Available online at www.sciencedirect.com
ScienceDirect
Procedia Engineering 00 (2014) 000–000
www.elsevier.com/locate/procedia
“APISAT2014”, 2014 Asia-Pacific International Symposium on Aerospace Technology,
APISAT2014
The Effects of Heat Exchanger on Ozone Retention
KE Peng* , YANG Min, ZHANG Shuguang, YANG Chunxin
Beihang University, Beijing, 100191, China
Abstract
There has been an increase in considering the health and aviation safety problem related to bleed air contamination, such as
ozone, with the continued growth of aviation transportation. In this paper, the DPM (Distributed Parameters Model) was
employed and extended to numerically investigate the ozone spread in HX (Heat Exchanger), the key component of aircraft ECS
(Environmental Control System). The equation of mass conservation for each control volume was build, where the flux of the
ozone deposited to the surface was determined by its local concentration in air and the deposition velocities, which refers to
analogy analysis between mass transfer and heat transfer. The physical surface deposition and chemical reactions between ozone
and heat exchanger were both studied with DPM to determine the ozone retention ratio, which could consider the effects of local
flow and heat transfer condition with affordable computation burden. This study presented some numerical tools to analyze the
ozone problem in airplane ECS and reveals that the ECS may have a significant effect on airplane ozone removal.
© 2014 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of Chinese Society of Aeronautics and Astronautics (CSAA).
Keywords: cabin air quality, ozone, Heat Exchanger, numerical simulation;
1. Introduction
The modern large commercial aircraft equipped with environmental control system to provide outside fresh air
and regulate the temperature, pressure and humidity in the cabin to create a comfortable environment for passengers.
The outside air from the bleed air system of engine is directed to the cabin through the ECS, a relative complicated
system working under some extreme conditions. The engine bleed air is prone to contamination from outside, or
engine oil. The contaminants, such as ozone, solid particle, oil vapor and so on will be transmitted to the cabin
* Corresponding author. Tel.: +86-13366836673; .
E-mail address: p.ke@buaa.edu.cn
1877-7058 © 2014 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of Chinese Society of Aeronautics and Astronautics (CSAA).
2
KE Peng / Procedia Engineering 00 (2014) 000–000
through the ECS. Such contamination may degrade the quality of bleeding air and be threats to the heath of
occupants and safety of flights [1]. Ozone is a strong oxidant among these contaminants, which can cause eye and
respiratory irritation effect of short-term exposure at low concentrations, prolonged exposure to high concentrations
cause respiratory system diseases[2]. Flight crew and cabin attendants of civil airlines have reported some
uncomfortable or partially incapacitating symptoms associated with it. The limits of ozone concentration in aircraft
cabin has been regulated in airworthiness standards [3], and corresponding means of compliance has been presented
in advisory circulars, such as FAA AC 120-38 [4].
Currently, most researches focused on the final ozone concentration of in aircraft cabin[5], while there is very
limited knowledge of the transportation and deposition process of ozone during the whole flow path in ECS of
aircraft. ECS in aircraft will generally absorb those contaminants, such as ozone, solid dust particles, oil vapor and
gaseous pollutants, as revealed by experiments conducted by UK CAA and Austrian DSTO[8]. HAVC system in
building is similar to ECS in aircraft, many authors [6]~ [7] found that the HAVC system will contribute much to the
total ozone removal. So ECS will contribute much to the ozone retention in cabin, especially the widely applied
component Heat exchangers (HX), there is still short of tools to investigate the ozone retention inside HX.
Analytic methods [9] and numerical methods [10] have been developed for heat exchangers based on the integral
parameter model (IPM) to evaluate heat transfer inside HX, and then the ozone transportation could also be analyzed
with averaged mass transfer using IPM. However, it has inherent difficulties to characterize the different thermo
physical properties of every point inside HX, which usually works in extreme conditions, such as the fouling or wet
conditions. CFD based simulations [11] have been broadly utilized in recent years to yield accurate results through
solving Navier-Stokes equation with proper flow models, such as wall function, viscosity models and turbulence
models. However, huge computational burden caused by complex geometry and unclear deposition mechanism
limits its application. The DPM [12] is a reduced-order model for fast CFD analysis with much less computation
burden and acceptable precision with the consideration of local thermal physical properties and heat and mass
transfer characteristics, which has been validated in the flow and thermal analysis of ducts and HX. Besides that,
Yang et al [13] have analyzed ozone retention in ducts with DPM successfully.
