- Email: [email protected]
Use of compound desiccant to develop high performance desiccant cooling system
International Journal of Refrigeration 30 (2007) 345e353 www.elsevier.com/locate/ijrefrig
Use of compound desiccant to develop high performance desiccant cooling system C.X. Jia, Y.J. Dai*, J.Y. Wu, R.Z. Wang Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Huashan Road 1954, Shanghai, 200030, PR China Received 16 June 2005; received in revised form 24 March 2006; accepted 5 April 2006 Available online 12 June 2006
Abstract The paper is aimed to develop a high performance rotary solid desiccant cooling system using a novel compound desiccant wheel (DW). The unique feature of the desiccant wheel is that it can work well under a lower regeneration temperature and have a higher dehumidification capacity due to the contribution of the new compound desiccant materials. Experimental results indicate that the novel desiccant wheel under practical operation can remove more moisture from the process air by about 20e40% over the desiccant wheel employing regular silica gel. A mathematical model that is used to predict the system performance has been validated with the test results. By integrating the desiccant wheel with evaporative cooling, heat recovery and heating for regeneration sections, a solid desiccant cooling system can be formed. Simulation results show that because of the use of the new compound desiccant, the desiccant cooling system can work under much lower regeneration temperature and have a relative high COP, thus low grade thermal energy resources, such as solar energy, waste heat, etc., can be efficiently utilized to drive such a cooling cycle. Ó 2006 Elsevier Ltd and IIR. All rights reserved. Keywords: Air conditioning; Desiccant wheel; Adsorption; Enhancement; Desiccant; Experiment; Modelling; Performance
Utilisation d’un compose´ de´shydratant afin de de´velopper un syste`me a` de´shydratant tre`s performant Mots cle´s : Conditionnement d’air ; Roue de´shydratante ; Adsorption ; Ame´lioration ; De´shydratant ; Expe´rimentation ; Mode´lisation ; Performance
1. Introduction Desiccant cooling systems work on the principles of desiccant dehumidification and evaporative cooling. The unique merit they have is that the sensible and the latent
* Corresponding author. Tel./fax: þ86 21 6293 3250. E-mail address: [email protected] (Y.J. Dai). 0140-7007/$35.00 Ó 2006 Elsevier Ltd and IIR. All rights reserved. doi:10.1016/j.ijrefrig.2006.04.001
heat can be processed separately. Also, they are advantageous particularly for use in hot and humid climates, and can have access to various low grade thermal energy resources, such as solar energy, waste heat, etc. During the past decades, many efforts have been made to develop such kind of cooling device [1e3]. It is found that desiccant systems are quite efficient in dealing with the latent load, but considerably less so with regard to the sensible load [4]. There are two solutions to overcome such defect. One
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
346
Nomenclature T RH m Y h COP Qc Nu Sh u cp
Dry-bulb temperature ( C) Relative humidity Mass flow rate of air (kg/s) Humidity ration of air (kg/kg) Enthalpy of air (kJ/kg) Coefficient of performance Refrigerating output (kW) Nusselt number Sherwood number Rotation speed (r/h) Specific heat (J/kg K)
is for the desiccant cooling system to deal with the latent heat while an electric-powered heat pump deals with the remaining sensible heat. The other is to improve the performance of the desiccant wheel and make the process air sufficient dry so that the sensible heat can be removed entirely by means of evaporative cooling. Nevertheless, one has to face the fact that the size of the desiccant wheel must be big enough if the conventional desiccant rotors are utilized, or other high performance desiccant wheel should be developed by identifying new desiccant materials in order to make the system compact and lower the initial and operating costs. It is known that the advanced desiccant materials may give improved sorption capacity, better moisture and heat diffusion rates, as well as favorable equilibrium isotherms, of which an ideal Type 1M isotherm shape (Modified Langmuir Type 1) was proposed by Collier et al. [5] as early as 1986. In general, the chosen adsorbent used in desiccant wheel should have both high hygroscopic capacity and saturated adsorption rate. At present, commercially available desiccants include silica gel, activated alumina, natural and synthetic zeolites, titanium silicate, lithium chloride, and synthetic polymers. Such kinds of devices are widely utilized in industrial and commercial application for dehumidification operation. Since Pennington [6] has patented the first desiccant cooling system (DCS), there have been many reports on desiccant cooling performance so far. Parameters optimization on desiccant cooling systems was one of important works [7e9]. Refs. [10e12] report the open cycle systems experimentally. However, to date, coefficient of performance (COP) for DCS has been usually about 0.5e1.0 [13]. It is the key for improvement in COP of DCS to develop novel desiccant wheels with advanced desiccant materials. In this paper, a novel desiccant wheel fabricated with a new kind of composite desiccant is developed and utilized in a desiccant cooling system. The main objectives are to report the relevant work and analyze the performance of the DCS for the purpose of obtaining a better COP through both numerical calculation and experimental analysis.
a b fd fm r L
Half height of desiccant channel (m) Half width of desiccant channel (m) Mass of desiccant unit length (kg/m) Mass of matrix unit length (kg/m) Radius (m) Latent heat of water vapor (kJ/kg)
Subscripts r Regeneration air and rejected air p Process air side a Ambient air s Supply air
Fig. 1. The experimental desiccant wheel.
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
347
Radius/130mm
Thickness/100mm 0m 68
1. Heat insulation felt ring 2. Steel hoop of fixing felt ring 3. Heat insulation felt 4. Regeneration air conduit 5. Steel plate of fixing felt 6. Steel ring of fixing felt ring
m
150mm
170mm
1 1064mm
4
Fig. 3. Gearing system.
0m
6 Thickness/1mm
m
rotor and gearing system should be considered. Fig. 3 presents the gearing system of the studied desiccant wheel. The photo of the experimental equipment is shown in Fig. 4. 31
1 Thickness/2mm 2 Thickness/1mm 3 Thickness/1mm 4 Angle/90°C
2 1.wheel Radius/90mm 2.chain 3.gear 3 4.electromotor Radius/70mm
5 Thickness/1mm
Fig. 2. Equipment of airproof and heat insulation.
