Building and Environment 48 (2012) 128e134

Contents lists available at SciVerse ScienceDirect

Building and Environment

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

Development of solid desiccant dehumidification using electro-osmosis

regeneration method for HVAC application
B. Li, Q.Y. Lin, Y.Y. Yan*
Energy & Sustainability Research Division, Faculty of Engineering, University of Nottingham, Lenton Firs Building, University Park, Nottingham NG7 2RD, UK

a r t i c l e i n f o a b s t r a c t

Article history: In this paper a novel method of electro-osmosis (EO) based regeneration for solid desiccant is reported,
Received 9 May 2011 and the potential of the method for application in HVAC particularly for dehumidification process in air
Received in revised form conditioning system is discussed. The results of the measurement of water removal rate inside solid
5 September 2011
desiccant array show that the mass flow rate of EO driven mass flow is achievable at 0.953 g m2 s
1 with
Accepted 7 September 2011
20 V DC voltage on the installed electrodes. By comparing the theoretical and experimental results, the
current limitations of theory and experimental apparatus were explored. The experimental results show
Keywords:
that the energy consumption in EO integrated air conditioning system is averagely 23.3% lower than the
Solid desiccant dehumidification
Electro-osmosis
conventional air conditioning system in respect of the different configurations in air handling process. In
HVAC the case study, among the three potential locations for EO device, the position in Point C is the best
Microporous media Latent heat recovery choice with an ideal COP of 12.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Although latent heat counts less ratio of total load from ambient
normally, but it is relatively difficult to remove them effectively in
Among the energy consumptions in the buildings, HVAC the process due to the phase change and heat exhausting from it.
systems currently cover about 30e50% of total amount consump- Moreover, considering the indoor air quality and energy use issues,
tion. Due to higher demand of thermal comfort inside building and which significantly influence the application of HVAC system, the
larger energy consumptions within air conditioning system, the problems from high humidity need to be re-considered as around
improvements of the whole system using less energy and being 80% of the dehumidification load comes from outside fresh air and
more efficient is crucial to achieve sustainability in building service it can affect the performance of the air conditioning system
sector. Since the 1990s, the air conditioning industry has been faced dominantly [6]. Hence, dealing with the latent cooling load, which
with a number of challenges including the increased energy effi- can be commonly achieved by dehumidification process, is a critical
ciencies, the improved indoor air quality, the growing concern for and decisive stage in air conditioning process that can largely
improved comfort and environmental control, the increased influence the whole energy consumption. In commercial buildings,
ventilation requirements, the reduction of chlorofluorocarbons the traditional way to control the humidity is to cool the air down
(CFCs), and the rising peak demand charges [1,2]. New approaches to dew point temperature till condensation and then reheat them
to space conditioning will be required urgently to resolve these to the design temperature. It is obviously that the reheating process
economic, environmental, and regulatory issues [3e5]. To achieve in the cycle costs large amount of energy inevitably. The industry is,
cooling purpose, the cooling coil is supplied by vapour compression however, quickly learning that by controlling the humidity to 50%
cycle using refrigerant. In conventional air conditioning system, the RH or less, and independent of temperature, dry bulb temperature
cooling capacity needs to match the cooling load, which consists of can be raised to 25e26  C in consideration of thermal comfort,
sensible heat and latent heat generated by occupants and some yielding great energy savings in the meantime. For every degree of
facilities. A common commercial air conditioning system mainly the set points in temperature is raised, a 5% reduction in air
consists of a cooling coil and heating coil with supply/extract fans conditioning energy is realized [7]. However, both solid desiccant
as shown in Fig. 1. and liquid desiccant dehumidification techniques in existing
industrial application require considerable heat and electricity to
operate the regeneration process which is less competitive in
* Corresponding author. Tel.: þ44 0115 846 7263. domestic dwellings and office buildings [8]. Hence, the develop-
E-mail address: yuying.yan@nottingham.ac.uk (Y.Y. Yan). ment of dehumidification techniques, in retrospect, is urgently to

