Content uploaded by Xiangdong Li
Author content
All content in this area was uploaded by Xiangdong Li on Feb 25, 2016
Content may be subject to copyright.
Research Article Indoor/Outdoor Airflow
and Air Quality
E-mail: jiyuan.tu@rmit.edu.au
CFD study of the effects of furniture layout on indoor air quality under
typical office ventilation schemes
Ruining Zhuang1, Xiangdong Li1, JiyuanTu1,2 ()
1. School of Aerospace, Mechanical and Manufacturing Engineering, RMIT University, PO Box 71, Bundoora, VIC 3083, Australia
2. Department of Building Science, School of Architecture, Tsinghua University, PO Box 1021, Beijing 100084, China
Abstract
The relative freshness of indoor air in breathing zone can be measured by ventilation effectiveness.
Numerous research articles in literature have investigated ventilation effectiveness under different
ventilation schemes, different inlet/outlet positions, and different diffusor types. These researches
seem to have a goal to find a solution to optimize ventilation effectiveness through manipulating
ventilation system. In reality, however, the occupants of a rented office room have no right to
manipulate the ventilation system; instead, they have to accept whatever rented to them. An
important issue thus arises: how to improve ventilation effectiveness without changing ventilation
system? This paper has built a CFD model about a typical office room, validated it by published
experimental data in literature, and then applied it to twelve typical office situations/cases of
different furniture layouts under different ventilation schemes. The simulation results of twelve
cases show that furniture layout is an important factor in indoor airflow and temperature fields,
and the quality of air in breathing zone can be significantly improved by adjusting furniture layout
without making any change in ventilation system.
Keywords
furniture layout,
indoor air quality,
breathing zone,
ventilation effectiveness,
office CFD model
Article History
Received: 22 January 2013
Revised: 30 April 2013
Accepted: 20 May 2013
© Tsinghua University Press and
Springer-Verlag Berlin Heidelberg
2013
1 Introduction
Indoor air quality (IAQ) is an important concern among
many modern communities as people are spending more
time indoors than ever with many spending 80% to 90%
of their lifetime indoors (Gee 2001). Some researchers
conducted a synthetic literature survey of indoor air quality
and summarized that indoor air quality could be affected by
a number of pollutants such as second-hand smoke, volatile
organic compounds (VOCs), asbestos fibres, biological
particles, radon, carbon monoxide, etc. (Austin et al. 2002).
Indoor air quality can be measured by absolute value of
contaminant concentration (Yang et al. 2009), and also can
be measured by relative indicators such as air age (Buratti
et al. 2011; Li et al. 2003) and ventilation effectiveness (Coffey
and Hunt 2007; Pereira et al. 2009; Rim and Novoselac
2010). Larger ventilation rate (air change per hour) would
reduce pollutant concentration and improve IAQ, but
in the same time would consume more energy and create
stronger indoor airflow that might cause discomfort to
some occupants. Therefore, a balanced solution needs to be
considered.
Due to the fact that pollutants are usually distributed
unevenly in the room, ventilation effectiveness provides a
key to the balanced solution because what really matters is
the pollutant concentration at breathing zone. Ventilation
effectiveness is the ratio of the concentration of pollutant in
the ventilation outlet to the concentration of pollutant in
the breathing zone (Rim and Novoselac 2010) and has a
form of
outlet
breathing_zone
VE C
C
= (1)
where C represents concentration of pollutant at different
locations. For a given ventilation rate (i.e. fixed energy con-
sumption of ventilation) combined with a given pollutant
emission rate, the concentration of pollutant at the ventilation
outlet is a constant regardless of other conditions due to
BUILD SIMUL (2014) 7: 263–275
DOI 10.1007/s12273-013-0144-5
Zhuang et al. / Building Simulation / Vol. 7, No. 3
264
mass conservation. However, the concentration of pollutant
in breathing zone might be significantly affected by other
conditions such as position of pollutant source, interior
furniture arrangement, and the details of ventilation system
(including the basic ventilation scheme, positions of inlet and
outlet, and diffusor type). Therefore at least two approaches
exist to improve air quality in breathing zone while keep
energy consumption of ventilation unchanged: (1) manipulate
details of ventilation system, or (2) change interior arrange-
ment such as location of pollutant source, sitting position,
and overall furniture layout.
The majority of published articles on this issue pertain
to the first approach. Although many researchers claimed
that displacement ventilation scheme is generally better
than mixing ventilation scheme in terms of ventilation
effectiveness (Lin et al. 2005; Zoon et al. 2011), some others
found that displacement ventilation scheme is not always
better than mixing ventilation scheme in some situations
such as in a large public transport interchange (Lin et al.
2006) or when the outlet is at lower height (Yin et al. 2009).
