Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations

Energy xxx (2017) 1e8

Contents lists available at ScienceDirect

Energy journal homepage: www.elsevier.com/locate/energy

Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations Seung Jin Oh a, Kim Choon Ng b, Wongee Chun c, Kian Jon Ernest Chua a, * a

Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117576, Singapore Water Desalination and Reuse Centre, King Abdullah University of Science & Technology, Thuwal 23955-6900, Saudi Arabia c Department of Nuclear and Energy Engineering, Jeju National University, 1 Ara 1 dong, Jeju 63243, South Korea b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 September 2016 Received in revised form 31 January 2017 Accepted 11 February 2017 Available online xxx

This paper presents the evaluation of a solid desiccant dehumidifier equipped with adsorbent powder coated heat exchangers (PCHX). The main component of the solid desiccant dehumidifier includes two heat exchangers that are coated with silica gel RD type powders in order to increase water adsorption uptake by improving its heat and mass transfer. A series of experiment are conducted to evaluate two key performance indices, namely, moisture removal capacity (MRC) and thermal coefficient performance (COPth), under various hot and humid air conditions. Conventional granular adsorbent packed heat exchangers (GPHX) are employed to benchmark the performance of the adsorbent coated heat exchanger (PCHX). Results reveal that the PCHX exhibits higher uptake performance due to better heat and mass transfer. It is found that the moisture removal capacity increases from 7.4 g/kg to 11.0 g/kg with air flow rates of 35 kg/h, resulting in the extended contact time of the water vapor. Experiments also demonstrate that the moisture removal capacity is highly affected by inlet air humidity ratio. In addition, marked improvement in COPth can be achieved by a lowered hot water regeneration temperature. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Dehumidifier Adsorbent coated heat exchanger Silica gel Moisture removal capacity Thermal coefficient performance

1. Introduction A humid tropical climate region is basically defined as a region where its monthly latent cooling load of outdoor air exceeds its monthly sensible cooling load. As a city-state with little land mass, the use of air-conditioning in Singapore has become a norm in its buildings. The electricity demands for cooling in building sectors accounted for 31 ± 2% of total electricity consumption in all sectors in Singapore [1e3]. Conventional mechanical vapor compression cycles (MVC) are widely used in order to overcome sensible loads and latent loads in air-conditioning systems. In a humid region, however, MVC typically operates in a way to lower supply air temperature below the dew-point temperature in order to remove high humidity, which leads to the waste of energy. Moreover, conventional air conditioning system utilizes refrigerants such as CFCs which are harmful to our ozone layer as well as they affect the respiratory systems of humans [4]. Desiccant dehumidification is one possible solution that can

* Corresponding author. E-mail address: [email protected] (K.J.E. Chua).

minimize the aforementioned problems via a de-coupling cooling concept that separates latent cooling from sensible cooling in the air handling units (AHUs) of buildings. Liquid desiccants have been used in industrial application since 1930s and they are more flexible in their application. They have lower pressure drop and lower regeneration temperature as compared to solid desiccants [5,6]. However, such liquid desiccant dehumidifications are also toxic and corrosive in nature, which renders them unsuitable for airconditioning application. Liquid desiccants were also found to have carryover effects. As such, with less energy input, solid desiccants are more compact and less subjected to corrosion and carryover effects A solid desiccant dehumidification system is typically classified into two types by packing method, namely, a fixed bed type and a rotary wheel type that is the most appropriate dehumidification type for air-conditioning application [7e10]. However, the fixed bed design has not been used well for air-conditioning system because it is required to stop the process once the desiccant have reached its saturation state [11]. The rotary wheel design is the most widely used in as HVAC systems. Some investigations have been carried out on novel hybrid cooling systems that incorporate a heat pump and a membrane

http://dx.doi.org/10.1016/j.energy.2017.02.169 0360-5442/© 2017 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Oh SJ, et al., Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations, Energy (2017), http://dx.doi.org/10.1016/j.energy.2017.02.169