0360-1323/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.buildenv.2011.09.008
 B. Li et al. / Building and Environment 48 (2012) 128e134 129

applied electric field [22]. It has also been used to characterize and
Nomenclature design the salt rejection properties of reverse osmosis membranes
and to help understand the behaviour of biological membranes. In
D Diameter of the cylindrical container this paper, the EOF theory has been used to guide the regenerate
L The length of EO process solid desiccant by regarding desiccant structure as micro-channel
veo Average velocity of EO flow and applying electric field between the desiccants membrane.
3 Electric permittivity The moisture adsorbed by desiccant can be pumped out in a certain
z Zeta potential rate.
h Dynamic viscosity Generally, electro-osmosis flow is a flow in porous media or
Df Electrical potential drop micro-channel structure, driven by so-called EO force with applied
E Electric field electrical field. The EOF dynamics in porous media has been studied
mion Ionic mobility much both theoretically and numerically. Most of papers are based
meo Electro-osmosis mobility on the theory of double layer model which was developed by Stern
j Porosity and Guoy-Chapman, respectively. For most EO flow, an electric
_ eo
m Mass flow rate for EO driven flow per m2 double layer (EDL) will be formed spontaneously at the liquid/solid
rwater Density of water interface due to the electrochemical reaction. As depicted in Fig. 2,
the charged surface will attract positive ions present in the water
and repel negative ions owing to the polarization effect of porous
be expended in domestic application as replacement of current media and this region at liquid/solid interface is the so called EDL.
products in the market [9]. Particularly, in some extremely hot and In the inner layer, the counter ions near the wall are called the Stern
humid climate area, the current existing technology cannot meet layer while the outer diffuse layer is called Gouy-Chapman layer
the comfort requirements due to the humidity removal capacity of [23]. The positive ions will therefore predominate in the Debye
the system, and the crystallization, corrosion problems of haloids sheath next to the charged surface so application of an external
on the wheel or desiccant pool [10]. Therefore, the study on EO electric field results in a net migration towards the cathode of ions
dehumidification method will be the first of its kind in domestic in the surface water layer. Due to viscous drag, the water in pores is
application and definitely it will keep the industry sustainable. drawn by ions and therefore flows through the porous media. The
Based on our reviews, with respect to this novel application, the schematic diagram of EOF in porous media and its velocity profile
system design is a dominant factor for validating the electro- has been shown in Fig. 3.
chemical phenomenon. Qi et al. designed an EO system and used In EO flow, the process will gradually become weaker due to
desiccant powders in a humid environmental chamber to achieve several reasons. One of the most significant reasons is the corrosion
dehumidification [11,12]. On their system, the flow rate of water in electrodes and the process will be interrupted when electrodes
removal was measured at 0.0021 g s
1. Although DC pulse was are fully corroded and the pH of the flow will also change. This
applied into the application to reduce the Joule effect and erosions occurs due to the oxidation and reduction reactions at the anode
on the electrodes, there are still some obstacles to obtain higher and cathode [21]. The anode will become acidic as carbon acid is
capacity and longer durability which were dominant by the system formed by the reaction due to carbon dioxide with water and
design and configuration. Hence, a more dedicated system design is cathode becomes basic. Due to the movement of ions from anode to
needed to precisely and simultaneously measure the actual EO cathode, as the pH become more basic, some composites like
performance. complex calcium precipitates will be formed and these composites
The electro-osmotic flow (EOF) theory with a historical retro- will fill the voids along the porous material, which will significantly
spect based on the electrokinetic effects is observed by F.F. Reuss in reduce the EO flow and finally interrupt the flow. For most EO flow,
1809 when doing an experimental investigation on porous clay, and it is influenced by several forces such as gravity, hydraulic pressure,
it has been currently widely studied and analyzed in areas such as viscosity, temperature driven force and EO but the hydrostatic
electrochemistry, physics and vascular plant biology [13e16]. Due pressure and electro-osmotic force are dominant components.
to its importance, EOF theory has currently been applied in various Therefore, the total flow rate becomes the sum of EO flow and
fields such as chemical engineering [17,18] and civil engineering hydrodynamic flow [24].
including mine tailing water prevention and waste sludge treat- As one of the popular alternatives dealing with separate
ment [19e21]. Basically, EOF is the motion of polar liquid passing humidity control, the desiccant system which has a large potential
through porous structure, micro-channel or other membrane under market including applications in schools, restaurants, hotels and

Fig. 1. Schematic diagram for a conventional air conditioning system.