2. Description of the new desiccant wheel The studied novel desiccant wheel is not new but a rotary dehumidifier fabricated with one kind of new composite desiccant material. The composite desiccant is a two-layered material that consists of a host matrix with open pores (silica gel) and a hygroscopic substance (lithium chloride) impregnated into its pores. Due to its physical structure the composite desiccant takes an intermediate position between solid adsorbent and pure hygroscopic salt and can be organized in a way to demonstrate the best features of both systems. The pore surface area of the composite desiccant is 194 m2/g and the pore diameter is 3.98 nm. To form a desiccant wheel, a honeycombed matrix, which can adhere the desiccant materials and have a mass of parallel micro-air channels, is extremely necessary. The air channel, for which the walls are coated with abundant desiccant materials, is capable of removing the moisture from the passing process air. The air channels and the fabricated desiccant wheel are shown in Fig. 1. Nevertheless, when the desiccant is saturated with water vapor, the unit cannot dehumidify the process air any more, thus a regeneration section, which can drive the adsorbed water vapor out by heating and make the desiccant active again, should be considered. As shown in Fig. 2, the work herein makes a division, and lets 1/4 of the surface area of the wheel exposed to regeneration air. Also the airproof and heat insulation in the air tunnel are important to ensure the good performance of a rotary desiccant dehumidifier. Here, insulation felt, fiberglass plastic and fixed steel ring are necessary for the airproof and heat insulation. To drive the wheel, a retardment
3. Test comparison In order to identify the unique features of the composite desiccant over the conventional materials, for example, silica gel, test comparisons on both thermal gravity analysis and water vapor adsorption isotherms, as well as the dehumidification performances, are made in the following. 3.1. Thermal gravity analysis and isotherms Fig. 5 depicts the thermal gravity curve of the two tested samples, namely, silica gel and composite desiccant. The thermal gravimetric analysis (TGA) can study the thermal stability, adsorption/desorption, oxidation/reduction, reaction kinetic, etc. It is seen that the dehydration rates of composite desiccant are bigger than the silica gel under any given temperature conditions. These curves indicate that
Fig. 4. The photo of the experimental equipment.
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
348 105
1.0 Silica gel Composite desiccant
0.9
Adsorption capacity(g/g)
Weight( )
100
95
90
85
80
0.8 0.7 0.6 0.5
composite desiccant :Ta=15°C silica gel:Ta=15°C composite desiccant:Ta=25°C silica gel:Ta=25°C composite desiccant:Ta=35°C silica gel:Ta=35°C composite desiccant:Ta=40°C silica gel:Ta=40°C
0.4 0.3 0.2 0.1
0
20
40
60
80
100
120
0.0 0.0
Temperature (°C)
0.2
0.4
0.6
0.8
1.0
Relative humidity Fig. 5. TG curve of the three tested samples. Fig. 6. The equilibrium adsorption curves.
the composite desiccant is easier for regeneration than the regular silica gel. Water vapor adsorption isotherms were investigated to evaluate the property of the composite desiccant substrate. The tests were conducted in a thermo-humidistat chamber, in which a constant temperature and humidity environment can be created and maintained. Fig. 6 presents the isotherms for silica gel and composite desiccants under different ambient temperature. Analyzing the equilibrium water vapor adsorption isotherm, the following features can be concluded: (1) the equilibrium adsorption capacities of composite desiccant are one to two times higher than that of silica gel under given conditions; (2) the composite desiccant is more hygroscopic than silica gel under low relative humidity conditions. For example the equilibrium adsorption capacity of composite desiccant is one time higher than that of silica gel under high relative humidity, however, under low relative humidity the equilibrium adsorption capacity of composite desiccant is two times higher than that of silica gel; and (3) the adsorption capacity increases almost linearly with humidity under lower relative humidity conditions and rises rapidly under high humidity environment.
matrix. The dimensional sizes and operation parameters of the two desiccant wheels are listed in Table 1. Fig. 7 shows the schematic diagram of the test rig for a rotary desiccant dehumidifier. A commercially available electro-heater and an electro-heating humidifier are installed in the unit; the unit could control temperature and humidity through the console. In the regeneration section, an electroheater is added to provide the necessary heat. The temperature of regeneration air could be adjusted by means of the console controlling quality of heating. In order to avoid the dust particles that may lower the efficiency of the desiccant, two air filters will be installed, respectively, at the inlet positions. In the experiment, the temperature and relative humidity were measured by using a series of high accuracy and multifunction Digital Thermo/Hygrometer. The type of sensor used to measure the temperature and humidify of process air is THT-N263A. The upper limit of measured temperature is 60 C, the lower limit is 10 C, and the precision of temperature sensor is 0.5 C. Relative humidity monitored with the sensor ranges from 10 to 100%RH and its precision is 2%RH. After Keithley data logger collects the electric signal, the data can be saved in the computer. At regeneration airside, a suit of temperature and humidity digital network detector (Shinyei TRH 7X) with two probes are arranged at the inlet and the outlet of the desiccant wheel. The measured range is 20 to 80 C and 0e95%RH and the precisions are 0.3 C, 2%RH.
3.2. Performance comparison of two desiccant wheels Further comparison was made between two complete desiccant wheels, one is from the regular silica gel, and the other is from composite desiccant. Both wheels are fabricated in our laboratory, and have the same supporting
Table 1 Operating and structure parameters Diameter (mm)
Thickness (mm)
Flow rate of process air (m3/h)
Flow rate of regeneration air (m3/h)
Power of regeneration air fan (kW)
Power of process air fan (kW)
Power of regeneration heater (kW)
Power of motor (kW)
260
100
360
120
0.23
0.25
3.9
0.12
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
349
Data collector
Rotary wheel
Humidifier
Yp1
Yp2
Fan Process air
Heater
hr2
hr1 heater
Regeneration air Electromotor
Fig. 7. The test rig.
Fig. 8 presents the comparison results under various regeneration temperature conditions. Here, the rotation speed is set as 12 r/h. It is seen that the new composite desiccant wheel can remove more moisture from the air, about 40e50% higher over the silica gel one, indicating a better performance for using new materials.