2

S.J. Oh et al. / Energy xxx (2017) 1e8

technologies [12e17]. Chen et al. employed a heat pump as a heat source at 40e50
C to regenerate a composite desiccant wheel that was placed between an evaporator and a condenser to conduct a second stage of dehumidification [12]. A lower power consumption of 1.86 kW has been achieved with the hybrid configuration as compared to a conventional condensing or desiccant wheel. A heat pump has been also applied to a liquid desiccant dehumidification system by Xie et al. [14]. Authors developed theoretical models of key components and analyzed the system energy performance. They concluded that COP can be increased from 5.1 to 5.5 by adopting a multi-stage heat pump. Das et al. investigated the performance of membrane contractors for liquid desiccant cooling system [16]. Their simulation results showed that the performance of the contactors are significantly affected by the membrane properties such as porosity, pore size and thickness. Zhang et al. applied both a heat pump and a hollow fiber membrane to twostage liquid desiccant dehumidification system [17]. The proposed system consists of cross-flow hollow fiber membrane modules, which acts as two dehumidifiers and two regenerators, and a compression heat pump system. The authors concluded that the COP can be increased by about 20% under the typical hot and humid conditions. Recently, many studies have been focused on the development of desiccant materials to improve adsorption capacity and thus the performance of the solid-based desiccant dehumidification system [18e24]. In spite of the advancement of the rotary wheel configuration and the novel composite desiccant material, it was pointed out that the removal of adsorption heat generated during dehumidification was a critical issue to be solved for further improvement of a solid-based desiccant dehumidification [30,31]. The adsorption heat attenuates the adsorption capacity of the system, being away from the ideal isothermal system, which leads to more irreversibility loss, generating high entropy during the operation processes that severely affect the coefficient of performance of the system. Some researchers proposed the concept of adsorbentcoated cross-cooled heat exchanger [32e34]. Prior to this, a finned tube exchanger where the granular adsorbent was packed by a wire mesh was introduced in industrial applications such as adsorption chillers and adsorption desalination systems [35,36]. This granular adsorbent packed heat exchanger was also employed in the dehumidification system [37]. However, it was found that the performance hasn’t improved as much as expected due to the poor

contact not only between adsorbent and metal fins, but also among adsorbent solids [38], inducing very low heat transfer efficiency. Fig. 1 shows examples of the adsorbent coated heat exchanger with different materials and coating methods. Many investigation also have been carried out mathmatically and experimentally to study thermal coefficient of performance (COPth), defined in Eq. (3), of various dehumidification systems [39,40]. Typical COPth of solid desiccant dehumidification systems was found to varies from 0.02 to 0.67 depending on the operating conditions and desiccant materials. For a hybrid cooling system where the solid desiccant dehumidifier is incorporated with MVC systems, evaporative cooling systems or thermal driven chiller such as AB and AD chiller, their COP are expected to be above 1.0 [41e43]. The objective of this study is to improve the overall performance of a solid desiccant dehumidifier by incorporating the adsorbent coated heat exchangers. Its performance is judiciously investigated under different operating conditions.

2. Adsorbent powder and sorption characteristics Silica gel has relatively high moisture adsorption capacity and low regenerative temperature of 50
Ce90
C. Among various types of silica gel, type ‘RD’ silica gel, manufactured by Fuji Silysia Chemical Ltd., Japan, was selected as a desiccant for dehumidifying humid air because it exhibited higher uptake capacity under higher relative humidity in the range of 70%e90% when it was mixed with a binder material. Its thermophysical properties are summarized in Table 1 [35]. In the present work, we reproduced powder silica gel of 0.07 mm in diameter by using a grinder and a mesh sieve (no.200) since the manufacturer does not provide a micronized silica gel. The powder adsorbents allows for coating a thin layer onto the metal fins and evenly mixed powder with the binder. It is worth noticing that both the granular silica gel and powder silica gel have the identical thermophysical properties except for apparent diameter. Fig. 2 shows water equilibrium uptake on the silica gel type RD powder with respect to water vapor pressure and relative humidity in the temperature range of 25e80
C. The measurements were conducted based on the gravimetric method using the Aquadyne DVS analyzer manufactured by Quantachrome Instruments. It is observed from Fig. 2(a) that the equilibrium uptake is proportional

(a)

(b)

(c)

(d)

(e)

(f)

Fig. 1. Examples of the adsorbent coated heat exchanger with different desiccant materials and coating methods: (a) silica-gel coating with binder by dip-coating method [25], (b) in-situ crystallization of SAPO-34 zeolite [25], (c) SPAO-34 zoloite coating with binder by dip-coating method [26], (d) Li-modified desiccant coating by electrostatic spraying method [27] (e) AQSOA Z01 coating with binder by dip-coating method [28], and (f) silica-gel coating with binder by dip-coating method [29].

Please cite this article in press as: Oh SJ, et al., Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations, Energy (2017), http://dx.doi.org/10.1016/j.energy.2017.02.169

S.J. Oh et al. / Energy xxx (2017) 1e8 Table 1 Thermo-physical properties of silica gel [35]. Property

Value

Specific surface area (m2/g) Porous volume (ml/g) Average pore diameter (A) Apparent density (kg/m3) pH value Water content (wt%) Specific heat capacity (kJ/kg K) Thermal conductivity (W/mK) Mesh size

650 0.35 21 800 4.0 e 0.921 0.198 10e20

to the partial pressure of water vapor while it is inversely proportional to the temperature of adsorbent. It is noteworthy that the highest partial pressure of water vapor available in a tropical climate is 5.6 kPa under 35
C dry bulb temperature and 100% RH. In a typical active desiccant wheel (DW) where returned air is used for regeneration process, the temperature of regeneration air is set to about 80
C to lower its relative humidity, while the conditions of process air deepen on the conditions of outdoor air [44]. Therefore, relevant relative humidity conditions of process air and