This paper extended the DPM further to investigate the ozone spread in HX numerically. It was organized as
follows, the DPM model for HX Simulation will be developed firstly in Part II, and where the special numerical
tool, DPM, will be introduced briefly with the basic configuration of HX and the theoretic models for such
interaction will be presented and validated. More results were given in Part III along with some applications. The
conclusion will be drawn at last.
2. DPM model for HX Simulation
2.1. Introduction to DPM
1
4
Tin
2
Tex
3
hotstream
coldstream
Fig. 1 Mesh generation [12]
The mesh generation method according to different fin surfaces will be illustrated by a rectangle-plain-fined heat
exchanger, as shown in Fig.1, where the mesh as the dash line and one of the basic elements acquired is in the right
part. It has definite physical significance, where the width and height of element are fin pitch and height, and its
length is equal to fin pitch of the other stream passage.
KE Peng / Procedia Engineering 00 (2014) 000–000
3
The governing equation of element based on energy balance to solve the temperature field is,
4
gm Ac c p Tout  Tin     KF C,i Ti  TC 
(1)
i 1
Where, i  1, 2,3, 4 presents the four adjacent elements, Ac is cross-sectional area of the element, K and F are
the heat transfer coefficient and area. TC can be presented by the average temperature of inlet and outlet, as,
TC  Tin  Tout  2 . Then, TC can be derived as,
4  KF 
4  KF 

 

C,i Ti
C,i
(2)
TC  Tin  2  
  2  

i 1 g m Ac c p Tin 
i 1 g m Ac c p 

 

In term of Eq.(2), temperature of all elements can be got by stepwise iteration until the system becomes to be
stable.
2.2. Additional equations for mass transfer
2.2.1. Governing equations
Here we add the equation of mass conservation to the basic DPM presented by Zhang [12] to solve the ozone
concentration in whole domain. The equation of mass conservation for the fluid in the element k, as shown in Figure
1, is
(3)
mout, k  min,k  lt vd CE , k dx
Using the ozone concentration at inlet and outlet for each element, Eq. (3) could be rewrote as,
Cout,k  Cin,k  (2 / ru )vd ( x)CE ,k dx
Where, m is the ozone mass flow （ kg/s ） , lt is the circumference of flow passages, C E , k
(4)
is the ozone
concentration at element k, which could be determined using the average concentration at the inlet and outlet, that is,
CE , k  (Cin, k  C out, k ) / 2 .
2.2.2. Determination of deposition velocity vd
Given ozone concentration at inlet, the concentration could be determined easily at each point after solving the
Eq. (4) based on the method of calculate surface deposition velocity vd ( x) , which is defined as 1/ vd  1/ vs  1/ vt ,
where the surface adsorption velocity vs is vs    v  /4 and the transportation limit velocity vt is
vt (x)  DSh(x) / d , Where  is the mass accommodation coefficient (or surface reaction probability), D is the
contaminant diffusion coefficient in air and related to the kind of the dispersion medium, temperature and pressure,
Sh(x) is the local Sherwood number.
Based on the similarity of the mass transfer process and the heat transfer process, the analogy theory is referred
to o establish the analysis model of Sh which shall be guaranteed to have similar flow and boundary conditions. For
the deposition of mass transfer problems, lots of fitting relations established in field of heat transfer can be
referenced by introducing an analogy factor n, so the corresponding mass transfer calculation formula can be written
as, Sh(x)  ( Sc / Pr) n Nu . More details about the suitable correlations can be found in Reference 13.
3. Applications and results
A typical air-air HX in ECS is taken as an example here, where the dimension is 190×114×262(mm) and the
fins used in two sides are both louvered-fin surfaces. During the performance evaluation, the inlet temperature of hot
and could side stream are 221℃ and 118℃, and the mass flux of those are 0.169 kg/s and 0.338 kg/s. The
temperature distributions of HX under steady state are demonstrated in Figure 2, where the temperature varied from
150 to 220 ℃.