5. Performance analysis
4. Model validation By employing the mathematical model developed by Zhang et al. [14], some simulation and analysis work can be carried out. Table 2 lists the relevant parameters that will be used in the computation. The adsorption isotherms of the new composite desiccant materials are like in the aforementioned part of this paper. Here, we validate our model using one group of test data, which is shown in Fig. 9. It is seen that the agreement of the simulated and the experimentally obtained results is generally good for different regeneration temperature levels. The discrepancy between the two results is about 15e20%. The reason is that the estimated heat and mass transfer coefficient, as well as some other parameters, do not still approach the real value. 8 Slica gel Composite desiccant
Dehumidification amount(g)
7 6
5.1. Analyses on desiccant wheels In this paper, the performance of composite desiccant wheel and silica gel desiccant wheel is evaluated based on two different criteria: (1) moisture removal capacity, and (2) dehumidification coefficient of performance. The moisture removal capacity can be expressed as: D ¼ Yp1 Yp2 where D is the moisture removal capacity, Yp1 and Yp2 are the humidity ratio of inlet and outlet process air, respectively; the dehumidification coefficient of performance (DCOP) of desiccant wheel is particularly defined. The parameter reflects the dehumidification performance and energy consumption synthetically, and can be presented by: m_ p L Yp1 Yp2 DCOP ¼ m_ r ðhr1 hr2 Þ Table 2 Parameters used in computation
5 4 3 2 1 0 40
In the analysis, the Nusselt number and the Sherwood number are both taken as 20, which agree with the experimental observation in this study. Hence, in the following, the parameters listed in Table 2 will be utilized for further analysis.
50
60
70
80
90
100
Regeneration temperature(°C) Fig. 8. Comparison of two wheels (12 r/h).
110
120
Specific heat of composite desiccant cpd (J/kg K) Half height of desiccant channel a (m) Half width of desiccant channel b (m) Nuz Shz Rotation speed u (r/h) Specific heat of ceramic matrix cpm (J/kg K) Thickness of desiccant wheel (m) Radius r (m) Mass of desiccant unit length fd (kg/m) Mass of ceramic matrix unit length fm (kg/m)
921 0.001 0.001 20 20 16.0 880 0.1 0.13 0.005 0.003
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
350 8
8
6 5 4 3 2
6 5 4 3 2 1
1 0 40
0 0.3 50
60
70
80
90
100
Silica gel wheel Composite desiccant wheel
7
Dehumidification amount(g)
Dehumidification amount(g)
(a)
Simulation data Experimental data
7
0.4
0.5
110
Regeneration temperature (°C)
(b)
where m_ p and m_ r are flow rate of process air and regeneration air, respectively (kg/s); L is the latent heat of water vapor (J/kg); hr1 and hr2 are enthalpy of inlet and outlet regeneration air, respectively (J/kg). For an ideal
Dehumidification amounts(g/kg)
(a)
3.5
0.8
0.9
1.0
Silica gel desiccant wheel Composite desiccant wheel
3.0 2.5 2.0 1.5
8
1.0
Silica gel desiccant wheel Composite desiccant wheel
7
Rotation speed =16rph Regeneration temperature =100 °C Inlet temperature=21 °C
0.5
6
0.0 0.3
5
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Relative humidity 4
Fig. 11. Impact of inlet air relative humidity on dehumidification performance.
3 2 Rotation speed =16rph Regeneration temperature =100 °C Relative humidity=40
1 0 18
20
22
24
26
28
30
32
34
36
38
Inlet temperature of process air (°C)
(b)
0.7
4.0
Dcop
Fig. 9. Comparison of simulated and experimental results.
0.6
Relative humidity
isenthalpic dehumidification process, though DCOP tends towards infinity, for an actual process, it would have a definite value such that 0 DCOP < N. It is seen that, higher value of DCOP indicates better performance of the rotary wheel.
5 Silica gel desiccant wheel Composite desiccant wheel
8
Dehumidification amount(g)
4
Dcop
3
2
1
0 18
Rotation speed =16rph Regeneration temperature =100 °C Relative humidity=40 20
22
24
26
28
30
32
34
36
Composite desiccant wheel Silica gel wheel
7 6 5 4 3 2 1
38
Inlet temperature of process air (°C) Fig. 10. Impact of inlet air temperature on dehumidification performance.
0
0
50
100
150
200
Thickness of the desiccant wheel(mm) Fig. 12. Impact of wheel thickness on dehumidification amounts.
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
351
Evaporative cooler
2
5 6
Reject air
Supply air
Heat exchanger
1 Desiccant wheel
Ambient air
4 Heater
3 Evaporative cooler
Fig. 13. Schematic diagram of a desiccant cooling system.
The impact of the wheel thickness on the dehumidification performance is shown in Fig. 12. It is seen that there is an optimum value with respect to dehumidification amount under given conditions. The optimum thickness is about 80e100 mm. The maximum dehumidification amount can be achieved corresponding to this point, for both the composite desiccant wheel and the conventional silica gel wheel.
Variations of dehumidification amount and DCOP with the inlet air temperatures are plotted in Fig. 10. Significant improvement in both D and DCOP can be found under given operation conditions listed in the figure. Similar to all the results obtained in the aforementioned work, the composite desiccant rotor can remove more moisture than the silica gel one by about 50% corresponding to 100 C regeneration temperature, and have a much higher DCOP. The higher the inlet air temperature, the higher DCOP will be, and the much better the composite desiccant will have. The influence of inlet air relative humidity on dehumidification amount and DCOP are presented in Fig. 11. Under given conditions, namely, the regeneration temperature is 100 C, inlet air temperature is 21 C. The dehumidification amount of the new rotors is bigger than that of silica gel one by about 30e50%. Particularly, there exists a peak DCOP, when the inlet air relative humidity is about 80%. This is explainable in terms of different adsorption mechanisms involved. It is found that the moisture removal capacity of composite desiccant wheel is almost one and half times as high as that of silica gel wheel when the relative humidity is 40%. But with the increase in relative humidity, the dehumidification predominance of composite desiccant wheel weakens. That is because relative humidity is too low; the chemical reaction dominates the solid adsorption. However, under high relative humidity conditions, not only chemical reaction but also solid adsorption and capillary condensation occur, so the dehumidification predominance of composite desiccant wheel is not so obvious.
5.2. Desiccant cooling cycle analysis Since the use of composite desiccant materials is positively helpful to improve the dehumidification performance of a desiccant wheel, it will also be possible for adoption of such kind of wheel to configure a high performance desiccant cooling apparatus. Fig. 13 presents a desiccant cooling system, which consists of a desiccant wheel, a regenerative heat exchanger, two evaporative coolers, etc. The system works in open cycle, and two parts of air are involved in the cooling process. The process air comes from room outside, thereafter is dehumidified through desiccant wheel, and then is directly cooled by a direct evaporative cooler to a reasonable temperature and humidity level, at last is sent to the conditioned space. The regeneration air, which is used to regenerate the desiccant wheel, is from the return air of the conditioned room. The thermal energy to regenerate the desiccant wheel can be provided by solar air collector, waste heat recovery unit, gas burner, etc.