For a given ventilation scheme, ventilation effectiveness may
be significantly affected by inlet or outlet positions (Lee et
al. 2009; Xamán et al. 2009; Xing et al. 2001). Different types
of diffusor may also have strong influences on ventilation
effectiveness (Zhang et al. 2009) even though basic ventilation
scheme and positions of inlet/outlet remain the same. Some
researchers recommended stratum ventilation scheme
(Tian et al. 2010) as a better option because fresh air is
horizontally blown into the room at occupant’s breathing
height, while some others proposed personalized ventilation
scheme (Halvonova and Melikov 2010; Li et al. 2010) which
supplies fresh air directly to occupant’s breathing zone. It is
also found that the size of pollutant particle might affect
performance of ventilation system. A certain ventilation
scheme might have better ventilation effectiveness than
another ventilation scheme when particle size is small, but
might have worse ventilation effectiveness when particle
size is large (Pereira et al. 2009; Rim and Novoselac 2010).
Due to many contradictory results in literature, the effects
of ventilation system on ventilation effectiveness remain
extremely complex and unclear. There seems no universally
“best” ventilation system which can achieve optimal
ventilation effectiveness under any situation. Solution must
be worked out case by case by considering all factors in the
ventilation system.
In reality, however, most companies have no building
of their own thus have to rent office rooms in large office
buildings. They have no right to change ventilation system
in their office rooms but to accept whatever existing. An
important issue thus arises: how to improve ventilation
effectiveness under given ventilation system? The aforemen-
tioned second approach is considered.
Adjusting the location of pollutant source is an option
(Rim and Novoselac 2010), while changing occupant’s
position is another choice (Xamán et al. 2009). Some
researchers claimed that room furniture has great influence
on the indoor airflow and pollutant distribution (Cheong
et al. 2003; Cheong and Phua 2006; Mortensen et al. 2008),
while some others ranked furniture arrangement and
occupant position as the lowest impact factors to indoor air
quality as long as they do not obstruct airflow in a large
office room (Lau and Chen 2007). In all of these researches
in literature, however, the impact of furniture layout on
ventilation effectiveness has not been particularly investigated;
it is just a subsidiary topic in discussions of how ventilation
schemes and other factors may affect indoor airflow,
temperature and pollutant distributions.
A few researchers had done particular investigation
upon the effects of positions of contaminant source and
occupied regions. Li and Zhao (2004) proposed two indices
(accessibility of supply air and accessibility of contaminant
source) to measure the relative effects of spread of supply air
or contaminant in a 2D space; Zhang et al. (2006) further
developed these ideas into a novel concept called spatial
flow influence factor (SFIF), which tried to obtained optimal
interior arrangement via mathematical approach for a 2D
situation.
Due to the fact that most office users cannot manipulate
ventilation system to improve air quality in breathing
zone, and due to the lack of published articles in literature
which particularly focus on influence of furniture layout
on ventilation effectiveness, this paper is doing necessary
research on this issue. In this study, a validated CFD model
is applied to a typical office room under twelve different
combinations of furniture layout and ventilation scheme.
Results show that furniture layout is an important factor in
ventilation effectiveness, and the quality of air in breathing
zone can be significantly improved by adjusting furniture
layout without making any change in ventilation system.
2 Methodology
This article applied experimentally validated Computational
Fluid Dynamics (CFD) model to investigate formaldehyde
distribution in a typical office room under different com-
binations of furniture layout and ventilation scheme. Three
typical furniture layouts in Fig. 1 and four typical ventilation
schemes in Fig. 2 have been chosen, which make twelve
combinations (i.e. twelve cases) in Table 1.
The bookshelf is assumed to be a new piece of furniture,
which is emitting formaldehyde (one of common VOCs
associated with new furniture) and is selected as a repre-
sentative of some common pollutant sources in office room.
CFD simulations were run for all twelve cases, and ventilation
Zhuang et al. / Building Simulation / Vol. 7, No. 3
265
Table 1 Twelve cases of this study
Layout A Layout B Layout C
MV1 A-MV1 B-MV1 C-MV1
MV2 A-MV2 B-MV2 C-MV2
DV1 A-DV1 B-DV1 C-DV1
DV2 A-DV2 B-DV2 C-DV2
effectiveness of all cases were calculated and compared.
Through the comparison of the results of different cases, the
effects of furniture layout on indoor air quality in breathing
zone (i.e. ventilation effectiveness) were expected to be found.
The commercial program CFX in ANSYS 13.0 has been
chosen to run CFD simulations for all twelve cases. This com-
mercial program has incorporated many thermal dynamics
equations and models, some of which need to be selected in
CFX-Pre according to conditions and requirements of this
particular study.
Fig. 1 Three typical layouts of office room
Fig. 2 Four ventilation schemes (in Layout B as examples)
Zhuang et al. / Building Simulation / Vol. 7, No. 3
266
Due to very low pollutant concentrations in this study
as well as the validation model (formaldehyde about
10–7 kg/m3 and CO2 about 1.0×10–3 –2.0×10–3 kg/m3), the
transport of pollutants is assumed to be controlled by the
airflow thus the existence of pollutants in the air has no
effect on the airflow field. Both formaldehyde and CO2 are
modelled by additional variables in CFX-Pre and transport
equation for additional variable has been chosen, which
has a form of
() () ( )
Φ
ρ
ρρDS
t+⋅ =⋅ +U (2)
where ρ is the mixture density in mass per unit volume,
which approximately equals to the air density due to the
very low pollutant concentration. Φ is the pollutant quantity
per unit volume while
Φ
ρ
= is the pollutant mass con-
centration.
