÞ @ ðqeui Þ @ lt @e qc1 e @ui @uj tively define the momentum change rate of particles in the vertical
þ ¼ lþ þ þ c2
@t @xi @xj re @xj 2 @xj @xi and tangential directions after collision with the wall; that is,
qe2 t2;n
pffiffiffiffiffiffi ð5Þ en ¼ ð9Þ
Kþ ve t2;n
where
t2;t
g et ¼ ð10Þ
rk ¼ 1; re ¼ 1:2; c2 ¼ 1:9c1 ¼ max 0:43; t1;t
gþ5
2 1=2 k @ui @uj where ‘‘100 and ‘‘2” represent the quantities before and after the col-
g¼ 2Eij Eij ¼ 0:5 þ
e @xj @xi lision, respectively. If the normal restitution coefficient is 1.0, the
direction of a solid particle changes only after collision with the
2 wall, and the velocity perpendicular to the wall remains unchanged.
k
lt ¼ c l q If the tangential restitution coefficient is 1.0, it means that the tan-
e gential velocity direction remains unchanged after the particle col-
lides with the wall. The default values of two recovery coefficients
are 1.0; that is, there is no momentum loss and the collision is com-
(2) Governing equation of particle phase pletely elastic.
Trap denotes the capture boundary condition, which means
In FLUENT, the trajectories of dissociated particles can be that the particle is captured by the wall and the orbital calculation
obtained by integrating the differential equation of particle force is terminated.
in Lagrange coordinates. According to Newton’s law of motion, Escape denotes an escape boundary condition, where particles
the force balance equation of particles in Cartesian coordinates is escape and terminate the orbit calculation. Escape boundary condi-
established as tions are generally used on the exit surface.
Interior denotes the internal boundary condition, where parti-
dupi
¼ Fi ð6Þ cles will pass through the internal flow areas.
dt In this paper, the boundary condition of the air supply (return)
where upi is the velocity of particles in direction i (m/s); and Fi is the outlet was escape, while the boundary condition of other walls was
unit mass forces applied to particles in direction i (m/s2): trap.
!
dupi 18 C D Re q (4) Phase coupling
¼ ui upi þ g i 1 þ F ai ð7Þ
dt qp dp 24
2 qp
In calculations involving fluid and particle phases, the problems
where F ai denotes the other external forces received by the unit par- of single-phase and two-phase coupling arise. If only the influence
ticle mass, which are related to the flow field conditions and parti- of fluid relative to the discrete particle phase is considered, and the
cle properties. With different particle properties and flow fields, the influence of particles relative to the fluid phase is not considered,
magnitudes of these forces vary with respect to each other. Based single-phase coupling is used. The orbit calculation in single-
on an analysis of particle force and motion characteristics, for the phase coupling is based on the flow field calculation results for a
particle trajectory model in indoor environment, the pressure gradi- continuous fluid phase. When the particle load is relatively high,
ent force, virtual mass force, and Basset force can be ignored, and not only does the influence of the fluid relative to the particles
the fluid drag force, gravity, Brownian force, and Saffman force need to be considered, but also the influence of discrete particles
should be considered. Then, Eq. (7) can be written as relative to fluid phase; that is, bidirectional coupling. Two-phase
! coupling has higher accuracy than single-phase coupling, but the
18l C D Re q
Fi ¼ ui upi þ g i 1 þ F bi þ F si ð8Þ mathematical model is more complex.
qp dp 2 24 qp The solution procedure for two-phase coupling is shown in
Fig. 6. When the particles used for calculation contain high mass
In order to simplify the simulation processes, the following
and momentum loads, the two-phase coupling method must be
assumptions were adopted in this study:
used in order to calculate the influence of the relative discrete con-
tinuous phase.
The mass and heat transfer between the particle phase and gas
phase are not considered;
There is no coagulation or splitting of particles (i.e., the particle 4.2.2. CFD numerical model
size remains unchanged in the calculation process);
All particles are spherical. All particles of any shape can be (1) Physical model
equivalent to a sphere through use of the ‘‘aerodynamic equiv- Taking an office with the under-floor air distribution system as
alent diameter”. the research object, geometric simplification and boundary condi-
(3) Wall conditions of particle phase tion setting were carried out. The room size was 6 m 7 m 2.