4
KE Peng / Procedia Engineering 00 (2014) 000–000
145.2
118.2
51
154.5
163.8
163.8
173.1
182.4
41
191.7
201.0
210.3
31
219.6
173.1
154.5
21
182.4
191.7
11
201.0
210.3
1
Grid index in the direction of cold stream
Grid index in the direction of cold stream
51
124.2
148.1
160.0
41
130.2
142.1
136.1
142.1
154.1
148.1
154.1
160.0
31
166.0
21
136.1
130.2
11
124.2
1
1
11
21
31
41
51
1
11
Grid index in the direction of hot stream
21
31
41
51
Grid index in the direction of hot stream
(a) 12th layer hot stream
(b) 13th layer cold stream
Fig.2. Temperature contour inside the HX,℃
3.1. Clean HX without fouling on wall surface
50
0.9998
0.9989
0.9979
0.9970
0.9960
0.9951
0.9942
0.9932
0.9923
0.9913
0.9904
0.9895
0.9885
0.9876
0.9866
0.9857
40
30
20
10
10
20
30
40
Grid index in the direction of cold stream
Grid index in the direction of cold stream
For the clean HX manufactured with aluminium, the surface deposition will be only considered and the general
the wall surface reaction probability  is about 3.4 105 . The ozone retention ratio is about 0.98 for one flow
passage in the middle layer, as shown in Fig. 3(a). The trends of retention ratio in each flow passage are same
because of the constant material properties were used here. However, due to the huge variation of the distribution of
temperature inside the HX, the material properties, such as density, viscosity and molecular diffusion coefficient of
ozone in air, will influenced too and cause the fluctuation of the ozone retention ratio in each flow passage, as
shown in Fig. 3(b). The retention ratio will decrease heavily where the outlet temperature is high, and the removal
rate will be twice at the passage near the outlet of cold side than the inlet.
50
40
30
20
10
50
10
Grid index in the direction of hot stream
(a)
Constant material properties
0.9995
0.9976
0.9957
0.9938
0.9919
0.9900
0.9881
0.9862
0.9842
0.9823
0.9804
0.9785
0.9766
0.9747
0.9728
0.9709
20
30
40
50
Grid index in the direction of hot stream
(b)
Variable material properties
  3.4  105
Fig. 3.Ozone retention ratio in 12th hot layer, where
Further investigations on the material properties are given in Fig. 4, where we can find that, the retention ratio
will decrease with the increasing of the wall surface reaction probability  and the trends is nonlinear. Considering
the effects of the material properties, the retention ratio will decrease much more.
KE Peng / Procedia Engineering 00 (2014) 000–000
5
1.01
1.00
ozone retention ratio
0.99
0.98
0.97
0.96
0.95
Constant material properties
Variable material properties
0.94
0.93
1E-8
1E-7
1E-6
1E-5
1E-4
1E-3
mass accommodation coeff of the surface
Fig 4. Effects of material properties and wall surface reaction probability on retention ratio
3.2. Used HX with fouling on wall surface
50
0.9769
0.9661
0.9553
0.9445
0.9337
0.9230
0.9122
0.9014
0.8906
0.8798
0.8690
0.8582
0.8474
0.8366
0.8258
0.8150
40
30
20
10
10
20
30
40
50
Grid index in the direction of hot stream
(a) Constant material properties
Grid index in the direction of cold stream
Grid index in the direction of cold stream
For the used HX with fouling on wall surface, the bi-molecular chemical reaction need to be considered,
because the HX may become reservoirs of atmospheric dust, engine oils and adsorbed organic compounds, which
can react with ozone.
Given the first order rate constant of the hypothetical reaction Kb is the level of 10-3, the ozone retention ratio
could reach 0.87 for one flow passage in the middle layer, as shown in Fig. 5. The ozone removal rate is relative
high comparing to Fig. 3 due to its larger rate of chemical reaction between the fouling on wall surface. The trends
of retention ratio in each flow passage are same because of the constant material properties were used here. The
material properties will cause the fluctuation of the ozone retention ratio in each flow passage, as shown in Fig. 5(b) ,
due to the huge variation of the distribution of temperature inside the HX. The retention ratio will decreased heavily
where the outlet temperature is high, and the removal rate is much higher at the passage near the outlet of cold side
than the inlet.