130 B. Li et al. / Building and Environment 48 (2012) 128e134

now available. These existing desiccant systems are not very effi-
cient and quite simple in terms of function and operational range in
that they use rejected heat from the condenser to regenerate the
desiccant wheel [1]. Although the heat source can be supplied using
renewable energy such as solar thermal, the energy requirement
still makes the whole system less competitive. Therefore, the novel
idea of using EO theory to regenerate solid desiccant has been
technically addressed here. Based on this new idea, the EO
phenomena in nature and the applications in Micro-
electromechanical systems (MEMS) fields have been conclusively
reviewed. It is also promising that applies this technique into
a larger scale like dehumidification in buildings as it consumes low
DC voltage which can be easily fetched through various means.
Hence, this can be potentially improved by using a new regener-
ating method: Electro-osmosis regeneration method instead of
conventional heat source mentioned above. The general principle
of EOF in air conditioning application is that the moisture can be
Fig. 2. Charge distribution and water affinity form at interface with solid desiccants.
absorbed by a membrane composed of a desiccant, being removed
across the membrane and rejected on the other side by electro-
osmotic pumping. Based on electro-osmotic counter pressure
food stores as it can significantly reduce operational costs, improve concept, the liquid flow occurs in the direction of the electric field.
indoor air quality and achieve precise humidity control [25]. Positive ions move towards cathode, dragging water molecules.
However, the current desiccant cooling systems have not been used
widely as it should because its efficiency is still below break-even.
2. Experimental setup and method
The energy invested in running the system was more than the
traditional air conditioning system. And it is this reason that may
In order to evaluate the performance of EO process on regen-
lead to a narrow market and less popular than existing techniques.
erating solid desiccant and optimise our design and configuration,
Unlike traditional vapour compression cooling systems which are
an experiment is carried out using Zeolite as desiccant forming the
driven by electrical power to cool and dehumidify air, desiccant
micro-channel structure. To achieve precise measurement and
systems, instead, use thermal energy to accomplish the same effect.
avoid vapour fluctuating from testing chamber as some investiga-
Generally, a desiccant cooling system may use either solid or liquid
tors carried in the past, the separation of generated bubble and the
desiccant to dry the air by adsorbing moisture. And then this air is
actual contact area are reinvestigated [14]. Fig. 4 shows the
cooled down by evaporative cooling method and sent to the air-
improved schematic diagram of the EO regeneration system. The
conditioned space finally. The saturated desiccant is usually re-
apparatus includes the EO device, the electronic scale and a data
heated by gas or electric heating to release moisture, which is
logger. Zeolite has been filled into the EO device, a simple cylin-
defined as the conventional regeneration method and prevailing in
drical container (D 16 mm  L 12 mm) between two electrodes
the current HVAC market [26].
rings. Two pieces of filter paper are used between the electrodes in
The rapid development of desiccant air conditioning technology,
both sides to block leakage from the container. Considering the
which can handle sensible and latent heat loads independently, has
thickness of the electrodes and filter paper, the actual length for EO
expanded desiccant applications into hospitals, supermarkets,
process is set as 10 mm. The water used in the experiment is de-
restaurants, theatres, schools, office buildings, stores or even agri-
cultural related but it still consumes a large amount of energy for
reactivation of desiccants [27e31]. It should be noted that whereas
the reactivation heat in the original desiccant systems was
produced by burning natural gas, all-electric desiccant systems are

Fig. 3. Principle of EOF. Fig. 4. Schematic diagram of the EO regeneration system.