Table 3 Performance of the desiccant cooling system under typical climate conditions (composite desiccant (line 1), silica gel (line 2), regeneration temperature: 95 C) Ambient conditions
Indoor environment
Supplied air
Rejected air from DW
Performance index
Ta ( C)
RHa (%)
Ti ( C)
RHi (%)
Ts ( C)
RHs (%)
Tr ( C)
RHr (%)
COP
Qc (kW)
35
40
26.7
50
20 22
67 72
48 50
32 24
1.28 0.95
2.53 1.81
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
352
In this study, based on the validated aforementioned mathematical model for the desiccant wheel, and some simplified models for both the heat exchanger and the evaporative cooler, as mentioned in Ref. [15], the system performance of the studied desiccant cooling system has been analyzed. Here, the effectiveness of the regenerative heat exchanger is assumed as 0.8, the evaporative cooling efficiency is 0.6. The mass flow rates for the process air and the regeneration air are 360 m3/h and 120 m3/h, respectively. The other parameters are taken from Table 2. Particularly, for the heat exchangers, both the cold and the warm air flow rates are equal, that is to say, about 360 m3/h, but some amount of the cold air will be rejected, and only small part of air is heated to regenerate the desiccant wheels. To evaluate the performance of desiccant cooling system, two performance indexes have been defined. One of them is the refrigerating output, Qc ¼ mp ðha hs Þ, the other is the coefficient of performance, COP ¼
mp ðha hs Þ mr ðh6 h5 Þ
where the subscripts correspond to the state point in the desiccant cooling system. The analyzed results are summarized in Table 3. The working conditions of the system are also included in the table. The supplied air is from the outlet of the process air, and will be sent to the conditioned space. The rejected air is taken from the outlet of regeneration area of the desiccant wheel. The detailed analysis results are also presented in Table 3. It is shown that the desiccant cooling system using composite desiccant performs better than the regular silica gel system. For the former, COP is 1.28, about 35% higher than the latter one. The temperature of supplied air is about 20 C, and the relative humidity is about 67% for the composite desiccant system, while those for regular silica gel is 22 C and 72%. Also should be noted is that the regeneration air temperature is only 95 C, hence solar energy or waste heat can be efficiently utilized. 6. Conclusions In this study, a new composite desiccant material, which presents merits of both porous physical adsorption and moisture hydration effect, is used in the desiccant wheels. Performance comparison between the desiccant wheels using composite desiccant and regular silica gel is made. A desiccant cooling cycle, which can adjust the process air to a reasonable delivery conditions for the conditioned space through utilization of new desiccant material, is also developed and analyzed. As a brief summary, the following conclusions can be made: (1) Compared with the regular silica gel, the composite desiccant is more hygroscopic and more easily regenerated.
It may mean a compact size and low cost for use of such kind of material in desiccant dehumidification system; (2) Under given conditions, test results indicate that the new desiccant wheel can remove more moisture from the process air by about 20e30% than the regular silica gel one; (3) Dehumidification coefficient of performance (DCOP) is a comprehensive performance index to evaluate the desiccant wheel. Simulation results using a validated model show that the DCOP for the new wheel is greatly better than the conventional one, and there may exist an optimum wheel thickness, which may maximize the dehumidification amount, is about 80e100 mm in this study; (4) A desiccant cooling system with new desiccant wheel is developed and analyzed, the system COP may reach 1.28, about 35% higher over the system using silica gel, and the supplied air can be adjusted to about 20 C, 67%, reasonable for air conditioning purposes.
Acknowledgment The research was supported by Daikin Co. and Shanghai Commission of Science and Technology under Contract No.05qmx1431.
References [1] H.M. Henning, T. Erpenbeck, C. Hindenberg, et al., The potential of solar energy use in desiccant cooling cycles, International Journal of Refrigeration 24 (2001) 220e229. [2] J. Wurm, D. Kosar, T. Clemens, Solid desiccant technology review, Bulletin of the International Institute of Refrigeration 82 (2002) 2e31. [3] N. Subramanyam, M.P. Maiya, et al., Application of desiccant wheel to control humidity in air-conditioning systems, Applied Thermal Engineering 24 (2004) 2777e2788. [4] K. Gommed, G. Grossman, A liquid desiccant system for solar cooling and dehumidification, in: Proceedings of the International Sorption Heat Pump Conference, September 24e27; Shanghai, P.R. China 2002 [pp. 655e660]. [5] R.K. Collier, T.S. Cale, Z. Lavan, Advanced desiccant material assessment, Gas Research Institute Report GRI-86/0182. [6] N.A. Pennington, Humidity Changer for Air Conditioning, U.S. Patent No. 2, 700, 537, January 1955. [7] E. Van den Bulck, The use of dehumidifiers in desiccant cooling and dehumidification systems, ASME Journal of Heat Transfer 108 (1986) 684e692. [8] R.K. Collier Jr, D. Novosel, W.M. Worek, Performance analysis of open-cycle desiccant cooling systems, ASHRAE Transactions 96 (1990) 1262e1268. [9] J.J. Jurinak, J.W. Mitchell, W.A. Beckman, Open-cycle desiccant air conditioning as an alternative to vapor compression cooling in residential applications, ASME Journal of Solar Energy Engineering 106 (1984) 252e260.
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353 [10] W. Casas, Experiences with a gas driven, desiccant assisted air conditioning system with geothermal energy for an office building, Energy and Buildings 37 (2005) 493e501. [11] S. Techajunta, S. Chirattananon, R.H.B. Exell, Experiments in a solar simulator on solid desiccant regeneration and air dehumidification for air conditioning in a tropical humid climate, Renewable Energy 17 (1999) 549e568. [12] M. Kanoqlu, M.O. Carpinlioqlu, M. Yildirim, Energy and exergy analyses of an experimental open-cycle desiccant cooling system, Applied Thermal Engineering 24 (2004) 919e932.
353
[13] D.G. Waugaman, A. Kini, C.F. Kettleborough, A review of desiccant cooling systems, Journal of Energy Resources Technology 115 (1993) 1e8. [14] X.J. Zhang, Y.J. Dai, R.Z. Wang, A simulation study of heat and mass transfer in a honeycombed rotary desiccant dehumidifier, Applied Thermal Engineering 23 (2003) 989e1003. [15] Y.J. Dai, R.Z. Wang, H.F. Zhang, Use of liquid desiccant cooling to improve the performance of vapor compression air conditioning, Applied Thermal Engineering 21 (2001) 1185e1202.