(a)

3

regeneration air under a tropical climate are 80% (Tdb ¼ 35
C) and 3% (Tdb ¼ 80
C) respectively. Therefore, it can be seen from Fig. 2(b) that the difference of equilibrium uptake between the dehumidification and regeneration processes is 0.36752 kg/kg, which is the maximum possible water removal capacity of the type RD silica gel operating under the relative humidity conditions. Fig. 3 illustrates the microscopic photos of granular RD and powder RD taken by Field Emission Scanning Electron Microscopy (FESEM) method. The granular RD solids are clearly visualized at 55 times magnification, while powder RD is clearly visualized at 1500 times magnification. It has an average diameter of 800 mm while the powder form has an average diameter of 4 mm. 3. Experimental setup 3.1. Silica-gel coated heat exchanger Conventional granular adsorbent packed heat exchanger (GPHX) has been widely used for adsorption chillers and desalination plants as well as air-conditioning systems [26e28]. The major drawback of this packing method, however, is the poor contact not only between adsorbent and metal fins, but also among adsorbent solids, which induces very low heat transfer efficiency. For these reasons, we developed the new adsorbent coated heat exchanger by making a thin silica gel layer onto the fins of heat exchanger with a dip-coating method. To coat the fins of the heat exchanger with the fine silica-gel powder, Hydroxyethyl cellulose (HEC) was used as a binder since it exhibits the best binding force between the type RD silica-gel powder and the metal fins as well as it didn’t affect the adsorbate uptake capacity [24]. During the dip-coating process, the heat exchanger was mounted onto a rotation machine to ensure the uniform coating layer. After that, the heat exchanger was placed into an oven for curing at 120
C for 12 h. A total of 70 g of RD powder was coated on both sides of the aluminum fins with average coating thickness of 0.1 mm as shown in Fig. 4. The structure of the silica-gel powder also can be visualized at 50,000 times magnification as shown in Fig. 4. Experimental results of the powder adsorbent coted heat exchanger (PCHX) are benchmarked with the conventional granular adsorbent packed heat exchanger (GPHX). The finned tube heat exchanger has a dimension of 200 mm  150 mm x 22 mm, fin thickness 0.1 mm and spacing 1.5 mm, and 4 tube passes with tube diameter 9.5 mm. It is noteworthy that the masses of these two heat exchangers are not the same since the amount of adsorbent is different. 3.2. Description of experimental setup

(b) Fig. 2. Isothermal water-vapor sorption characteristics of silica gel type RD powder: (a) equilibrium uptake against different water vapor pressure and (b) equilibrium uptake against relative humidity at temperature from 25 to 80
C.

The solid desiccant dehumidification system, shown in Fig. 5 comprises an adsorption bed and a desorption bed operating in absorption and desorption modes alternatively. The bed comprises two heat exchangers embedded with type-RD silica gel. The heat exchangers act important role in both dehumidification and regeneration processes. It keeps the silica gel at lower temperature by removing the adsorption heat by cooling water flowing through heat exchanger coil during dehumidification, while it heats up the silica gel to release the water vapor during regeneration. The heat exchangers are installed at an angle of 30
in a V-shaped configuration in the direction of the air flow in order to not only increase the frontal interaction area between desiccant and humid air, but also allow the moist air to pass through with a low Reynolds number and thus laminar flow is developed as the air velocity decreases. In the case of desiccant wheels, airflow through the straight passages is laminar, but resistance tends to increases in proportion to wheel depth, increasing fan energy costs. The

Please cite this article in press as: Oh SJ, et al., Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations, Energy (2017), http://dx.doi.org/10.1016/j.energy.2017.02.169

4

S.J. Oh et al. / Energy xxx (2017) 1e8

(a)

dehumidification is completed by combining two consecutive processes within a given cycle time. These two processes refer to dehumidification (adsorption) and regeneration (desorption) processes. The experimental procedure is described as follows: outdoor air (state 1) is drawn into the system by fan 1, and flowed through the duct (state 2). If relative humidity of the air did not meet the preset experimental condition, it is then pass through the humidifier to uptake more water vapor (state 30 ). The humid process air is allowed to enter the chamber and contacted with two-heat exchangers (state 4). The process air is dehumidified and simultaneously heated by the adsorbed heat released by silica gel. During the dehumidification process, cooling water flowed through the tubes of the two exchangers in order to remove the released heat. Subsequently the warm dry process air is made to exit the system by fan 2 (state 5). Thereafter, the process air flowed through the bypass line during the switching period for which air from the conditioned space is mixed with the outdoor air to constitute regeneration (state 10 ). Hot water at the temperature of 50e80
C flowed through the tubes of the two exchangers in order to release the absorbed moisture from adsorbing silica gel. Once the process air is warmed and dried, it passed through the two heat exchangers, adsorbing the water vapor released by the silica gel. Finally, the hot humid outlet regenerative air is exhausted from the system by fan 2 (state 5) to complete one cycle. 3.3. Performance indices Two different performance indices were adopted to analyze the performance of the dehumidifier with the installed adsorbent coated heat exchangers. One of the indices is moisture removal capacity (MRC) while the other is thermal coefficient of performance (COPth). MRC is defined as the time average of the transient moisture removal during the dehumidification process (tad), and it can be calculated by the following equation:

MRC ¼

(b) Fig. 3. Field emission scanning electron microscopic (FESEM) photos of type RD silica gel granules (a) and powder (b).

1 tad

Ztad dWad dt

(1)

0

Transient moisture removal (dW) is defined by the following equation:

dW ¼ Wad;in  Wad;out

(2)

where Wad,in and Wad,out represent the inlet and outlet humidity ratio (kgwater vapor/kgdry air) of process air during dehumidification process, respectively. Thermal coefficient of performance (COPth) is defined as the latent cooling capacity (Qc) divided by the total input thermal energy (Qreg) employed for regeneration process, and it represents the overall system capacity and energy efficiency. The total input thermal energy is the thermal energy of hot water to heat up silica gel to release the moisture. In this work, indoor air is mixed with outdoor air to make the same condition as a returned air from the conditioned space. COPth can be computed using following equation:

COPth

Fig. 4. Pictorial views of the silica gel powder coated heat exchanger with the coating thickness of 0.1 mm.

  Ma Wad;in  Wad;out hv Qc   ¼ ¼ Qreg Mhw Cp;hw Thw;out  Thw;in

(3)

where Qc and Qreg represent the latent cooling capacity (kW) and regeneration heating power (kW), respectively. Ma and Mhw are the mass flow rate (kg/s) of the process/regeneration air and hot water respectively. hv is the latent heat of water (kJ/kg). Thw,out and Thw,in are the respective temperature (
C) of how water at inlet and outlet

Please cite this article in press as: Oh SJ, et al., Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations, Energy (2017), http://dx.doi.org/10.1016/j.energy.2017.02.169

S.J. Oh et al. / Energy xxx (2017) 1e8

5

Fig. 5. Schematic diagram of the solid desiccant dehumidifier incorporating two-adsorbent coated heat exchangers.

of the heat exchanger. Cp,hw is the specific heat (kJ/kg
C) of the hot water at constant pressure. Controllable experimental conditions in this study included the air flow rate, process air inlet dry-bulb temperature, process air inlet humidity ratio, and cooling water/hot water supply temperatures. During the regeneration process, the inlet air dry-bulb temperature and humidity ratio were kept close to 35
C and 12 g/kg. Cycle time was fixed at 300 s for all the cases. Table 2 shows the conditions for experimental investigation under humid tropical climate. As shown in table, the respective base line conditions were judiciously kept constant while other parameters were regulated.

3.4. Instrument Dry-bulb and wet-bulb temperatures at both inlet and outlet of the system were measured using 10 k thermistors (±0.15) with a 3-s interval time during the experiment. The air flow rates were monitored using electromagnetic flow meters (0.5% of reading). Measured data were stored in a computer using Agilent data acquisition system (34970A) and LabVIEW software. Relative humidity and the humidity ratio were calculated using both dry-andwet bulb temperatures.

4. Results and discussions The experimental results of the desiccant dehumidifier equipped with the adsorbent coated heat exchangers are analyzed under various ambient conditions and operating conditions.

Table 2 Operation conditions for experiment under hot and humid climate. Operation parameters

Baseline

Values

Process air inlet dry-bulb temperature [
C] Process air relative humidity [%] Process/regeneration air mass flow rate [kg/h] Cooling water inlet temperature [
C] Hot water inlet temperature [
C] Cycle time [sec]

30 80 55 30 80 300

28, 65, 35, 25, 50,

30, 75, 45, 30, 60,

32, 80, 55, 35 70,

34 85 65 80

Fig. 6. Dynamic variation of moisture water removal capacity of the solid desiccant dehumidifier equipped with GPHX and PCHX.