 B. Li et al. / Building and Environment 48 (2012) 128e134 131

ionized water to make the ions concentration at low level and keep applying 20 V DC, simultaneously extra droplets are created and
the control volume unaffected by the outside changes of concen- the mass of collected water is ramped up to a rate of 0.0154 g/min
tration especially for the charges from solution. over a period of 70 min. As the flow rate created by gravitational
To begin with the experiment, water was continuously adsorbed force is 0.038 g/min, then the result due to electro-osmosis process
by the Zeolite before applying the electric field and the mass of is calculated as 0.0115 g/min by deducting flow rate value caused by
droplet at the bottom was collected and measured by the electric gravitational force factor from the total flow rate recorded here. By
weight scale to evaluate the effect caused by gravity. When mass specifying the flow density of electro-osmosis, it is easily calculated
flow reached stable and Zeolite was saturated, 20 V DC voltage is as 0.953 g m2 s
1 which is divided by the actual testing volume
applied and the mass of droplet is recorded by a data logger. from electro-osmosis flow rate in the pump. In fact, the effect of EO
Theoretically, the mass flow rate can be calculated by multi- pumping force is significant from the results and it shows a great
plying the theoretical total volume flow rate driven by EO process potential to regenerate desiccant.
with water density, where the total volume flow rate is the sum of Theoretically, the mass flow rate value can be calculated by
flow rate in all the micro-channels formed by Zeolite particles. Due multiplying the theoretical total volume flow rate driven by EO
to the complexity of the actual micro-channel in porous medium process with water density, where the total volume flow rate is the
formed by the Zeolite particles and the irregular shape of each sum of flow rate in all the micro-channels formed by Zeolite
desiccant, an average EO velocity is calculated macroscopically particles. Due to the complexity of the actual micro-channel in
[32]; and the velocity driven by EO force, veo, is defined by the porous medium formed by the Zeolite particles and irregular shape
HelmholteSmoluchowski relation as: of each desiccant, an average EO velocity is calculated
macroscopically.
z
3 Df In the experiment, although the de-ionized water has been
veo ¼ E ¼ meo E ¼ meo (1)
h L used, there could still be a low ion concentration due to the ions in
solid particles are partly soluble in liquid. At room temperature, the
Therefore, considering the porosity of Zeolite, the velocity
mobility, mion, has been shown in Table 1 [32] and meo in the
equation could be adjusted as:
experiment is cited as 6.6  10
8 m2 (Vs)
1. Therefore, the theo-
z
3 Df retical mass flow rate for EO driven flow per m2 is calculated as:
veo ¼ j E ¼ jmeo (2)
h L
Df
_ eo ¼ rwater veo ¼ jrwater meo
m ¼ 19:8 g m2 s
1 (4)
Thus, the theoretical mass flow rate for EO driven flow per m2 L
becomes:
It is evident that the actual mass recorded in testing is much less
Df
_ eo ¼ rwater veo ¼ jrwater meo
m (3) than the theoretical value as calculated above. Some reasons may
L contribute to the discrepancy: The main reason is the bubbling
effect at both interfaces of the electrodes which continuously
3. Results and discussion generates small amount of hydrogen at cathode and oxygen at
anode during the testing. Not only these gases may block the inlet
Following the experimental procedure using the apparatus, the and outlet of functional EO pump but also create random voids
results for both cumulative collected mass of water and pure EO inside the solid particle channel which leads to the reduction of
regeneration rate have been shown in Fig. 5. contact area. Consequently, the actual DC voltage, which applied
As depicted in Fig. 6, at the beginning of the experiment, the on the solid desiccant, would vary significantly that affects the
mass of collected water is relatively slow, approximately 0.038 performance of solid desiccant regeneration. The solution to avoid
g/min. This may be contributed by the adsorption ability of Zeolite. this problem or relieve the erosion of the electrodes is that change
When the Zeolite is saturated, the water vapour is condensed to the active metallic electrodes to the carbon fibre or graphite
small droplets which are collected by a container. According to the electrodes which exerts less ions during the testing. Alternatively,
results, the mass of the collected water is up to 0.0038 g/min. After it can be modified by releasing a fixed volume above the elec-
trode, which is controllable by valves, in order to give enough
space to the bubble floating up. Another critical reason is the
interaction of configuration inside the EO pump. It is found that
the best performance of regeneration happens when the filter is
installed at outside surface of the electrodes which means the
solid particles are wrapped by the attached electrodes and
Perspex container. The filters are acted as the membranes that
keep the desiccant particles inside the container, and also the
contained water could be freely pumped outside. Currently, the
optimal design is undergoing that identifies the interactions
between membrane and electrodes. The last possible reason is
that the filter meshing size. As the Zeolite is synthetic (non-
polarised) and purified (dealumination) at size below 45 nm
which could challenge our and membrane in use, it is suggested
that the more dedicated filter should be applied to maximally
avoid the contents loss during operation.
However, in our testing, it is still encouraging that the actual
removal rate by our prototype is acceptable for the threshold of
solid desiccant dehumidification. Hence, a specific evaluation is
carried out in next chapter for assessing the feasibility of this
Fig. 5. Change of mass of collected water. testing result in real air conditioning application.
132 B. Li et al. / Building and Environment 48 (2012) 128e134

Fig. 6. Psychrometric cooling processes for (a) Conventional system. (b) EO (L1). (c) EO (L2). (d) EO (L3).