Use of compound desiccant to develop high performance desiccant cooling system C.X. Jia, Y.J. Dai*, J.Y. Wu, R.Z. Wang Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Huashan Road 1954, Shanghai, 200030, PR China Received 16 June 2005; received in revised form 24 March 2006; accepted 5 April 2006 Available online 12 June 2006
Abstract The paper is aimed to develop a high performance rotary solid desiccant cooling system using a novel compound desiccant wheel (DW). The unique feature of the desiccant wheel is that it can work well under a lower regeneration temperature and have a higher dehumidification capacity due to the contribution of the new compound desiccant materials. Experimental results indicate that the novel desiccant wheel under practical operation can remove more moisture from the process air by about 20e40% over the desiccant wheel employing regular silica gel. A mathematical model that is used to predict the system performance has been validated with the test results. By integrating the desiccant wheel with evaporative cooling, heat recovery and heating for regeneration sections, a solid desiccant cooling system can be formed. Simulation results show that because of the use of the new compound desiccant, the desiccant cooling system can work under much lower regeneration temperature and have a relative high COP, thus low grade thermal energy resources, such as solar energy, waste heat, etc., can be efficiently utilized to drive such a cooling cycle. Ó 2006 Elsevier Ltd and IIR. All rights reserved. Keywords: Air conditioning; Desiccant wheel; Adsorption; Enhancement; Desiccant; Experiment; Modelling; Performance
Utilisation d’un compose´ de´shydratant afin de de´velopper un syste`me a` de´shydratant tre`s performant Mots cle´s : Conditionnement d’air ; Roue de´shydratante ; Adsorption ; Ame´lioration ; De´shydratant ; Expe´rimentation ; Mode´lisation ; Performance
1. Introduction Desiccant cooling systems work on the principles of desiccant dehumidification and evaporative cooling. The unique merit they have is that the sensible and the latent
* Corresponding author. Tel./fax: þ86 21 6293 3250. E-mail address: [email protected] (Y.J. Dai). 0140-7007/$35.00 Ó 2006 Elsevier Ltd and IIR. All rights reserved. doi:10.1016/j.ijrefrig.2006.04.001
heat can be processed separately. Also, they are advantageous particularly for use in hot and humid climates, and can have access to various low grade thermal energy resources, such as solar energy, waste heat, etc. During the past decades, many efforts have been made to develop such kind of cooling device [1e3]. It is found that desiccant systems are quite efficient in dealing with the latent load, but considerably less so with regard to the sensible load [4]. There are two solutions to overcome such defect. One
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
346
Nomenclature T RH m Y h COP Qc Nu Sh u cp
Dry-bulb temperature ( C) Relative humidity Mass flow rate of air (kg/s) Humidity ration of air (kg/kg) Enthalpy of air (kJ/kg) Coefficient of performance Refrigerating output (kW) Nusselt number Sherwood number Rotation speed (r/h) Specific heat (J/kg K)
is for the desiccant cooling system to deal with the latent heat while an electric-powered heat pump deals with the remaining sensible heat. The other is to improve the performance of the desiccant wheel and make the process air sufficient dry so that the sensible heat can be removed entirely by means of evaporative cooling. Nevertheless, one has to face the fact that the size of the desiccant wheel must be big enough if the conventional desiccant rotors are utilized, or other high performance desiccant wheel should be developed by identifying new desiccant materials in order to make the system compact and lower the initial and operating costs. It is known that the advanced desiccant materials may give improved sorption capacity, better moisture and heat diffusion rates, as well as favorable equilibrium isotherms, of which an ideal Type 1M isotherm shape (Modified Langmuir Type 1) was proposed by Collier et al. [5] as early as 1986. In general, the chosen adsorbent used in desiccant wheel should have both high hygroscopic capacity and saturated adsorption rate. At present, commercially available desiccants include silica gel, activated alumina, natural and synthetic zeolites, titanium silicate, lithium chloride, and synthetic polymers. Such kinds of devices are widely utilized in industrial and commercial application for dehumidification operation. Since Pennington [6] has patented the first desiccant cooling system (DCS), there have been many reports on desiccant cooling performance so far. Parameters optimization on desiccant cooling systems was one of important works [7e9]. Refs. [10e12] report the open cycle systems experimentally. However, to date, coefficient of performance (COP) for DCS has been usually about 0.5e1.0 [13]. It is the key for improvement in COP of DCS to develop novel desiccant wheels with advanced desiccant materials. In this paper, a novel desiccant wheel fabricated with a new kind of composite desiccant is developed and utilized in a desiccant cooling system. The main objectives are to report the relevant work and analyze the performance of the DCS for the purpose of obtaining a better COP through both numerical calculation and experimental analysis.
a b fd fm r L
Half height of desiccant channel (m) Half width of desiccant channel (m) Mass of desiccant unit length (kg/m) Mass of matrix unit length (kg/m) Radius (m) Latent heat of water vapor (kJ/kg)
Subscripts r Regeneration air and rejected air p Process air side a Ambient air s Supply air
Fig. 1. The experimental desiccant wheel.
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
347
Radius/130mm
Thickness/100mm 0m 68
1. Heat insulation felt ring 2. Steel hoop of fixing felt ring 3. Heat insulation felt 4. Regeneration air conduit 5. Steel plate of fixing felt 6. Steel ring of fixing felt ring
m
150mm
170mm
1 1064mm
4
Fig. 3. Gearing system.
0m
6 Thickness/1mm
m
rotor and gearing system should be considered. Fig. 3 presents the gearing system of the studied desiccant wheel. The photo of the experimental equipment is shown in Fig. 4. 31
1 Thickness/2mm 2 Thickness/1mm 3 Thickness/1mm 4 Angle/90°C
2 1.wheel Radius/90mm 2.chain 3.gear 3 4.electromotor Radius/70mm
5 Thickness/1mm
Fig. 2. Equipment of airproof and heat insulation.