8 m, with a diameter of 0.2 m air-returning outlet close to the ceil-
The boundary conditions of the discrete phase model can be ing of the room, 1.2 m away from the north wall, and 2.5 m away
defined separately in each area. The boundary conditions of the from the east wall. The setting of the air supply outlet was the
granular phase involved in FLUENT mainly include reflect, trap, same as in the experiment. Considering the actual situation during
escape, and interior. the experimental test, the room was equipped with eight central-
Reflect denotes the reflection boundary condition, where the ized heat sources, including two sitting human bodies, two com-
particles collide with the wall and bounce back, resulting in puters, and four lights. The heat dissipation capacity of the
momentum change. The amount of momentum change is deter- human body was assumed to be 75 W/person, the heat dissipation
mined by the normal restitution coefficient en and the tangential capacity of the computers was 100 W/set, and that of the lights
restitution coefficient et (see Fig. 6). The former and latter respec- was 60 W/lamp. The basic structure of the room is shown in Fig. 7.
10
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
Table 7
Indoor equipment parameters.
11
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
12
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
sion. Therefore, they are easily suspended indoors and carried by Table 10 and Fig. 12 show the comparison of discharge, sedi-
the air flow to the upper part of the room for discharging at the mentation, and suspension quantities of particles with different
exhaust outlet, resulting in more suspension and discharge trajec- particle sizes. From the above figures, it can be seen that small-
tories. Due to their high gravity and sedimentation velocity, the sized particles were easier to desorb from the wall and rolled into
large-sized particles tended to return to the ground before reach- the fluid, and were also easier to discharge from the air outlet
ing the exhaust outlet in the suspension process, thus forming under the action of the upward flow formed by the thermal plume
many sedimentation trajectories. of the floor air supply. Larger particles settled more on the floor
13
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
Table 9 Table 10
Initial conditions of particles. Comparison of orbital numbers of particles with different sizes.
Particle Mass of particles per Trajectory Particles number of Particle size/ Discharge Sedimentation Suspended
size/lm trajectory/lg number per trajectory lm trajectory trajectory trajectory
0.3 1.2101 200 8.084106 0.3 123 23 54
1 1.2101 200 2.183105 1 98 56 46
2.5 1.2101 200 1.397104 2.5 43 122 35
5 1.2101 200 1.746103 5 26 148 26
10 1.2101 200 2.183102 10 9 172 19
14
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
Fig. 12. Comparison of particle trajectory numbers with different particle sizes.
particles more easily move with the fluid, and the air quality in
the working area is enhanced.
Fig. 14. Concentration cloud diagram of mosquito-repellent incense diffusion process (X = 5.5 m).
Fig. 17. Temperature comparison at the measurement points (X = 5.5 m, Z = 2.7 m).
16
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
ues at the height of 1.1 m were 0.135 m/s and 0.1 m/s, respectively. area of the room, with horizontal flow direction and rela-
The temperature at test points was obviously layered in the height tively low flow velocity. Therefore, more particles are sent
direction of the room, and there was a large temperature jump to upper part of the room through the thermal plume, which
near the floor. The temperature rose with each layer from 0.1 m more easily move with the fluid, leading to better air quality
to 1.5 m. The temperature increases for the simulated and test in the working area.
values in this area were 2.96 ℃ and 2.73 ℃, respectively. The (3) The smoke diffusion experiment using mosquito-repellent
increase in temperature at the top of the room was not obvious incense revealed that the number and mass concentration
(Figs. 16–18). of fine particles below 0.5 lm obviously increased; however,
From the experimental comparison between the simulated and the indoor air quality can meet the design requirements
measured values of particle concentration, it was found that the using the underfloor air distribution. The results of numeri-
simulated value was generally lower, for which four main reasons cal simulation were consistent with those of the experimen-
can be noted. First, only the ground dust was considered in the tal measurements, proving that the Lagrange method is an
numerical simulation and the influence of wall surfaces was effective means to simulate the steady-state and dynamic
ignored. Second, the amount of dust caused by the testers was distribution of indoor particles, and has applicability and
not included in the calculation. Third, the setting of no-slip and reliability for the improvement of building environments.