4. EO integrated air conditioning system regenerating solid desiccant continuously which replace the heat
source used for regeneration in conventional solid desiccant cool-
4.1. Analysis on EO process in air conditioning application ing system. Besides this, the simple configuration without moving
parts makes it better to fit for various requirements. Different from
As stated above, the energy consumption in air conditioning the reheating process of the desiccant wheel, the desiccant
system is to overcome the cooling load involving two major factors. powders in the EO unit is continuously retreated by the DC power
One is the sensible cooling load which affects the change of dry which means there is a maintained mass gradient between the
bulb temperature; the other is the latent cooling load which fluc- circumstance and inside the EO unit. Hence, the regeneration
tuates with the change of moisture content generated by metabolic method needs no rotary wheel and motor. In EO integrated air
activity of human body and ambient inherent humid air. Taking the conditioning system, the air can be dehumidified directly and the
advantage of the specific characteristic of EO phenomena, it already process can be assumed as isenthalpic change. Due to the EO
has been approved that the great potential application in dehumidification effect, the moisture can be conditioned to
a certain level before being supplied into the room. Generally, there
Table 1 are three possible locations for the EO dehumidification device,
Ionic mobility for small ions at small concentration. namely EO (L1) located in Point C, EO (L2) located in Point W and EO
(L3) in Point L.
Agþ Kþ Liþ Naþ Br
Cl
F
I

At stage of EO (L1), the mixed air passes through the dehu-
Mobility 6.42 7.62 4.01 5.19 8.09 7.91 5.7 7.96
10
8 m2(Vs)
1
midification device and reaches to Point C1 where the moisture
content is the same as Point O (L). It means that the mixed air at
 B. Li et al. / Building and Environment 48 (2012) 128e134 133