4.1. Dynamic dehumidification performance Fig. 6 demonstrates the dynamic the variation of moisture removal capacity during the dehumidification and regeneration processes in four cycles against time when the solid desiccant dehumidification system was equipped with two different adsorbent embedded heat exchangers. A constant elapsed time of 3 s exists between each point plotted. In this cycle, the average temperatures of the inlet process air and its regeneration process are 30
C and 35
C, respectively. The average temperatures of cooling and hot water are set at 30
C and 80
C respectively. During dehumidification process, the inlet humidity ratio was kept close to 20 g/kg while it was kept to close to 12 g/kg during regeneration process in order to maintain the outdoor and returned air conditions. Results show that the difference in humidity ratio between the inlet and outlet of the solid desiccant dehumidifier increases drastically within 60 s from the initial state to the peak state during dehumidification process. Subsequently, it decreases at a lower rate as the adsorbent reaches its saturation state. In the dehumidification process, the largest difference in humidity ratio reaches 0.0135 kg/kg for PCHX and 0.00726 kg/kg

Please cite this article in press as: Oh SJ, et al., Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations, Energy (2017), http://dx.doi.org/10.1016/j.energy.2017.02.169

6

S.J. Oh et al. / Energy xxx (2017) 1e8

for GPHX. For the regeneration process, the largest difference in humidity ratios of PCHX and GPHX reach 0.0233 kg/kg and 0.0107 kg/kg. Although GPHX contains silica gel 5 times more than PCHX, its adsorption/desorption capacity is 2 times lower than those of PCHX. It is also observed that the desorption rate is even faster than adsorption rate for the both cases, resulting in a shorter process time. Equilibrium represents a state in which the rate of adsorption of adsorbate (water vapor) onto the surface of the silica gel is exactly counterbalanced by the rate of desorption of liquid phase back into the gas phase. However, it is noteworthy that the desorption rate is a function of the adsorbent’s surface temperature. Therefore, as the surface temperature rises when the supply of hot water flows through the heat exchanger, the more water vapor is released at faster rate. Hence, the regeneration time becomes shorter at higher surface temperature. This explains the reason for supplying the cooling water to the heat exchanger to enable a cooler adsorbent that is capable of adsorbing more moisture than the desorbing process [40].

(a)

4.2. Performance indices The performance of the solid desiccant dehumidifier incorporating with two different adsorbent embedded heat exchangers were experimentally investigated and judiciously compared through a series of runs under various humid tropical climate conditions. Fig. 7(a) depicts the experimental results under different air flow rate while other parameters are fixed the same as the baseline conditions. It is found that as air flow rate increases from 35 kg/h to 60 kg/h, MRC of the GPHX and PCHX drops from 11.0 g/kg to 7.4 g/ kg and 3.75 to 3 g/kg respectively. COPth also drops from 0.41 to 0.34; and for GPHX 0.33 to 0.28. This is mainly attributed to the shortened residence time of the process air in between fins when the air flow rates are higher. Fig. 7(b) shows the effect of the dry-bulb temperature on the moisture removal capacity and COPth. The dry-bulb temperature varies from 28
C to 34
C while the other parameters are kept constant as in the baseline conditions. It is observed that both MRC and COPth of GPHX and PCHX are markedly affected by process air inlet dry-bulb temperature. This is mainly because the air carries more water vapor at higher temperature although the relative humidity remains the same. This is also applicable to the case of varying in relative humidity as depicted in Fig. 7(c). It can be found that MRC and COPth for the two heat exchangers increases greatly with RH varying from 65% to 85% under the constant dry-bulb temperature. MRC and COPth for PCHX increase about 4.0 g/kg and 0.12 respectively; and for GPHX about 1.7 g/kg and 0.12. While the dry-bulb temperature remaining constant, increasing the relative humidity implies the higher humidity ratio and thus higher driving force between the humid air and the adsorbent. Fig. 8 illustrates the effects of varying cooling water and hot water temperatures on the water vapor removal capacity and COPth of the dehumidifier. Cooling water temperature varies from 25
C to 35
C while the hot water temperature varies from 50
C to 80
C. The volume flow rate of cooling water and hot water are kept at 4 L/ min for both the dehumidification process and the regeneration process during all the experiment. For both heat exchangers, MRC and COPth decrease with the higher cooling water temperature due to the adsorption heat generation. It is clear from the adsorption isotherm of type RD silica gel shown in Fig. 2(a) that the uptake of water vapor onto silica gel decreases with increasing its temperature at the same water vapor pressure. The MRC of GPHX and PCHX decrease about 0.4 g/kg and 2.7 g/kg respectively. COPth of GPHX and PCHX also decrease by 2.5% and 10.5% respectively. It is noteworthy that although the MRC and COPth are the highest at 25
C, in

(b)

(c) Fig. 7. Effect of change in the process air inlet conditions on moisture removal capacity and COPth: (a) change in air flow rate (35, 45, 50 and 60 kg/h), (b) change in dry-bulb temperature (28, 30, 32 and 34
C), (c) change in relative humidity (65, 75, 80 and 85%).

practical sense, the cooling water of 25
C is not available from the cooling tower in Singapore since its temperature is unable to go below wet-bulb temperature (26.7
C). From Fig. 8(b), it is observed that a marked COPth improvement can be achieved by simply lowering the hot water temperature. As the hot water temperature decrease from 80
C to 50
C, the COPth

Please cite this article in press as: Oh SJ, et al., Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations, Energy (2017), http://dx.doi.org/10.1016/j.energy.2017.02.169