capturing boundary conditions for the solid-wall velocity simpli- (4) The trajectory simulation of indoor particles showed that the
fied the complex behavior of particles and the flow field on the side small-sized particles had a relatively high number of sus-
wall (e.g., partial slip, rebound, and so on), such that the random pension and discharge trajectories. As the particle size
disturbance in the calculation results was weak. Fourth, due to became bigger, the number of suspended and discharged
the limitations of the particle concentration measurement instru- particles reduced, while the settled particles increased.
ment (multi-point measurement cannot be synchronized), as well (5) Simulation of the unsteady diffusion process of the
as those of the test conditions, the few measurement sample points mosquito-repellent incense smoke further proved that
and single-point sampling times, the measured data cannot fully small-sized particles easily diffuse to the upper area through
meet the accuracy needs. However, considering the rules and trend the thermal plume and are mostly discharged from the
changes reflected by the simulation and test values, the test was exhaust outlet. Meanwhile, large-sized particles tend to re-
relatively successful and can meet research needs, to a certain settle on the ground, due to their high gravity and sedimen-
extent. tation speed.
(6) We discussed the powerful diffusion and movement forces
5. Discussion of particles, which seem to imply that the potential threat
of aerosol transmission of respiratory infectious diseases
5.1. Findings cannot be ignored in relatively enclosed indoor environ-
ment. The experiment indicated that, once particles form
In this study, we took the inhalable particles in an indoor envi- aerosols—especially in narrow and confined spaces, such as
ronment with an underfloor air distribution (UFAD) system as the enclosed offices, elevators, carriages, cinemas, and so on—a
research object, in order to establish mathematical models. Based greater risk of infection is posed. Therefore, it is necessary
on the analysis of numerical models, the realizable k e two- to conduct indoor ventilation research, in order to reduce
equation model and discrete phase model in the Lagrange method the aerosol concentration and, consequently, the virus
were used to simulate the indoor turbulence and particle phase. density.
Simulation of the interior airflow field, particle trajectories, and
particle diffusion is useful for predicting the floor air supply purifi-
cation process and controlling pollutants through air distribution 5.2. Limitations and future directions
methods. Finally, the numerical simulation results were compared
with experimental results, in terms of the temperature distribution Based on our results, there are several limitations to be noted,
in the vertical dimension of the room, the axial velocity of the tuy- which also provide directions for future research.
ere, and the indoor particle concentration when the air supply
reached a stable state, which verified the feasibility and correct- (1) Complementing the theoretical study on particle suspension
ness of the numerical models. from wall to turbulence, the mechanism of particle desorp-
Based on the theoretical review, experimental measures, and tion can be further analyzed and a particle desorption model
numerical simulation, our main findings are as follows: can be established.
(2) Establishing an experimental system that can be precisely
(1) The experimental study demonstrated that fine particles, controlled to eliminate more interference factors, such as
with size below 0.5 lm, are mainly affected by Brownian the dust raised by the testers, the particles infiltrating from
forces and have a great impact on the indoor particle con- outdoors into indoors, and multiple working conditions and
centration. Meanwhile, the coarse particles, with size above measuring points, in order to comprehensively analyze the
2.5 lm, are greatly affected by gravity and have an obvious influence of ventilation, indoor heat sources, particle source
impact on the total mass concentration of indoor particles. locations, and other factors on the distribution of the indoor
Viruses can be transmitted by particles, while the UFAD particle concentration.
can significantly reduce the indoor particle concentration (3) Due to the limitations of computing in numerical simulation,
and create a cleaner indoor environment. some settings were simplified in this research. In order to
(2) Experimental analysis of the indoor flow field found that the simulate the motion distribution of indoor particles more
small-sized particles gradually diffused to the upper area of accurately, more particle tracks can be set while considering
the room and were discharged from the exhaust outlet the turbulent diffusion of particles and monitoring the depo-
under the action of the UFAD. We also found that the air dis- sition of particles on the wall, in order to obtain more appli-
tribution with the UFAD formed a cold-air lake in the lower cable motion rules.