S is the volumetric source term of the pollutant,
and Φ
D is the kinematic diffusivity of the pollutant through
the air.
The Re-Normalisation Group (RNG) k-ε model was
regarded as the most accurate model for indoor airflow
computation amongst eight different turbulence models
investigated by Chen (1995). Therefore this study has selected
RNG k-ε model as turbulence model for all twelve cases as
well as the validation model.
2.1 Geometry and mesh
The office room in this study is 4.0 m (L)×3.2 m (W)×2.8 m (H)
in size while the bookshelf is 1.0 m (L)×0.5 m (W)×2.0 m (H)
(Fig. 1 and Fig. 2). The sedentary human model, which is
in size of an adult man with nose height at 1.15 m and a
shoulder width of 60 cm, is seating near a desk with sizes
of 1.5 m (L)×0.9 m (W)×0.7 m (H) (Fig. 3). The computer
case is 0.40 m×0.36 m×0.10 m while the LCD screen is
0.55 m×0.36 m×0.03 m. A cylindrical momentum source
of 0.10 m (D)×0.10 m (L) is placed in the computer case to
simulate the computer cooling fan that inhales air in the
front and blows hot air in the rear. A square directional
diffusor of 0.3 m×0.3 m (Fig. 2) is applied to all mixing
ventilation schemes (i.e. MV1 and MV2 in Fig. 2) while a
simple rectangle inlet of 0.4 m×0.3 m in the middle of the
base line of left wall is applied to all displacement ventilation
schemes (i.e. DV1 and DV2 in Fig. 2). The outlet in MV1 is
0.5 m×0.15 m in size at the bottom corner of the right wall
(at the bottom of the door for many office rooms), while
the outlets in MV2, DV1 and DV2 are 0.4 m×0.4 m in size at
different positions on the ceiling (the center of outlet of MV1
is 1 m from right wall, while the centers of outlets of DV1
and DV2 are 0.3 m to left wall and right wall respectively).
Fig. 3 Details of human model, desk, chair and computer
Unstructured mesh is adopted in this study. Inflation
layers (prisms) are generated around heat sources, while
tetrahedrons fill out the room space. The total numbers of
mesh elements are around 1.8–2.2 million for twelve different
cases in Table 1. Mesh independence has been checked and
confirmed.
2.2 Boundary conditions and solver control
Boundary conditions were set in ANSYS 13.0 CFX-Pre.
The fluid material is set as “Air Ideal Gas” with buoyancy
reference density of 1.185 kg/m3. Turbulence model is set
as RNG k-ε. Walls, ceiling and floor are all set as fixed
temperature at 25℃. Air change rate is 400% per hour,
thus air speed at inlet is set as 0.44 m/s for all mixing
ventilation cases and is set as 0.32 m/s for all displacement
ventilation cases. Air temperature from the inlet is set as
25℃. All heat sources and their heat loads are summarized
in Table 2.
The bookshelf is assumed to be new and emits for-
maldehyde, which is set as an additional variable (scalar) with
a kinematic diffusivity of 1.9×10–5 m2/s. It is found that
different building/furniture materials may have different
formaldehyde emission rates with order of magnitude ranging
from 1.0×10–10 to 1.0×10–8 kg/(m2·s) (Kim et al. 2010). This
study set 1.0×10–9 kg/(m2·s) as formaldehyde emission rate
for the bookshelf, and total formaldehyde-emitting area is
3.5 m2. In fact, the accuracy of emission rate is not important
in this study because this study is only concerned about
ventilation effectiveness, which is the ratio of two con-
centrations at two different places and does not depend on
absolute value of pollutant emission rate. Transport equation
for additional variable has been chosen for the computation
of formaldehyde diffusion in the room.
Table 2 Heat sources in studied cases
Heat source Heat load (W)
Human 70
Computer 130
Lights 40
2
Zhuang et al. / Building Simulation / Vol. 7, No. 3
267
The cylindrical momentum source in the computer
case, which models the computer cooling fan, is one of the
special features of this study. To the best of our knowledge,
there seems no article in literature about modelling a
computer fan in CFD indoor simulations. It is also very hard
to find experimental data in literature about how much
airflows through an ordinary computer case. However,
specifications of computer fans can be easily found by
visiting seller’s websites of computer accessories in google
search. Airflow-creating ability of a computer fan is usually
measured by CFM (cubic feet per minute), which is usually
ranged from 4 CFM to more than 80 CFM according to our
google search. Therefore the boundary condition of the
cylindrical momentum source in our study was set in such
a way that ensures the airflow through the computer case is
equivalent to 10 CFM as a typical value. The issue of the
effects of computer fan on indoor airflow and air quality
will be investigated in our future study.
For the solver control, residual target is set as 1E–4 and
conservation targets for additional variable (i.e. formaldehyde
in this study) and energy are set as 0.01. At least 6000
iterations are needed for each of twelve cases in this study.