Point C1 could be cooled directly to Point O (L) without super system and modified system using EO dehumidification device in
cooling to dew point to obtain the moisture by conventional three locations. The outside conditions are set as: 35  C dry bulb
condensation process using chilled water. As well known, it costs temperature (tw), 60 % relatively humidity (RHw%) and 101,325 Pa
large amount of energy consumption in super-cooling process for pressure. Then the sensible cooling load (Qs) is 11.3 kW and the
resulting from the occurrence of phase change heat. However, total number of occupants is 60 with latent heat rate 45 W/person.
scenes are different in EO (L2) process that the fresh air is condi- Based on the functions of the building, the design condition can be
tioned separately before being mixed with returned air while the determined based on CIBSE Guide A [33]. Hence, the design room
following processes are the same as EO (L1). If EO process is set at temperature (TN) is 24  0.5  C and design relative humidity (RHN%)
point L3, the mixed air is designed to be conditioned to the dew is 55%. The required fresh air flow rate is set as 10 L/s per person. As
point temperature and then it is dehumidified within the EO the allowed error in the room is 0.5  C, the temperature difference
dehumidification device. At this stage, most of moisture in terms of between the supply air temperature and room design temperature
the latent load in supply air is removed that the condensate is is set as DTo ¼ 5  C. Therefore, the supply air temperature is
collected and repelled to the outside system. All these processes To ¼ 24
5 ¼ 19  C. The fresh air ratio is set at 15% from the total
have been shown in Fig. 6 volume of air supply at mixing point C.
Based on above analysis, the energy consumptions for each
configuration including conventional air conditioning system and 4.2.2. Energy performance evaluation
modified EO systems in different locations were concluded. According to Table 2 the detailed calculation for different situ-
Summarily, it mainly consists of cooling process and reheating ations can be carried out. Different systems integrated with EO
process to achieve accurate temperature control and thermal regeneration method with different locations have been compared
comfort condition. The energy consumption for EO integrated with conventional air conditioning system using vapour compres-
systems is mainly the combination results of adsorption process in sion cycle for cooling.
solid desiccant, regeneration process regarding to power It is distinctive that air conditioning system with EO dehumid-
consumption of EO electric supply and sensible cooling process in ification device is more efficient based on our calculations, which
terms of conventional air conditioning cycle. means that EO dehumidification method has brighter application in
The total energy consumption can be calculated using equations air conditioning technology. In this particular case, the percentage
shown in Table 2. Theoretically, the system using EO should be of energy saving is achieved as (54
41.4)/54¼23.3%.
more energy efficient. However, the actual location of the EO device It can be also seen that the electricity power consumption to
can also be reviewed in respect that complex configurations regenerate desiccant in EO (L2) process is slightly lower than EO
existing in reality. At EO (L2) stage, there is a noticeable change in (L1) process. However it could still be better due to the limit of
moisture content during the process W
W2 although the flow regeneration rate using EO and operational temperature. As the air
rate of the air is quite small. As the air velocity in air handling unit is velocity in air handling unit is relatively high, it may be difficult to
relatively high and the dehumidification capacity using EO is not achieve large moisture change in actual situation. In EO (L2)
relatively low, the huge change regarding to the moisture may not process, there is a large change of moisture content of 0.0122 kg/kg
be achieved completely in actual situation. Moreover, at stage of EO while the moisture content change in EO (L1) process is around
(L2), the dry bulb temperature of the air may rise significantly when 0.0034 kg/kg. Previous experimental result shows that the mois-
the moisture is supplied into desiccant arrays. It is noticed from the ture removal rate is around 0.953 g m2 s
1. The required EO device
Psychrometric chart that the increase of temperature is much contact area is nearly the same for EO (L2): 7.68 m2 and EO (L1):
higher than the location at EO (L1). Therefore, EO (L1) location 8.03 m2 in respectively. However, larger change of moisture content
seems more suitable choice for configuration of EO dehumidifica- means higher moisture removal rate for EO process with similar
tion device but actual performance needs to be validated in future. contact area. Hence, EO (L1) should be better than EO (L2) because
Last but not least, the comparison between EO (L1) and EO (L3) can it requires a smaller regeneration rate and can be easily achieved by
be carried out by analysing the working conditions for compressors EO regeneration process. In addition, in EO (L2) process, the dry
which could be different in terms of COP using conventional vapour bulb temperature of the air may rise to around as high as 65  C due
compression refrigerant cycle with different refrigerant through to phase change heat from condensation when the moisture in the
case study. air has been absorbed into desiccant, whereas the temperature for
EO (L1) may only rise from 27  C to 35.5  C. Conclusively, EO (L1) is
4.2. Case study more competitive with lower requirement of regeneration rate and
operational temperature. Comparing EO (L1) with EO (L3), the EO
4.2.1. Design condition (L3) process still stands out although the calculated energy inputs
In this section, an air conditioning system for a commercial are the same. Through the calculation of COP on R 134a and
building is designed to meet thermal comfort condition in summer ammonia, the ideal COP for compressor in EO (L3) is higher than EO
incorporating two systems virtually: conventional air conditioning (L1) and this finally leads to the reduction of energy input for the
compressor in that evaporative temperature is increased slightly
Table 2
due to EO dehumidification effect. However, further calculation and
Energy consumption for different systems. analysis needs to be done to evaluate the actual performance and
difference between EO (L1) and EO (L3).
Conventional air conditioning system EO (L1)
Process Energy input Process Energy input
In addition, the application of EO regeneration method has fairly
Cooling C
L _ c
hL Þ
QC ¼ mðh EO C
C1 Qelec1 positive effects on the basis of conventional desiccant cooling
Reheat L
O _ o
hL Þ
QR ¼ mðh _ c1
ho Þ
Cooling C1
O (L) QC1 ¼ mðh system. EO integrated air conditioning systems can reduce the
Total QAC ¼ QC þ QR Total QEO1 ¼ Qelec1 þ QC2 energy consumption (Table 3) of fossil fuels due to energy saving
EO (L2) EO (L3) with small electricity consumption for EO regeneration process
Process Energy input Process Energy input using simple low voltage DC supply rather than high voltage AC
EO W
W2 Qelec2 Cooling C
L3 _ c
hL3 Þ
QC3 ¼ mðh supply which means various energy sources can be alternatively
Cooling C2
O (L) _ c2
ho Þ
QC2 ¼ mðh EO L3
O Qelec3
Total QEO2 ¼ Qelec2 þ QC2 Total QEO3 ¼ Qelec3 þ QC3
used instead of heat source from the low grade energy form.
Moreover, all the process can be operated at normal temperature,
134 B. Li et al. / Building and Environment 48 (2012) 128e134

Table 3
Energy consumption for different configuration by EO integration.