17
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
18
X. Zhuang, Y. Xu, L. Zhang et al. Energy & Buildings 271 (2022) 112309
[5] Y.M. Fan, X.D. Li, Y.H. Yan, J.Y. Tu, Overall performance evaluation of underfloor [22] Z.G. Chang, S.H. Cao, Y. Lu, G.Q. Yu, Thermal stratification height simulation on
air distribution system with different heights of return vents, Energy and underfloor air distribution by EnergyPlus, Building Energy Efficiency. 315 (5)
Buildings. 147 (2017) 176–187, https://doi.org/10.1016/j.enbuild.2017.04.070. (2017) 22–24, In Chinese.
[6] B. Rahmati, A. Heidarian, A.M. Jadidi, Investigation in performance of a hybrid [23] H.D. Wang, Q.R.Chen, Y.Liu, Investigation of the quasi-periodic airflow
under-floor air distribution with improved desk displacement ventilation oscillation and ventilation performance of an enclosed environment with a
system in a small office, Applied Thermal Engineering. 138 (2018) 861–872, jet air supply, Energy and Buildings.265(2022)112104, . 2022.112104.
https://doi.org/10.1016/j.applthermaleng.2018.03.015. [24] A. Alajmi, W. El-Amer, Saving energy by using underfloor-air-distribution
[7] T. Zhang, Y. Zhang, A. Li, Y.i. Gao, Y. Rao, Q. Zhao, Study on the kinetic (UFAD) system in commercial buildings, Energy Conversion and Management.
characteristics of indoor air pollutants removal by ventilation, Building and 51 (8) (2010) 1637–1642, https://doi.org/10.1016/j.enconman.2009.12.040.
Environment. 207 (2022) 108535. [25] D.F. Ye, G.R. Yang, Design of VAV air conditioning system, Beijing: China
[8] S. Liu, M. Koupriyanov, D. Paskaruk, G. Fediuk, Q. Chen, Investigation of Construction Industry Press (2007:163,), https://doi.org/10.1016/j.
airborne particle exposure in an office with mixing and displacement enbuild.2010.10.017.
ventilation, Sustainable Cities and Society 79 (2022) 103718. [26] Y.X. Huang, Design of underfloor air distribution system for Lujiazui X3-2
[9] J. Kaiser, Mounting evidence indicts fine particle pollution, Science. 307 (2005) project in Shanghai. Refrigeration and Air-conditioning.11(3) (2011) 78-82. (In
1858–1861, https://doi.org/10.1126/science.307.5717.1858a. Chinese)
[10] M. Pirani, N. Best, M. Blangiardo, S. Liverani, R.W. Atkinson, G.W. Fuller, [27] N.S. Lopez, S.K. Galeos, B.R. Calderon, D.R. Dominguez, R. Lyengar,
Analysing the health effects of simultaneous exposure to physical and Computational fluid dynamics simulation of indoor air quality and thermal
chemical properties of airborne particles, Environ. Int. 79 (2015) 56–64, stratification of an underfloor air distribution system (UFAD)with various vent
https://doi.org/10.1016/j.envint.2015.02.010. layouts, Fluid Dynamics and Materials Processing. 17 (2) (2021) 333–347.
[11] G.D. Thurston, J.D. Spengler, A quantitative assessment of source contributions https://doi.org/10.32604/fdmp.2021.011213.
to inhalable particulate matter pollution in metropolitan Boston, Atmospheric [28] B. Zhao, Y. Zhang, X. Li, et al., Comparison of indoor aerosol particle
Environment (1967) 19 (1) (1985) 9–25. concentration and deposition in different ventilated rooms by numerical
[12] W. Huang, Research progress on the relationship between chemical method, Building and Environment 35 (1) (2004) 1–8, https://doi.org/10.1016/
composition of atmospheric particulate matter and health effects and its j.buildenv.2003.08.002.
mechanism, Health Research. 41 (2) (2012) 323–327, In Chinese. [29] L.X. Zhou, Theoretical and numerical simulation of turbulent gas particle two-
[13] W. Gu, S.H. Dong, J.Y. Zhang, Research on the status quo and optimization phase flow and combustion, Science Press, Beijing (1994) 177–196, In Chinese.