3 Results and discussions
3.1 CFD model validation results
For validation purpose of the CFD model in this study,
a model in a published journal article (Tian et al. 2010)
has been chosen, which has experimental data (and also
CFD simulation results) on velocity, temperature and CO2
concentration distributions in a test office room under
stratum ventilation scheme.
This published model (Fig. 4) has been chosen as the
validation model of this study for the following reasons: (1)
the published model is very similar to our model of this
paper: both models are about an office room with several
heat sources and one pollutant source, and both models
have similar geometry and boundary conditions; (2) both
CO2 in the published model and formaldehyde in our
model of this paper are gaseous contaminants which follow
the same set of physics and mathematical equations in CFD
computation.
This study has re-built the published model in ANSYS
13.0 CFX and run our own simulation. Some details of
geometry of the published model (i.e. the validation model
of this study) are in Fig. 4, which also shows the positions
of Lines 1–9 for experimental measurements. The air from
inlet (0.21 m×0.17 m) has a temperature of 19℃ and a velocity
of 1.19 m/s. Heat loads of human simulator, computer box
and lamps are 75 W, 180 W and 72 W×2 respectively. The
Fig. 4 Configurations and line positions (in mm) of test chamber
(Tian et al. 2010)
kinematic diffusivity of CO2 is 1.5×10–5 m2/s. For more details
on geometry and boundary conditions, please refer to the
original published article (Tian et al. 2010).
CFD simulation has been run by ANSYS 13.0 CFX-Solver,
and then the results were compared with experimental data
regarding velocity, temperature and CO2 concentration
(Fig. 5, Fig. 6 and Fig. 7). It is found that simulation results
are in acceptable agreement with experimental data. Therefore
this re-built model from the published article can be applied
to twelve cases of this study after some modifications in
geometry and boundary conditions which is descripted in
Methodology part.
3.2 Airflow fields of three sample cases
This study has run twelve cases of CFD simulation, and the
results on airflow, temperature and formaldehyde distribution
fields have been obtained. Due to limited space of this
journal article, three cases (A-DV2, B-DV2 and C-DV2 in
Table 1) under the same ventilation scheme (DV2 in Fig. 2)
Zhuang et al. / Building Simulation / Vol. 7, No. 3
268
Fig. 5 Measured (Tian et al. 2010) and simulated velocity profiles at 9 positions
Fig. 6 Measured (Tian et al. 2010) and simulated temperature profiles at 9 positions
Zhuang et al. / Building Simulation / Vol. 7, No. 3
269
are selected here as examples to show that change in
furniture layout will affect airflow field, temperature field
and formaldehyde concentration field.
The airflow field of Case A-DV2 on the middle vertical
plan (y1.60 m), which is shown by Fig. 8(a), was sampled
for discussion. In this case, the inlet is at the bottom of the
left wall while the outlet is on the ceiling near the right wall.
Both pollutant source (i.e. the bookshelf) and human model
are in the left part of the room, where a clock-wise eddy
can be identified here, which tends to trap formaldehyde
in the left part of the room for some time before driving it
to the ceiling along the left wall. In the middle part of the
room, an anti-clock-wise eddy can be seen, which might have
an effect of slowing down the transport of formaldehyde
to the right part of the room by holding it for some time
and then pushing it to the ceiling first. In the right part of
the room, the airflow from the inlet hits the right wall and
rebounds to form a small anti-clock-wise eddy above ground.
However an overall upward airflow pattern in the right part
of the room is very clear, which enables pollutant as well as
heat to be ventilated from the outlet in the ceiling of this
part of the room.
Correspondingly, the airflow field of Case B-DV2 on
the middle vertical plan (y1.60 m) is shown by Fig. 8(b).
The positions of inlet and outlet are the same as those of
previous case (i.e. A-DV2), but positions of furniture and
human model are all changed (refer to Fig. 1 and Fig. 2)
with the pollutant source (i.e. the bookshelf) being moved
to the right part of the room. Compared with the airflow
field of the previous case (Fig. 8(a)), the airflow field of this
case (Fig. 8(b)) still shows an eddy in the left part of the room
but the direction changes from clock-wise to anti-clock-wise,
which has an effect of resisting formaldehyde transport
from the right to the left. In the middle part of the room,
there is no longer an eddy but a broad rising airflow that
performs as a “firewall” to slow down the transport of
formaldehyde from the right to the left by blowing it to the
ceiling first. The buoyance by the hot computer as well as
the hot screen near the middle part of the reference vertical
plan (y
1.60 m) in Layout B (refer to Fig. 1) may be the
important driving force of the broad rising airflow here.