S.J. Oh et al. / Energy xxx (2017) 1e8

7

and investigated experimentally. Type RD silica gel powder was selected as desiccant material, and its sorption characteristic was analyzed with the gravimetric method. Micronized silica gel was reproduced and judiciously coated onto the heat exchangers’ fins for dehumidifying water vapor from humid process air. Two key performance indices (MRC and COPth) as well as the dynamic adsorption/desorption performance of the powder adsorbent coated heat exchanger (PCHX) were evaluated and compared with those of the granular adsorbent packed heat exchanger (GPHX) by a series of experiments under varying operating conditions. The following findings can be inferred from this study:

(a)

(1) Silica gel type RD powder has a high equilibrium water vapor uptake of about 0.4under higher relative humidity spanning 80%e90%, and thus it is appropriate to operate an efficient dehumidifier as far as tropical climate is concerned. (2) Although the amount of adsorbent used in PCHX is almost five times less, its adsorption capacity is almost two times higher when compared with GPHX. (3) MRC and COPth of the two heat exchangers decrease at higher air flow rates due to a shortened residence time of the water vapor, while MRC and COPth increase under higher temperature and relative humidity conditions. (4) Both MRC and COPth decreased with cooling water at higher temperatures due to the adsorption heat generation. Marked improvement in COPth, from 0.34 to 0.62, for PCHX can be achieved with hot water temperature of 50
C. The proposed system has no moving parts rendering less maintenance compared to a desiccant wheel. Furthermore, it is an energy-efficient means of dehumidification by adsorption process with a waste heat source as compared to other conventional airconditioning processes.

(b) Fig. 8. Effect of change in temperatures of cooling and heating water on moisture removal capacity and COPth: (a) change in cooling water temperature (25, 30, and 35
C), (b) change in hot water temperature (50, 60, 70 and 80
C).

of PCHX steadily climbs from 0.34 to 0.62; for GPHX from 0.23 to 0.52. However, MRC of PCHX and GPHX are observed to drop about 1.6 g/kg and 0.9 respectively. The improvement of COP is attributed to the reduced amount of thermal energy used to regenerate the desiccant while the drop of MRC is due to the insufficient regeneration. If the hot water temperature is low, the silica gel will not be completely regenerated in a given cycle time, resulting in a degradation of the desiccant’s adsorption capacity in a successive cycle. In other words, MRC is affected by the instant uptake capacity, which varies with (1) cooling water temperature and (2) the water content of silica gel from the previous cycle that is affected by how well the silica gel has been regenerated during regeneration process. That is, the lower the temperature of cooling water, the higher is uptake capacity during dehumidification. And, the higher the temperature of hot water, the more the desiccant releases water vapor during regeneration ensuring the better performance in a successive cycle. It is noteworthy to note that the COPth increases since the drop of the cooling power is relatively smaller than the saving of the input thermal energy.

Acknowledgement The authors gratefully acknowledged the financial support from National Research Foundation of Singapore (grant no. R-265-000466-281) and Korean National Research Foundation (grant no. 2014R1A2A1A01006421). Nomenclature

5. Conclusion

AHU COPth CFCs Cp DW FESEM GPHX HEC hv M MRC MVC PCHX Qc Qreg RD RH T W

air handling unit thermal coefficient of performance chlorofluorocarbon specific heat at constant pressure (kJ/kg
C) desiccant wheel field emission scanning electron microscopy granule-packed heat exchanger hydroxyethyl cellulose latent heat of water (kJ/kg) mass flow rate (kg/s) moisture removal capacity(gwater/kgair) mechanical vapor compressor cycle powder coated heat exchanger latent cooling capacity (kW) regeneration heating power (kW) regular density relative humidity ratio temperature (
C) humidity ratio (kgwater vapor/kgdry air)

In this study, a solid desiccant dehumidifier incorporating twoadsorbent silica-gel coated heat exchangers has been developed

Subscripts a air

Please cite this article in press as: Oh SJ, et al., Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations, Energy (2017), http://dx.doi.org/10.1016/j.energy.2017.02.169