2. Description of the new desiccant wheel The studied novel desiccant wheel is not new but a rotary dehumidifier fabricated with one kind of new composite desiccant material. The composite desiccant is a two-layered material that consists of a host matrix with open pores (silica gel) and a hygroscopic substance (lithium chloride) impregnated into its pores. Due to its physical structure the composite desiccant takes an intermediate position between solid adsorbent and pure hygroscopic salt and can be organized in a way to demonstrate the best features of both systems. The pore surface area of the composite desiccant is 194 m2/g and the pore diameter is 3.98 nm. To form a desiccant wheel, a honeycombed matrix, which can adhere the desiccant materials and have a mass of parallel micro-air channels, is extremely necessary. The air channel, for which the walls are coated with abundant desiccant materials, is capable of removing the moisture from the passing process air. The air channels and the fabricated desiccant wheel are shown in Fig. 1. Nevertheless, when the desiccant is saturated with water vapor, the unit cannot dehumidify the process air any more, thus a regeneration section, which can drive the adsorbed water vapor out by heating and make the desiccant active again, should be considered. As shown in Fig. 2, the work herein makes a division, and lets 1/4 of the surface area of the wheel exposed to regeneration air. Also the airproof and heat insulation in the air tunnel are important to ensure the good performance of a rotary desiccant dehumidifier. Here, insulation felt, fiberglass plastic and fixed steel ring are necessary for the airproof and heat insulation. To drive the wheel, a retardment
3. Test comparison In order to identify the unique features of the composite desiccant over the conventional materials, for example, silica gel, test comparisons on both thermal gravity analysis and water vapor adsorption isotherms, as well as the dehumidification performances, are made in the following. 3.1. Thermal gravity analysis and isotherms Fig. 5 depicts the thermal gravity curve of the two tested samples, namely, silica gel and composite desiccant. The thermal gravimetric analysis (TGA) can study the thermal stability, adsorption/desorption, oxidation/reduction, reaction kinetic, etc. It is seen that the dehydration rates of composite desiccant are bigger than the silica gel under any given temperature conditions. These curves indicate that
Fig. 4. The photo of the experimental equipment.
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
348 105
1.0 Silica gel Composite desiccant
0.9
Adsorption capacity(g/g)
Weight( )
100
95
90
85
80
0.8 0.7 0.6 0.5
composite desiccant :Ta=15°C silica gel:Ta=15°C composite desiccant:Ta=25°C silica gel:Ta=25°C composite desiccant:Ta=35°C silica gel:Ta=35°C composite desiccant:Ta=40°C silica gel:Ta=40°C
0.4 0.3 0.2 0.1
0
20
40
60
80
100
120
0.0 0.0
Temperature (°C)
0.2
0.4
0.6
0.8
1.0
Relative humidity Fig. 5. TG curve of the three tested samples. Fig. 6. The equilibrium adsorption curves.
the composite desiccant is easier for regeneration than the regular silica gel. Water vapor adsorption isotherms were investigated to evaluate the property of the composite desiccant substrate. The tests were conducted in a thermo-humidistat chamber, in which a constant temperature and humidity environment can be created and maintained. Fig. 6 presents the isotherms for silica gel and composite desiccants under different ambient temperature. Analyzing the equilibrium water vapor adsorption isotherm, the following features can be concluded: (1) the equilibrium adsorption capacities of composite desiccant are one to two times higher than that of silica gel under given conditions; (2) the composite desiccant is more hygroscopic than silica gel under low relative humidity conditions. For example the equilibrium adsorption capacity of composite desiccant is one time higher than that of silica gel under high relative humidity, however, under low relative humidity the equilibrium adsorption capacity of composite desiccant is two times higher than that of silica gel; and (3) the adsorption capacity increases almost linearly with humidity under lower relative humidity conditions and rises rapidly under high humidity environment.
matrix. The dimensional sizes and operation parameters of the two desiccant wheels are listed in Table 1. Fig. 7 shows the schematic diagram of the test rig for a rotary desiccant dehumidifier. A commercially available electro-heater and an electro-heating humidifier are installed in the unit; the unit could control temperature and humidity through the console. In the regeneration section, an electroheater is added to provide the necessary heat. The temperature of regeneration air could be adjusted by means of the console controlling quality of heating. In order to avoid the dust particles that may lower the efficiency of the desiccant, two air filters will be installed, respectively, at the inlet positions. In the experiment, the temperature and relative humidity were measured by using a series of high accuracy and multifunction Digital Thermo/Hygrometer. The type of sensor used to measure the temperature and humidify of process air is THT-N263A. The upper limit of measured temperature is 60 C, the lower limit is 10 C, and the precision of temperature sensor is 0.5 C. Relative humidity monitored with the sensor ranges from 10 to 100%RH and its precision is 2%RH. After Keithley data logger collects the electric signal, the data can be saved in the computer. At regeneration airside, a suit of temperature and humidity digital network detector (Shinyei TRH 7X) with two probes are arranged at the inlet and the outlet of the desiccant wheel. The measured range is 20 to 80 C and 0e95%RH and the precisions are 0.3 C, 2%RH.
3.2. Performance comparison of two desiccant wheels Further comparison was made between two complete desiccant wheels, one is from the regular silica gel, and the other is from composite desiccant. Both wheels are fabricated in our laboratory, and have the same supporting
Table 1 Operating and structure parameters Diameter (mm)
Thickness (mm)
Flow rate of process air (m3/h)
Flow rate of regeneration air (m3/h)
Power of regeneration air fan (kW)
Power of process air fan (kW)
Power of regeneration heater (kW)
Power of motor (kW)
260
100
360
120
0.23
0.25
3.9
0.12
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
349
Data collector
Rotary wheel
Humidifier
Yp1
Yp2
Fan Process air
Heater
hr2
hr1 heater
Regeneration air Electromotor
Fig. 7. The test rig.
Fig. 8 presents the comparison results under various regeneration temperature conditions. Here, the rotation speed is set as 12 r/h. It is seen that the new composite desiccant wheel can remove more moisture from the air, about 40e50% higher over the silica gel one, indicating a better performance for using new materials.
5. Performance analysis
4. Model validation By employing the mathematical model developed by Zhang et al. [14], some simulation and analysis work can be carried out. Table 2 lists the relevant parameters that will be used in the computation. The adsorption isotherms of the new composite desiccant materials are like in the aforementioned part of this paper. Here, we validate our model using one group of test data, which is shown in Fig. 9. It is seen that the agreement of the simulated and the experimentally obtained results is generally good for different regeneration temperature levels. The discrepancy between the two results is about 15e20%. The reason is that the estimated heat and mass transfer coefficient, as well as some other parameters, do not still approach the real value. 8 Slica gel Composite desiccant
Dehumidification amount(g)
7 6
5.1. Analyses on desiccant wheels In this paper, the performance of composite desiccant wheel and silica gel desiccant wheel is evaluated based on two different criteria: (1) moisture removal capacity, and (2) dehumidification coefficient of performance. The moisture removal capacity can be expressed as: D ¼ Yp1 Yp2 where D is the moisture removal capacity, Yp1 and Yp2 are the humidity ratio of inlet and outlet process air, respectively; the dehumidification coefficient of performance (DCOP) of desiccant wheel is particularly defined. The parameter reflects the dehumidification performance and energy consumption synthetically, and can be presented by: m_ p L Yp1 Yp2 DCOP ¼ m_ r ðhr1 hr2 Þ Table 2 Parameters used in computation
5 4 3 2 1 0 40
In the analysis, the Nusselt number and the Sherwood number are both taken as 20, which agree with the experimental observation in this study. Hence, in the following, the parameters listed in Table 2 will be utilized for further analysis.