In the right part of the room where the pollutant source
(i.e. bookshelf) and outlet are located, the airflow is more
complicated than that of the previous case. The airflow from
the inlet hits the right wall and rebounds to form a small
anti-clock-wise eddy above ground. However, other small
eddies of different directions also exist in this area; it is very
hard to identify any overall pattern in the right part of the
Fig. 7 Measured (Tian et al. 2010) and simulated CO2 concentration profiles at 9 positions
Zhuang et al. / Building Simulation / Vol. 7, No. 3
270
Fig. 8 Comparison of airflow fields on the middle vertical plan
(y=1.60 m)
room. Referring to the magnitude and direction of velocity
vectors in Fig. 8(b), it seems that small proportion of airflow
in this region can directly goes to the ventilation outlet in the
right side of the ceiling while large proportion of airflow
will travel to the left and join the broad rising airflow in the
middle part of the room before being blown to the ceiling.
The airflow field of Case C-DV2 on the same reference
plan (y1.60 m) is shown by Fig. 8(c). The positions of
inlet and outlet are the same as those of the two previous
cases (A-DV2 and B-DV2), but positions of furniture and
human model are all changed (refer to Fig. 1) with pollutant
source (i.e. the bookshelf) in the left and human in the right
but on difference side of the reference plan (y
1.60 m).
Although no systematic eddy can be seen in either the left
or the middle part of the vertical plan, a broad S-shape
downward airflow pattern in the left and the middle parts
is very clear. The airflow from the inlet hits the left side of
the desk and then goes up to join the horizontal airflow on
the top of the desk before converging with the strong rising
flow near the right wall that is driven by the buoyance effect
of the hot computer. Those red-colored vectors in the middle
of the computer case represent airflow from the simulated
computer fan. Considering the pollutant source (i.e. the
bookshelf) in the left part of the room and the human
model in the right part of the room, the overall airflow
pattern of Case C-DV2 creates a situation that the pollutant
source is in the upstream and the human model is in the
downstream. Such effect may increase formaldehyde con-
centration around the human model. However, due to the
fact that the pollutant source and the human model are in
different sides of the reference vertical plan (y
1.60 m),
the strong rising airflow on the reference plan and between
the hot computer and the ventilation outlet right above
creates a vacuum-cleaner effect that attracts formaldehyde
from the bookshelf side and propels it out of the room
immediately, thus will reduce the chance for formaldehyde
being transported to the human model side.
Let’s put all these information together and compare.
Due to the closeness of the pollutant source (i.e. the bookshelf)
to the human model in Case A-DV2, and the absence of
“firewall” or vacuum-cleaner effect in airflow field around
the human model, it can be expected that the formaldehyde
concentration in human’s breathing zone in Case A-DV2
would be the worst amongst three cases. The distance
between the pollutant source and the human model is almost
the same for Case B-DV2 and Case C-DV2, and buffering
mechanism presents in airflow fields of both cases (i.e.
“firewall” in Case B-DV2 and vacuum-cleaner effect in Case
C-DV2). However, due to the fact that the human model in
Case C-DV2 is in the downstream of the overall airflow
pattern with respect to the pollutant source, formaldehyde
concentration in human’s breathing zone in Case C-DV2
would be expected to be larger than that of Case B-DV2
(but still lower than that of Case A-DV2 due to the longer
distance from pollutant source to human model as well as
the vacuum-cleaner effect in airflow of Case C-DV2). These
expectations have been proved by the simulation results of
concentration fields as well as ventilation effectiveness in
the upcoming sections of this article.
In summary, the comparison of airflow fields on the
same reference plan (y
1.60 m) for three sample cases
(Layouts A, B and C under the same ventilation scheme
DV2) shows that the changes in furniture layout may cause
significantly changes in airflow field, which will further
affect pollutant distribution and ventilation effectiveness
in the typical office room of this study.
Zhuang et al. / Building Simulation / Vol. 7, No. 3
271
3.3 Temperature fields of three sample cases
The temperature fields on the middle vertical plan (y
1.60 m)
for Case A-DV2, Case B-DV2 and Case C-DV2 are displayed
by Figs. 9(a)–(c) respectively. These three graphs are very
similar in terms of positive vertical gradient in temperature
field caused by buoyance. In these three cases, cool air
comes into the room from the bottom of the left wall, and
then is heated by internal heat sources (i.e. the computer
set and human model). The heated air then goes up due
to buoyance, and converges with air heated by two lamps
on the ceiling. Despite of the similarity of three graphs,
Fig. 9 Comparison of temperature fields on the middle vertical
plan (y=1.60 m)
some differences, even though not remarkable, can also be
identified.
The differences in temperature fields of these three sample
cases would be more clear if the reference plan changes.
Figures 10(a)–(c) show temperature fields of these three
sample cases on the horizontal breathing plan (z
1.15 m).
Three hot areas can be seen around the human’s head, the
computer set, and at the intersection between the breathing
plan and the concave-up trajectory of hot airflow coming
Fig. 10 Comparison of temperature fields on the breathing plan
(z=1.15 m)
Zhuang et al. / Building Simulation / Vol. 7, No. 3
272
from the computer fan. Changes in furniture layout will move
the locations of these three hot regions on the temperature
contour, thus will certainly affect the overall airflow field,
pollutant distribution and ventilation effectiveness.
3.4 Formaldehyde concentration fields of three sample
cases
The formaldehyde concentration field of Case A-DV2 on
the middle vertical plan (y1.60 m) is shown by Fig. 11(a).