However, the retention ratio in conditions with variable material property is a little higher than that with constant
material property in Fig. 5, where the influence of the material properties to the retention ratio is a little different
between the Fig. 3. The reason is that the surface adsorption velocity in chemical reaction condition is much higher
than the transportation limited velocity, so the later one contributes much to the final deposition velocity, however it
is the contrary to the condition of surface deposition.
50
0.9871
0.9793
0.9716
0.9639
0.9561
0.9484
0.9407
0.9329
0.9252
0.9174
0.9097
0.9020
0.8942
0.8865
0.8787
0.8710
40
30
20
10
10
20
30
40
50
Grid index in the direction of hot stream
(b) Variable material properties
Fig.5 Ozone retention in 12th layer of hot side, where
kb  3.8 103
4. Conclusions
This paper extended the numerical methods, DPM, to investigate the ozone retention in heat exchanger,
considering both surface deposition and bi-molecular chemical reaction. It reveals more information of ozone
6
KE Peng / Procedia Engineering 00 (2014) 000–000
transportation process in HX and evaluates the factors affects the retention ratio, which will be helpful for the design
of the monitor and filter equipment in the bleed air system.
These tools has been partially validated by the experimentally or numerically through some simple structures with
different condition. However, such numerical tools need further validation with experiments.
Acknowledgements
The research presented in this paper was financially supported by the National Key Basic Research and
Development Program of China (the 973 Program) through grant No. 2012CB720100.
Reference
[1]Overfelt R.A., Jones B.W, Loo S.M., Haney R.L.,Sensors And Prognostics To Mitigate Bleed Air Contamination Events, Progress Report,
RITE-ACER-Coe-2012-05, 2012
[2]Zhu,G, Studies of the effects of ozone on human life, Science and Technology Information, 2007
[3]FAA, airworthiness standards: transport category airplanes, FAR part 25, 2014
[4]FAA, transport category airplanes cabin ozone concentrations, AC120-38, 1980.10
[5]Netten C, Aircraft Air Quality Incidents, Symptoms, Exposures and Possible Solutions, 2005
[6]Glenn C. M, William W. N, J. Alejandro C-R, Alfred T. H, Mark P. M, Indoor Air Quality Impacts of Ventilation Ducts: Ozone Removal and
Emissions of Volatile Organic Compounds, Journal of the Air & Waste Management Association, 2011
[7]Charles J. W, Helen C. S. The Influence of Ventilation on Reactions Among Indoor Pollutants: Modeling and Experimental Observations.
Indoor Air，2000
[8]Expert Panel on Aircraft Air Quality, Contamination of aircraft cabin air by bleed air – a review of the evidence , EPAAQ, 2009
[9]Yin J., Jenson M.K., Analytic model for transient heat exchanger response, International Journal of Heat Mass Transfer, Vol. 46, 2003,
pp.3255-3264
[10]Ansari M.R., and Mortazavi V., Simulation of dynamical response of a countercurrent heat exchanger to inlet temperature or mass flow rate
change, Applied Thermal Engineering, Vol. 26, pp.2401-2408, 2006.
[11]Galeazzo F.C.C., Miura R.Y., J.A.W. Gut, Experimental and numerical heat transfer in a plate heat exchanger. Chemical Engineering
Science. Vol. 61, pp. 7133-7138, 2006.
[12]Zhang L., Yang C., dynamic performance of compact heat exchangers under dry/wet conditions based on 3d distributed parameter model,
Sixth International Conference on Enhanced, Compact and Ultra-Compact Heat Exchangers: Science, Engineering and Technology, 2007
[13]YANG M, KE P, ZHANG S, Preliminary investigation of the effect of aircraft ECS to the bleed air contamination: numerical tool
development, ICES-2014-100, 44th International Conference on Environmental Systems, 13-17 July 2014, Tucson, Arizona