50
60
70
80
90
100
Regeneration temperature(°C) Fig. 8. Comparison of two wheels (12 r/h).
110
120
Specific heat of composite desiccant cpd (J/kg K) Half height of desiccant channel a (m) Half width of desiccant channel b (m) Nuz Shz Rotation speed u (r/h) Specific heat of ceramic matrix cpm (J/kg K) Thickness of desiccant wheel (m) Radius r (m) Mass of desiccant unit length fd (kg/m) Mass of ceramic matrix unit length fm (kg/m)
921 0.001 0.001 20 20 16.0 880 0.1 0.13 0.005 0.003
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
350 8
8
6 5 4 3 2
6 5 4 3 2 1
1 0 40
0 0.3 50
60
70
80
90
100
Silica gel wheel Composite desiccant wheel
7
Dehumidification amount(g)
Dehumidification amount(g)
(a)
Simulation data Experimental data
7
0.4
0.5
110
Regeneration temperature (°C)
(b)
where m_ p and m_ r are flow rate of process air and regeneration air, respectively (kg/s); L is the latent heat of water vapor (J/kg); hr1 and hr2 are enthalpy of inlet and outlet regeneration air, respectively (J/kg). For an ideal
Dehumidification amounts(g/kg)
(a)
3.5
0.8
0.9
1.0
Silica gel desiccant wheel Composite desiccant wheel
3.0 2.5 2.0 1.5
8
1.0
Silica gel desiccant wheel Composite desiccant wheel
7
Rotation speed =16rph Regeneration temperature =100 °C Inlet temperature=21 °C
0.5
6
0.0 0.3
5
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Relative humidity 4
Fig. 11. Impact of inlet air relative humidity on dehumidification performance.
3 2 Rotation speed =16rph Regeneration temperature =100 °C Relative humidity=40
1 0 18
20
22
24
26
28
30
32
34
36
38
Inlet temperature of process air (°C)
(b)
0.7
4.0
Dcop
Fig. 9. Comparison of simulated and experimental results.
0.6
Relative humidity
isenthalpic dehumidification process, though DCOP tends towards infinity, for an actual process, it would have a definite value such that 0 DCOP < N. It is seen that, higher value of DCOP indicates better performance of the rotary wheel.
5 Silica gel desiccant wheel Composite desiccant wheel
8
Dehumidification amount(g)
4
Dcop
3
2
1
0 18
Rotation speed =16rph Regeneration temperature =100 °C Relative humidity=40 20
22
24
26
28
30
32
34
36
Composite desiccant wheel Silica gel wheel
7 6 5 4 3 2 1
38
Inlet temperature of process air (°C) Fig. 10. Impact of inlet air temperature on dehumidification performance.
0
0
50
100
150
200
Thickness of the desiccant wheel(mm) Fig. 12. Impact of wheel thickness on dehumidification amounts.
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
351
Evaporative cooler
2
5 6
Reject air
Supply air
Heat exchanger
1 Desiccant wheel
Ambient air
4 Heater
3 Evaporative cooler
Fig. 13. Schematic diagram of a desiccant cooling system.
The impact of the wheel thickness on the dehumidification performance is shown in Fig. 12. It is seen that there is an optimum value with respect to dehumidification amount under given conditions. The optimum thickness is about 80e100 mm. The maximum dehumidification amount can be achieved corresponding to this point, for both the composite desiccant wheel and the conventional silica gel wheel.
Variations of dehumidification amount and DCOP with the inlet air temperatures are plotted in Fig. 10. Significant improvement in both D and DCOP can be found under given operation conditions listed in the figure. Similar to all the results obtained in the aforementioned work, the composite desiccant rotor can remove more moisture than the silica gel one by about 50% corresponding to 100 C regeneration temperature, and have a much higher DCOP. The higher the inlet air temperature, the higher DCOP will be, and the much better the composite desiccant will have. The influence of inlet air relative humidity on dehumidification amount and DCOP are presented in Fig. 11. Under given conditions, namely, the regeneration temperature is 100 C, inlet air temperature is 21 C. The dehumidification amount of the new rotors is bigger than that of silica gel one by about 30e50%. Particularly, there exists a peak DCOP, when the inlet air relative humidity is about 80%. This is explainable in terms of different adsorption mechanisms involved. It is found that the moisture removal capacity of composite desiccant wheel is almost one and half times as high as that of silica gel wheel when the relative humidity is 40%. But with the increase in relative humidity, the dehumidification predominance of composite desiccant wheel weakens. That is because relative humidity is too low; the chemical reaction dominates the solid adsorption. However, under high relative humidity conditions, not only chemical reaction but also solid adsorption and capillary condensation occur, so the dehumidification predominance of composite desiccant wheel is not so obvious.
5.2. Desiccant cooling cycle analysis Since the use of composite desiccant materials is positively helpful to improve the dehumidification performance of a desiccant wheel, it will also be possible for adoption of such kind of wheel to configure a high performance desiccant cooling apparatus. Fig. 13 presents a desiccant cooling system, which consists of a desiccant wheel, a regenerative heat exchanger, two evaporative coolers, etc. The system works in open cycle, and two parts of air are involved in the cooling process. The process air comes from room outside, thereafter is dehumidified through desiccant wheel, and then is directly cooled by a direct evaporative cooler to a reasonable temperature and humidity level, at last is sent to the conditioned space. The regeneration air, which is used to regenerate the desiccant wheel, is from the return air of the conditioned room. The thermal energy to regenerate the desiccant wheel can be provided by solar air collector, waste heat recovery unit, gas burner, etc.