This graph exhibits a slant gradient pattern: formaldehyde
concentration is higher near the left wall and the ceiling but
is lower at the right wall and the floor. A high-concentration
Fig. 11 Comparison of formaldehyde concentration fields on the
middle vertical plan (y=1.60 m)
center is found at human’s breathing height in the left part
of the room. Such pattern is in agreement with our airflow
results and discussions on Case A-DV2: a clock-wise eddy
in the left part of the room tends to trap formaldehyde in
this area for a certain period of time before driving it to
the ceiling along the left wall, and the second eddy in the
middle part of the room may slow down the transport of
formaldehyde from the left to the right by pushing it to the
ceiling first; fresh air from the inlet travels from the left to
the right along the floor and then hit and spread along the
right wall, making the floor and the lower part of the right
wall being lower concentration zones of formaldehyde.
The formaldehyde concentration field of Case B-DV2 on
the middle vertical plan (y
1.60 m) is shown by Fig. 11(b).
This graph shows a high-concentration center in the middle
part of the room at breathing height, but the magnitude is
much less than the high-concentration center in previous
case. The location of formaldehyde concentration center in
the middle part of the room in Case B-DV2 is in agreement
with our airflow results and discussions on Case B-DV2: the
transport of formaldehyde from the right to the left is slowed
down by the “firewall” (the broad rising airflow in the middle
part of the room); large proportion of airflow containing
formaldehyde will go to the middle part of the room first
before travelling to other parts of the room; as the middle
part of the room acts as a “transit” of formaldehyde transport,
a high-concentration center is thus expected in this area.
The formaldehyde concentration field of Case C-DV2 on
the middle vertical plan (y
1.60 m) is shown by Fig. 11(c).
A high-concentration center exists in the left part of the
room and has the same magnitude as that in Case A-DV2.
However, formaldehyde disperses to a much larger area in
Case C-DV2 than in Case A-DV2 and the low-concentration
region near the ground has been significantly reduced. Such
differences of formaldehyde distribution between Case
C-DV2 and Case A-DV2 are in agreement with our airflow
results and discussions on Case C-DV2 and Case A-DV2:
unlike local eddies in the airflow field of Case A-DV2 which
blowing the formaldehyde-containing air to the ceiling first,
the broad downward airflow across the left part and the
middle part of the room in Case C-DV2 blows formaldehyde
to the lower part of the room, thus extends the dispersion
of formaldehyde in the room.
The formaldehyde concentration fields of three sample
cases on the breathing plan (z1.15 m) are shown by
Figs. 12(a)–(c) respectively. By checking contour lines near
the human’s nose, it can be seen that formaldehyde con-
centration in breathing zone is the highest in Case A-DV2
and the lowest in Case B-DV2. These findings prove our
expectations at the end of airflow discussion section. It
makes Layout B the best furniture layout to achieve optimal
ventilation effectiveness under ventilation scheme DV2.
Zhuang et al. / Building Simulation / Vol. 7, No. 3
273
Fig. 12 Comparison of formaldehyde concentration fields on the
breathing plan (z=1.15 m)
3.5 Ventilation effectiveness of all twelve cases
After three case studies on airflow, temperature and for-
maldehyde concentration fields in the above, it is time to
discuss all twelve cases in terms of ventilation effectiveness.
Based on average formaldehyde concentrations near the nose
and at the outlet for each of the twelve CFD simulation cases,
corresponding ventilation effectiveness is calculated and
displayed in Table 3. Graphical presentations of ventilation
effectiveness of twelve cases are in Fig. 13.
It can be found from Table 3 that the difference between
the best and the worst ventilation effectiveness of twelve
cases is quite significant. Case B-DV2 (the combination of
Layout B and displacement ventilation No. 2) achieves
ventilation effectiveness of 2.689, which is almost 4 times of
ventilation effectiveness of 0.736 in the Case A-MV2. For a
given ventilation scheme, the difference between the best and
the worst ventilation effectiveness is also remarkable. For
example, ventilation effectiveness of 2.689 in Case B-DV2
is more than 2.5 times as much as ventilation effectiveness
of 1.061 in Case A-DV2. Such remarkable differences imply
that it is quite necessary to investigate the effects of furniture
layout on ventilation effectiveness and find an optimal
furniture layout to achieve fresher air in breathing zone for
office users when ventilation scheme is unchangeable.