8

S.J. Oh et al. / Energy xxx (2017) 1e8

ad db hw in out reg

adsorption dry bulb hot water inlet outlet regeneration

References [1] Singstat. Department of statistics Singapore: Singapore in figures - 2009. 2015. https://www.singstat.gov.sg/statistics/browse-by-theme. [2] EIA. Internation energy statistics. US Energy Inf Adm; 2016. [3] Oh SJ, Ng KC, Thu K, Chun W, Chua KJE. Forecasting long-term electricity demand for cooling of Singapore’s buildings incorporating an innovative airconditioning technology. Energy Build 2016;127:183e93. http://dx.doi.org/ 10.1016/j.enbuild.2016.05.073. [4] Jones RR. Ozone depletion and its effects on human populations. Br J Dermatol 1992;41:2e6. [5] Daou K, Wang RZ, Xia ZZ. Desiccant cooling air conditioning: a review. Renew Sustain Energy Rev 2006;10:55e77. http://dx.doi.org/10.1016/ j.rser.2004.09.010. [6] Katejanekarn T, Chirarattananon S, Kumar S. An experimental study of a solarregenerated liquid desiccant ventilation pre-conditioning system. Sol Energy 2009;83:920e33. http://dx.doi.org/10.1016/j.solener.2008.12.006. [7] Pesaran AA. Review of desiccant dehumidification technology. 1994. http:// dx.doi.org/10.2172/934386. [8] Ramzy A, Elawady WM, Abdelmeguid H. Modelling of heat and moisture transfer in desiccant packed bed utilizing spherical particles of clay impregnated with CaCl2. Appl Therm Eng 2014;66:499e506. http://dx.doi.org/ 10.1016/j.applthermaleng.2014.02.031. [9] Ramzy KA, Kadoli R, Ashok Babu TP. Improved utilization of desiccant material in packed bed dehumidifier using composite particles. Renew Energy 2011;36:732e42. http://dx.doi.org/10.1016/j.renene.2010.06.038. [10] Hamed AM, Abd-Elrahman WR, El-Emam SH, Awad MM. Theoretical and experimental investigation on the transient coupled heat and mass transfer in a radial flow desiccant packed bed. Energy Convers Manag 2013;65:262e71. http://dx.doi.org/10.1016/j.enconman.2012.07.025. [11] Harriman LG. The dehumidification handbook. 2002. [12] Chen CH, Hsu CY, Chen CC, Chiang YC, Chen SL. Silica gel/polymer composite desiccant wheel combined with heat pump for air-conditioning systems. Energy 2016;94:87e99. http://dx.doi.org/10.1016/j.energy.2015.10.139. [13] Abdel-Salam AH, Ge G, Simonson CJ. Thermo-economic performance of a solar membrane liquid desiccant air conditioning system. Sol Energy 2014;102: 56e73. http://dx.doi.org/10.1016/j.solener.2013.12.036. [14] Xie Y, Zhang T, Liu X. Performance investigation of a counter-flow heat pump driven liquid desiccant dehumidification system. Energy 2016;115:446e57. http://dx.doi.org/10.1016/j.energy.2016.09.037. [15] Nie J, Fang L, Zhang G, Sheng Y, Kong X, Zhang Y, et al. Theoretical study on volatile organic compound removal and energy performance of a novel heat pump assisted solid desiccant cooling system. Build Environ 2015;85:233e42. http://dx.doi.org/10.1016/j.buildenv.2014.11.034. [16] Das RS, Jain S. Performance characteristics of cross-flow membrane contactors for liquid desiccant systems. Appl Energy 2014;141:1e11. http://dx.doi.org/ 10.1016/j.apenergy.2014.12.014. [17] Zhang N, Yin SY, Zhang LZ. Performance study of a heat pump driven and hollow fiber membrane-based two-stage liquid desiccant air dehumidification system. Appl Energy 2016;179:727e37. http://dx.doi.org/10.1016/ j.apenergy.2016.07.037. [18] Aristov YI, Restuccia G, Cacciola G, Parmon VN. A family of new working materials for solid sorption air conditioning systems. Appl Therm Eng 2002;22:191e204. http://dx.doi.org/10.1016/S1359-4311(01)00072-2. [19] M D. Desiccant composition, n.d. [20] Zhang XJ, Sumathy K, Dai YJ, Wang RZ. Dynamic hygroscopic effect of the composite material used in desiccant rotary wheel. Sol Energy 2006;80: 1058e61. http://dx.doi.org/10.1016/j.solener.2005.07.008. [21] Zhang XJ, Qiu LM. Moisture transport and adsorption on silica gel-calcium chloride composite adsorbents. Energy Convers Manag 2007;48:320e6. http://dx.doi.org/10.1016/j.enconman.2006.04.001.