Conventional EO (L1) EO (L2) EO (L3)

Air temperature 27.0e16.5  C 35.5e19.0  C 35.5e19.0  C 27.5e16.5  C
Refrigerant temperature 32.0e11.5  C 40.5e14.0  C 40.5e14.0  C 32.5e11.5  C
Cooling (QC) QC ¼ 46.125kW QC1 ¼ 38.25kW QC2 ¼ 38.25kW QC3 ¼ 38.25kW
Reheat/EO (QR/Qelec) QR ¼ 7.875kW Qelec1 ¼ 3.24kW Qelec2 ¼ 3.13kW Qelec3 ¼ 3.24kW
Total QAC ¼ 54kW QEO1 ¼ 41.49kW QEO2 ¼ 41.38kW QEO3 ¼ 41.49kW
COP R 134a 12.19 9.25 9.25 12.03
Ammonia 12.30 9.48 9.48 12.04

EO regeneration method can also partly avoid using Chlorinated [8] Yu BF, Hu ZB, Liu M, Yang HL, Kong QX, Liu YH. Review of research on air-
conditioning systems and indoor air quality control for human health. Inter-
Fluorocarbon Compounds (CFCs) which has removed the latent
national Journal of Refrigeration 2009;32:3e20.
cooling load and the system only needs to overcome the sensible [9] Ghesmat K, Hassanzadeh H, Abedi J. The impact of geochemistry on convec-
cooling load. This could further reduce the size of compressors, tive mixing in a gravitationally unstable diffusive boundary layer in porous
fans, ductworks, etc. and finally reduce the capital cost of the media: CO2 storage in saline aquifers. Journal of Fluid Mechanics 2011;673:
480e512.
system. In future, the modified standalone system can control the [10] Hao X, Zhang G, Chen Y, Zou S, Moschandreas DJ. A combined system of
humidity of air independently of its temperature to achieve better chilled ceiling, displacement ventilation and desiccant dehumidification.
comfort level especially for those high humid climate zones where Building and Environment 2007;42:3298e308.
[11] Qi R, Tian C, Shao S. Experimental investigation on possibility of electro-
normal cooling system may fail to tackle with large amount of osmotic regeneration for solid desiccant. Applied Energy 2010;87:
latent heat from outside environment. 2266e72.
[12] Qi R, Tian C, Shao S, Tang M, Lu L. Experimental investigation on performance
improvement of electro-osmotic regeneration for solid desiccant. Applied
5. Conclusions Energy 2011;88:2816e23.
[13] Yan YY, Hall JB. The concept of electroosmotically driven flow and its appli-
Conclusively, this paper has introduced the development of cation to biomimetics. Journal of Bionic Engineering 2004;1(1):46e52.
[14] Yan YY, Zu YQ, Ren LQ, Li JQ. Numerical modelling of electroosmotically
a novel EO based regeneration method for solid desiccant which driven flow within the micro thin liquid layer near an earthworm surface e
has shown a great potential for application in HVAC particularly for a biomimetic approachProceedings of the Institution of Mechanical Engineers,
dehumidification process. Based on the present studying, it has Part C. Journal of Mechanical Engineering Science; 2007:1201e10.
[15] Wang MR, Chen SY. Electroosmosis in homogeneously charged micro- and
been found that the EO mass flow rate is achievable at nanoscale random porous media. Journal of Colloid and Interface Science
0.953 g m2 s
1. By comparing the theoretical and experimental 2007;314:264e73.
results, the current limitations of theory and experimental appa- [16] Zu YQ, Yan Y. Numerical simulation of electroosmotic flow near earthworm
surface. Journal of Bionic Engineering 2006;3:179e86.
ratus have been explored, and the reasons for constraining the [17] Bertolini L, Coppola L, Gastaldi M, Redaelli E. Electroosmotic transport in
development of EO performance, such as bubbling effect and porous construction materials and dehumidification of masonry. Construction
electrode erosion, have been identified. It is predictable that the and Building Materials 2009;23:254e63.
[18] Neve RS, Yan YY. Enhancement of heat exchanger performance using
introduction of graphite electrodes or carbon fibre electrodes will combined electrohydrodynamic and passive methods. International Journal of
make the operation more durable. Furthermore, the progress in Heat and Fluid Flow 1996;17(4):403e9.