strategy of indoor air quality in gyms, IOP Conference Series: Earth and [30] K. Zhong, X.F. Yang, Y.M. Kang, Effects of ventilation strategies and source
Environmental Science. 781 (2021) 32–47. locations on indoor particle deposition[J], Building and environment 45 (3)
[14] P.M.Fine, G. R. Cass, B. R. T. Simoneit, Characterization of fine particle (2010) 655–662, https://doi.org/10.1016/j.buildenv.2009.08.003.
emissions from burning church candles, Environment Science & [31] C. Li, M.S. Li, X.X. Ma, Simulation of Particulates Resuspension from Human
Technology.34(14)(1999) 2352-2362, http://doi.org/10.1021/es981039v. Walking in UFAD Room, Applied Mechanics and Materials. 2012 (1800) 170–
[15] S.C. Lee, W.M. Li, L.Y. Chan, Indoor air quality at restaurant with different styles 173, https://doi.org/10.4028/www.scientific.net/AMM.170-173.3639.
of cooking in metropolitan Hong Kong, The Science of the Total Environment. [32] M. Salmanzadeh, G.h. Zahedi, G. Ahmadi, D.R. Marr, M. Glauser, Computational
279 (1–3) (2001) 181–193, https://doi.org/10.1016/S0048-9697(01)00765-3. modeling of effects of thermal plume adjacent to the body on the indoor
[16] H.X. Wang, X.H. Chen, H.Y. Li, Particle size spectrum of submicron aerosol airflow and particle transport, Journal of Aerosol Science. 53 (2012), https://
produced by air and mosquito repellent incense combustion, Environmental doi.org/10.1016/j.jaerosci.2012.05.005.
Science & Technology. 30 (10) (2007) 34–36, https://doi.org/10.3969/j. [33] C. Zhuang, G. Yang, T. Long, Dean. Hu, Numerical comparison of removal and
issn.1003-6504.2007.10.013. (In Chinese). deposition for fully-distributed particles in central- and split-type air-
[17] M. Li, J.G. Dong, et al, Analysis of cigarette smoke aerosol by single particle conditioning rooms, Building and Environment. 112(2017)17-28, https://doi.
aerosol mass spectrometer, Analytical Chemistry. 40 (6) (2012) 936-939, org/10.1016/j.buildenv.2016.11.027.
http://doi.org/10.1016/S1872-2040(11)60555-0. [34] A.Morteza, A.Ehssan, P.Marcelo, Numerical simulation of airflow and particle
[18] Y.H. Zhou, G.F. Yang, X.N. Li, Indoor PM concentrations and students’ behavior distributions with floor circular swirl diffuser for underfloor air distribution
in primary school classrooms, Journal of Cleaner Production. 318 (2021), system in an office environment, Environmental science and pollution
https://doi.org/10.1016/j.jclepro.2021.128460 128460. research international.26(24)(2019)24552-24569, https://doi.org/10.1007/
[19] M. Ahmadzadeh, E. Farokhi, M. Shams, Investigating the effect of air s11356-019-05651-8.
conditioning on the distribution and transmission of COVID-19 virus [35] Z.M. Xiong, G.Q. Zhang, Research status of indoor inhalable particulate
particles, Journal of Cleaner Production. 316 (2021) 128–147, https://doi.org/ pollution, HVAC.34 (4) (2004)32-36, http://doi.org/10.3969/j.issn.1002-
10.1016/j.jclepro. 2021.28147. 8501.2004.04.008. (In Chinese).
[20] G.C. Yao, N.P. Li, Experimental study on the settlement characteristics of [36] S.B. Fred, Underfloor Air Distribution (UFAD) Design Guide, American Society
indoor inhalable particles in floor air supply system, Journal of Safety and of Heating Refrigerating and Air-Conditioning Engineer, USA, 2003, pp. 11–18.
Environment. 8 (1) (2008) 79–81, In Chinese. [37] T.L.Thatcher, D. W. Layton, Deposition, resuspension and penetration of
[21] H.H. Jin, Q.P. Li, et al., Experimental study on distribution and transport particles within a residence, Atmospheric Environment. 29(13) (1995)1487-
characteristics of indoor suspended particles, Journal of Zhejiang University. 1497, http://doi.org/10.1016/1352-2310(95)00016-R.
44 (9) (2010) 1793–1797, In Chinese.
19