Figure 13 shows that Layout A achieves the worst (i.e. the
lowest) ventilation effectiveness under all four ventilation
Table 3 Ventilation effectiveness in twelve cases
Case Average VOC at
outlet (kg/m3) Average VOC near
nose (kg/m3) Ventilation
effectiveness
A-MV1 8.52E–08 9.97E–08 0.854
A-MV2 8.75E–08 1.19E–07 0.736
A-DV1 8.63E–08 6.33E–08 1.363
A-DV2 8.79E–08 8.29E–08 1.061
B-MV1 8.80E–08 7.63E–08 1.154
B-MV2 8.61E–08 6.85E–08 1.258
B-DV1 8.83E–08 5.02E–08 1.758
B-DV2 8.36E–08 3.10E–08 2.698
C-MV1 8.99E–08 6.70E–08 1.342
C-MV2 8.83E–08 7.52E–08 1.173
C-DV1 8.31E–08 5.33E–08 1.558
C-DV2 8.67E–08 5.22E–08 1.661
Fig. 13 Ventilation effectiveness of 12 cases grouped by ventilation
scheme
Zhuang et al. / Building Simulation / Vol. 7, No. 3
274
schemes. The distance between the pollutant source (i.e. the
bookshelf) and the occupant, which is the shortest in Layout
A among three furniture layouts, may be an important
reason. Another important reason may be the lack of
buffering mechanism in the airfield field to reduce the
transport of formaldehyde from the bookshelf to the
occupant. Examples of such buffering mechanism have
been given in airflow discussion section of this article.
Figure 13 also shows that two displacement ventilation
schemes (i.e. DV1 and DV2) achieve better ventilation
effectiveness than two mixing ventilation schemes (i.e. MV1
and MV2) for all three furniture layouts in this study. In
literature, many articles showed that displacement ventilation
is better than mixing ventilation in terms of ventilation
effectiveness (Lin et al. 2005; Zoon et al. 2011) while some
others do not agree (Lin et al. 2006; Yin et al. 2009). The
results in this study seem to favour displacement ventilation
scheme.
In both Layout B and Layout C, the occupant is sitting
far away from formaldehyde-emitting bookshelf. However,
if without relating to ventilation scheme, it is hard to decide
which layout has the better ventilation effectiveness. Layout
B performs better in ventilation schemes MV2, DV1 and
DV2, but Layout C performs better in MV1. However, in
terms of average or overall ventilation effectiveness under all
ventilation schemes, Layout B seems to be the better layout
than Layout C. The reasons may lie in the airflow fields.
Referring to airflow and formaldehyde concentration analysis
sections, the human model is generally sitting in the upstream
with respect to the pollutant source in Layout B while sitting
in the downstream in Layout C under ventilation schemes
MV2, DV1 and DV2. Under ventilation scheme MV1,
however, the relative positions of the occupant and the
bookshelf in terms of upstream and downstream are not
clear for both Layout B and Layout C because the inlet is
located in the center of the ceiling (refer to MV1 in Fig. 2).
In short, it seems that better ventilation effectiveness may
be achieved if occupant is sitting in the upstream in the
airflow field with respect to the pollutant source. CFD
simulation is a powerful means to discover relative positions
of upstream and downstream through airflow analysis.
There is only one contaminant in this study (formaldehyde,
a kind of VOCs that affects indoor air quality) and only one
contaminant source (the bookshelf). Therefore this CFD
study is just a simple example about how furniture layout
might significantly affect indoor air quality in breathing
zone. In reality, however, there might be hundreds of VOCs
and multiple contaminant sources in indoor environment.
Future researches are needed to investigate these issues
under much more complicated settings. No matter how
complicated the setting is, the basic approach in this study is
expected to be effective. The numerical procedure developed
in this study has laid a platform on which further researches
could be readily conducted and the gaseous contaminant
distributions in an indoor environment could be quantitatively
predicted as long as the properties and release characteristics
of the VOCs in interests are known.
4 Conclusions
The above discussions on the results led to the following
conclusions:
(1) From the discussions about the airflow fields and the
temperature fields of three example cases, it can be con-
cluded that furniture layout is an important influential
factor of indoor airflow field and temperature field.
Changes in furniture layout will result in changes in
airflow field and temperature field thus will cause changes
in pollutant distribution field and ventilation effectiveness.
The remarkable difference shown in Table 3 and Fig. 13
between the best and the worst ventilation effectiveness
under a given ventilation scheme shows that the change
in furniture layout might remarkably improve the quality
of air in breathing zone even no change has been made in
ventilation system; it justifies the necessity of this study.
(2) However furniture layout is not the only decisive factor
of ventilation effectiveness; details in geometry and
ventilation system must be taken into account. In
order to achieve better ventilation effectiveness through
arrangement of furniture layout, two principles may
apply: (i) it is recommended to arrange the occupant’s
sitting position far away from the pollutant source,
and/or (ii) arrange the occupant’s sitting position in the
upstream of the overall airflow field with respect to the
pollutant source. Airflow field analysis by means of CFD
simulation can help us identify the upstream location
for occupant and the downstream position for the
pollutant source in a certain room.
(3) Although the advantage of displacement ventilation over
mixing ventilation is still open for debate in literature,
the results of this paper clearly favour displacement
ventilation scheme in terms of ventilation effectiveness.
It may be recommended that large office buildings
should try their best to provide displacement ventilation
scheme to room renters in order to give them more
flexibility in arranging furniture while maintaining
acceptable quality of air in breathing zone.
Acknowledgements
The financial supports provided by the Australian Research
Council (project ID LP110100140), the National Natural
Science Foundation of China (Grant No. 21277080) and
RMIT Research Scholarship are gratefully acknowledged.