[22] Jia CX. Study on adsorption mechanism and dehumidification property of composite desiccant. Adv Mater Res 2011;150e151:912e6. doi:10.4028, www.scientific.net/AMR.150-151.912. [23] Gordeeva LG, Grekova AD, Krieger TA, Aristov YI. Adsorption properties of composite materials (LiCl þ LiBr)/silica. Microporous Mesoporous Mater 2009;126:262e7. [24] Jia CX, Dai YJ, Wu JY, Wang RZ. Use of compound desiccant to develop high performance desiccant cooling system. Int J Refrig 2007;30:345e53. http:// dx.doi.org/10.1016/j.ijrefrig.2006.04.001. [25] SorTech AG. Layer Composite and Production Thereof. 2007. [26] Freni A, Bonaccorsi L, Calabrese L, Caprì A, Frazzica A, Sapienza A. SAPO-34 coated adsorbent heat exchanger for adsorption chillers. Appl Therm Eng 2015;82:1e7. http://dx.doi.org/10.1016/j.applthermaleng.2015.02.052. [27] Fang Y, Zuo S, Liang X, Cao Y, Gao X, Zhang Z. Preparation and performance of desiccant coating with modified ion exchange resin on finned tube heat exchanger. Appl Therm Eng 2016;93:36e42. http://dx.doi.org/10.1016/ j.applthermaleng.2015.09.044. [28] Okamoto K, Teduka M. The development of aqsoa water vapor adsorbent and aqsoa coated heat exchanger. Impres 2010:27e32. http://dx.doi.org/10.3850/ 978-981-08-7614-2. [29] Li A, Thu K, Bin Ismail A, Shahzad MW, Ng KC. Performance of adsorbentembedded heat exchangers using binder-coating method. Int J Heat Mass Transf 2016;92:149e57. http://dx.doi.org/10.1016/j.ijheatmasstrans fer.2015.08.097. [30] Ge TS, Dai YJ, Wang RZ, Peng ZZ. Experimental comparison and analysis on silica gel and polymer coated fin-tube heat exchangers. Energy 2010;35: 2893e900. http://dx.doi.org/10.1016/j.energy.2010.03.020. [31] Huan Z, Jianlei N. Two-stage desiccant cooling system using low-temperature heat. Build Serv Eng Res Technol 1999;20:51e5. [32] Weixing Y, Yi Z, Xiaoru L, Xiugan Y. Study of a new modified cross-cooled compact solid desiccant dehumidifier. Appl Therm Eng 2008;28:2257e66. http://dx.doi.org/10.1016/j.applthermaleng.2008.01.006. [33] Zheng Y, Yuan W, Wang H, Yuan X. Experiments on dynamic dehumidification of internally cooling compact solid dehumidifier. J Beijing Univ Aeronaut Astronaut 2006;32:1100e3. [34] Fathalah K, Aly SE. Study of a waste heat driven modified packed desiccant bed dehumidifier. Energy Convers Manag 1996;37:457e71. http://dx.doi.org/ 10.1016/0196-8904(95)00201-4. [35] Ng KC, Chua HT, Chung CY, Loke CH, Kashiwagi T, Akisawa A, et al. Experimental investigation of the silica gel-water adsorption isotherm characteristics. Appl Therm Eng 2001;21:1631e42. http://dx.doi.org/10.1016/S13594311(01)00039-4. [36] Shahzad MW, Thu K, Kim Y deuk, Ng KC. An experimental investigation on MEDAD hybrid desalination cycle. Appl Energy 2015;148:273e81. http:// dx.doi.org/10.1016/j.apenergy.2015.03.062. [37] Myat A, Thu K, Choon NK. The experimental investigation on the performance of a low temperature waste heat-driven multi-bed desiccant dehumidifier (MBDD) and minimization of entropy generation. Appl Therm Eng 2012;39: 70e7. http://dx.doi.org/10.1016/j.applthermaleng.2012.01.041. [38] Cacciola G, Restuccia G. Progress on adsorption heat pumps. Heat Recover Syst CHP 1994;14:409e20. http://dx.doi.org/10.1016/0890-4332(94)90044-2. [39] Taweekun J, Akvanich V. The experiment and simulation of solid desiccant dehumidification for air-conditioning system in a tropical humid climate. Engineering 2013;5:146e53. http://dx.doi.org/10.4236/eng.2013.51A021. [40] Performance comparison between desiccant cooling system and vapor compression system n.d. [41] Jani DB, Mishra M, Sahoo PK. Experimental investigation on solid desiccantvapor compression hybrid air-conditioning system in hot and humid weather. Appl Therm Eng 2016;104:556e64. http://dx.doi.org/10.1016/ j.applthermaleng.2016.05.104. [42] Subramanyam N, Maiya MP, Murthy SS. Parametric studies on a desiccant assisted air-conditioner. Appl Therm Eng 2004;24:2679e88. http://dx.doi.org/ 10.1016/j.applthermaleng.2004.03.020. [43] Sheng Y, Zhang Y, Deng N, Fang L, Nie J, Ma L. Experimental analysis on performance of high temperature heat pump and desiccant wheel system. Energy Build 2013;66:505e13. http://dx.doi.org/10.1016/ j.enbuild.2013.07.058. [44] Fischer BJ. The Impact of Relative Humidity, Heat of Adsorption and Carryover Heat The Impact of Relative Humidity, Heat of Adsorption and Carryover Heat n.d.

Please cite this article in press as: Oh SJ, et al., Evaluation of a dehumidifier with adsorbent coated heat exchangers for tropical climate operations, Energy (2017), http://dx.doi.org/10.1016/j.energy.2017.02.169