fined meshing of filter will promote the EO pump system efficiently. [19] Mahmoud A, Olivier J, Vaxelaire J, Hoadley AFA. Electrical field: A historical
review of its application and contributions in wastewater sludge dewatering.
In the last part, comparing different configuration feasibilities in air
Water Research 2010;44:2381e407.
handling unit, the energy consumption for EO integrated air [20] Raats MHM, Van Diemen AJG, Lavèn J, Stein HN. Full scale electrokinetic
conditioning system has been evaluated with reference to the dewatering of waste sludge. Colloids and Surfaces A: Physicochemical and
testing result. A case study with detailed designs of using different Engineering Aspects 2002;210:231e41.
[21] Morefield SW, McInerney MK, Hock VF, Marshall OS. Interface modeling for
systems involved in both conventional air conditioning systems electro-osmosis in subgrade structures; 2004.
and the EO integrated systems (with different locations of the EO [22] David WG. Electro-osmosis for dehumidification. School of Mechanical Engi-
device) has been carried out. The case study shows that the neering, Georgia Institute of Technology; 2006.
[23] Zeng SL, Chen CH, Mikkelsen JC, Santiago JG. Fabrication and characterization
adoption of the EO regeneration method can offer higher energy of electroosmotic micropumps. Sensors and Actuators B: Chemical 2001;79:
efficiency by reducing 23% energy consumption. 107e14.
[24] Li Yan. Solid desiccant dehumidification techniques inspired from natural
electroosmosis phenomena. Journal of Bionic Engineering 2011;8:90e7.
References [25] Mei VC, Chen FC, Lavan Z, Collier RK, Meckler G. An assessment of desiccant
cooling and dehumidification technoloy; 1992.
[1] Pesaran AA, Penney TT, Czanderna AW. Desiccant cooling state-of-the-art [26] Harriman L. In: The ASHRAE guide for buildings in hot and humid climates.
assessment. National Renewable Energy Laboratory, A Division of Midwest 2nd ed. ASHRAE; 2009.
Research Institute; 1992. [27] Calton DS. Application of a desiccant cooling system to supermarkets. ASHRAE
[2] Ma Z, Wang S. Building energy research in Hong Kong: a review. Renewable Transactions 1985;91(1B):441e6.
and Sustainable Energy Reviews 2009;13:1870e83. [28] Burns PR, Mitchell RB, Bechman WA. Hybrid desiccant cooling systems in
[3] Dorer V, Weber A. Energy and CO2 emissions performance assessment of supermarket applications. ASHRAE Transactions 1985;91(1B):457e68.
residential micro-cogeneration systems with dynamic whole-building simu- [29] Mazzei P, Minichiello F, Palma D. HVAC dehumidification systems for
lation programs. Energy Conversion and Management 2009;50:648e57. thermal comfort: a critical review. Applied Thermal Engineering 2005;25:
[4] Diaconu BM. Thermal energy savings in buildings with PCM-enhanced 677e707.
envelope: influence of occupancy pattern and ventilation. Energy and Build- [30] Bellia L, Mazzei P, Minichiello F, Palma D. Air conditioning systems with
ings 2011;43:101e7. desiccant wheel for Italian climates. International Journal on Architectural
[5] Beerepoot M, Beerepoot N. Government regulation as an impetus for inno- Science 2000;4:193e213.
vation: evidence from energy performance regulation in the Dutch residential [31] Mei L, Dai Y. A technical review on use of liquid-desiccant dehumidification
building sector. Energy Policy 2007;35:4812e25. for air-conditioning application. Renewable and Sustainable Energy Reviews
[6] La D, Dai YJ, Li Y, Wang RZ, Ge TS. Technical development of rotary desiccant 2008;12:662e89.
dehumidification and air conditioning: a review. Renewable and Sustainable [32] Bruus H. Theoretical microfluidics. Oxford University Press; 2008.
Energy Reviews 2010;14:130e47. [33] CIBSE. CIBSE guide A, environmental design. Norwich, Norfolk NR6 6SA, Great
[7] Daou K, Wang R, Xia Z. Desiccant cooling air conditioning: a review. Renew- Britain/London: Page Bros. (Norwich) Ltd./The Chartered Institution of
able and Sustainable Energy Reviews 2006;10:55e77. Building Services Engineers; 2006.