Table 3 Performance of the desiccant cooling system under typical climate conditions (composite desiccant (line 1), silica gel (line 2), regeneration temperature: 95 C) Ambient conditions
Indoor environment
Supplied air
Rejected air from DW
Performance index
Ta ( C)
RHa (%)
Ti ( C)
RHi (%)
Ts ( C)
RHs (%)
Tr ( C)
RHr (%)
COP
Qc (kW)
35
40
26.7
50
20 22
67 72
48 50
32 24
1.28 0.95
2.53 1.81
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353
352
In this study, based on the validated aforementioned mathematical model for the desiccant wheel, and some simplified models for both the heat exchanger and the evaporative cooler, as mentioned in Ref. [15], the system performance of the studied desiccant cooling system has been analyzed. Here, the effectiveness of the regenerative heat exchanger is assumed as 0.8, the evaporative cooling efficiency is 0.6. The mass flow rates for the process air and the regeneration air are 360 m3/h and 120 m3/h, respectively. The other parameters are taken from Table 2. Particularly, for the heat exchangers, both the cold and the warm air flow rates are equal, that is to say, about 360 m3/h, but some amount of the cold air will be rejected, and only small part of air is heated to regenerate the desiccant wheels. To evaluate the performance of desiccant cooling system, two performance indexes have been defined. One of them is the refrigerating output, Qc ¼ mp ðha hs Þ, the other is the coefficient of performance, COP ¼
mp ðha hs Þ mr ðh6 h5 Þ
where the subscripts correspond to the state point in the desiccant cooling system. The analyzed results are summarized in Table 3. The working conditions of the system are also included in the table. The supplied air is from the outlet of the process air, and will be sent to the conditioned space. The rejected air is taken from the outlet of regeneration area of the desiccant wheel. The detailed analysis results are also presented in Table 3. It is shown that the desiccant cooling system using composite desiccant performs better than the regular silica gel system. For the former, COP is 1.28, about 35% higher than the latter one. The temperature of supplied air is about 20 C, and the relative humidity is about 67% for the composite desiccant system, while those for regular silica gel is 22 C and 72%. Also should be noted is that the regeneration air temperature is only 95 C, hence solar energy or waste heat can be efficiently utilized. 6. Conclusions In this study, a new composite desiccant material, which presents merits of both porous physical adsorption and moisture hydration effect, is used in the desiccant wheels. Performance comparison between the desiccant wheels using composite desiccant and regular silica gel is made. A desiccant cooling cycle, which can adjust the process air to a reasonable delivery conditions for the conditioned space through utilization of new desiccant material, is also developed and analyzed. As a brief summary, the following conclusions can be made: (1) Compared with the regular silica gel, the composite desiccant is more hygroscopic and more easily regenerated.
It may mean a compact size and low cost for use of such kind of material in desiccant dehumidification system; (2) Under given conditions, test results indicate that the new desiccant wheel can remove more moisture from the process air by about 20e30% than the regular silica gel one; (3) Dehumidification coefficient of performance (DCOP) is a comprehensive performance index to evaluate the desiccant wheel. Simulation results using a validated model show that the DCOP for the new wheel is greatly better than the conventional one, and there may exist an optimum wheel thickness, which may maximize the dehumidification amount, is about 80e100 mm in this study; (4) A desiccant cooling system with new desiccant wheel is developed and analyzed, the system COP may reach 1.28, about 35% higher over the system using silica gel, and the supplied air can be adjusted to about 20 C, 67%, reasonable for air conditioning purposes.
Acknowledgment The research was supported by Daikin Co. and Shanghai Commission of Science and Technology under Contract No.05qmx1431.
References [1] H.M. Henning, T. Erpenbeck, C. Hindenberg, et al., The potential of solar energy use in desiccant cooling cycles, International Journal of Refrigeration 24 (2001) 220e229. [2] J. Wurm, D. Kosar, T. Clemens, Solid desiccant technology review, Bulletin of the International Institute of Refrigeration 82 (2002) 2e31. [3] N. Subramanyam, M.P. Maiya, et al., Application of desiccant wheel to control humidity in air-conditioning systems, Applied Thermal Engineering 24 (2004) 2777e2788. [4] K. Gommed, G. Grossman, A liquid desiccant system for solar cooling and dehumidification, in: Proceedings of the International Sorption Heat Pump Conference, September 24e27; Shanghai, P.R. China 2002 [pp. 655e660]. [5] R.K. Collier, T.S. Cale, Z. Lavan, Advanced desiccant material assessment, Gas Research Institute Report GRI-86/0182. [6] N.A. Pennington, Humidity Changer for Air Conditioning, U.S. Patent No. 2, 700, 537, January 1955. [7] E. Van den Bulck, The use of dehumidifiers in desiccant cooling and dehumidification systems, ASME Journal of Heat Transfer 108 (1986) 684e692. [8] R.K. Collier Jr, D. Novosel, W.M. Worek, Performance analysis of open-cycle desiccant cooling systems, ASHRAE Transactions 96 (1990) 1262e1268. [9] J.J. Jurinak, J.W. Mitchell, W.A. Beckman, Open-cycle desiccant air conditioning as an alternative to vapor compression cooling in residential applications, ASME Journal of Solar Energy Engineering 106 (1984) 252e260.
C.X. Jia et al. / International Journal of Refrigeration 30 (2007) 345e353 [10] W. Casas, Experiences with a gas driven, desiccant assisted air conditioning system with geothermal energy for an office building, Energy and Buildings 37 (2005) 493e501. [11] S. Techajunta, S. Chirattananon, R.H.B. Exell, Experiments in a solar simulator on solid desiccant regeneration and air dehumidification for air conditioning in a tropical humid climate, Renewable Energy 17 (1999) 549e568. [12] M. Kanoqlu, M.O. Carpinlioqlu, M. Yildirim, Energy and exergy analyses of an experimental open-cycle desiccant cooling system, Applied Thermal Engineering 24 (2004) 919e932.
353
[13] D.G. Waugaman, A. Kini, C.F. Kettleborough, A review of desiccant cooling systems, Journal of Energy Resources Technology 115 (1993) 1e8. [14] X.J. Zhang, Y.J. Dai, R.Z. Wang, A simulation study of heat and mass transfer in a honeycombed rotary desiccant dehumidifier, Applied Thermal Engineering 23 (2003) 989e1003. [15] Y.J. Dai, R.Z. Wang, H.F. Zhang, Use of liquid desiccant cooling to improve the performance of vapor compression air conditioning, Applied Thermal Engineering 21 (2001) 1185e1202.