Zhuang et al. / Building Simulation / Vol. 7, No. 3
275
References
Austin J, Brimblecombe P, Sturges W (2002). Air Pollution Science
for the 21st Century. Amsterdam: Elsevier Science.
Buratti C, Mariani R, Moretti E (2011). Mean age of air in a naturally
ventilated office: Experimental data and simulations. Energy and
Buildings, 43: 2021 2027.
Chen Q (1995). Comparison of different k-ε models for indoor airflow
computations. Numerical Heat Transfer, 28: 353 369.
Cheong KWD, Djunaedy E, Poh TK, Tham KW, Sekhar SC, Wong NH,
Ullah MB (2003). Measurements and computations of contaminant’s
distribution in an office environment. Building and Environment,
38: 135 145.
Cheong KWD, Phua SY (2006). Development of ventilation design
strategy for effective removal of pollutant in the isolation room
of a hospital. Building and Environment, 41: 1161
1170.
Coffey CJ, Hunt GR (2007). Ventilation effectiveness measures based
on heat removal: Part 1. Definitions. Building and Environment,
42: 2241 2248.
Gee I (2001). Monitoring indoor air pollution. Indoor and Built
Environment, 10: 123 124.
Halvonova B, Melikov A (2010). Performance of “ductless” personalized
ventilation in conjunction with displacement ventilation: Impact
of intake height. Building Environment, 45: 996 1005.
Kim KW, Kim S, Kim HJ, Park JC (2010). Formaldehyde and TVOC
emission behaviors according to finishing treatment with surface
materials using 20 L chamber and FLEC. Journal of Hazardous
Materials, 177: 90 94.
Lau J, Chen Q (2007). Floor-supply displacement ventilation for
workshops. Building and Environment, 42: 1718 1730.
Lee KS, Zhang T, Jiang Z, Chen Q (2009). Comparison of airflow and
contaminant distributions in rooms with traditional displacement
ventilation and under-floor air distribution systems. ASHRAE
Transactions, 115(2): 306 321.
Li R, Sekhar S, Melikov A (2010). Thermal comfort and IAQ assessment
of under-floor air distribution system integrated with personalized
ventilation in hot and humid climate. Building Environment, 45:
1906 1913.
Li X, Li D, Yang X, Yang J (2003). Total air age: An extension of the
air age concept. Building and Environment, 38: 1263
1269.
Li X, Zhao B (2004). Accessibility: A new concept to evaluate ventilation
performance in a finite period of time. Indoor and Built Environment,
13:287 293.
Lin Z, Chow TT, Fong KF, Tsang CF, Wang Q (2005). Comparison of
performances of displacement and mixing ventilations. Part II:
Indoor air quality. International Journal of Refrigeration, 28:
288
305.
Lin Z, Jiang F, Chow TT, Tsang CF, Lu WZ (2006). CFD analysis of
ventilation effectiveness in a public transport interchange. Building
and Environment, 41: 254
261.
Mortensen LH, Rode C, Peuhkuri R (2008). Investigation of airflow
patterns in a microclimate by particle image velocimetry (PIV).
Building and Environment, 43: 1929 1938.
Pereira ML, Graudenz G, Tribess A, Morawska L (2009). Determination
of particle concentration in the breathing zone for four different
types of office ventilation systems. Building and Environment,
44: 904
911.
Rim D, Novoselac A (2010). Ventilation effectiveness as an indicator
of occupant exposure to particles from indoor sources. Building
and Environment, 45: 1214
1224.
Tian L, Lin Z, Wang Q (2010). Comparison of gaseous contaminant
diffusion under stratum ventilation and under displacement
ventilation. Building and Environment, 45: 2035
2046.
Xamán J, Tun J, Álvarez G, Chávez Y, Noh F (2009). Optimum
ventilation based on the overall ventilation effectiveness for tem-
perature distribution in ventilated cavities. International Journal
of Thermal Sciences, 48: 1574 1585.
Xing H, Hatton A, Awbi HB (2001). A study of the air quality in the
breathing zone in a room with displacement ventilation. Building
and Environment, 36: 809
820.
Yang W, Sohn J, Kim J, Son B, Park J (2009). Indoor air quality
investigation according to age of the school buildings in Korea.
Journal of Environmental Management, 90: 348
354.
Yin Y, Xu W, Gupta JK, Guity A, Marmion P, Manning A, Gulick B,
Zhang X, Chen Q (2009). Experimental study on displacement
and mixing ventilation systems for a patient ward. HVAC&R
Research, 15: 1175
1191.
Zhang T, Lee KS, Chen Q (2009). A simplified approach to describe
complex diffusers in displacement ventilation for CFD simulations.
Indoor Air, 19: 255
267.
Zhang Y, Li X, Wang X, Deng W, Qian K (2006). Spatial flow
influence factor: A novel concept for indoor air pollutant control.
Science in China Series E-Engineering & Materials Science, 49:
115
128.
Zoon WAC, Loomans MGLC, Hensen JLM (2011). Testing the
effectiveness of operating room ventilation with regard to removal
of airborne bacteria. Building and Environment, 46: 2570
2577.
