ArticlePDF Available

Design, simulation and testing of a hybrid liquid desiccant for independent control of temperature and humidity

Authors:
DESIGN, SIMULATION AND TESTING OF A HYBRID LIQUID
DESICCANT FOR INDEPENDENT CONTROL OF TEMPERATURE
AND HUMIDITY
L. Alonso1, X. Peña1, C. Pascual1, J. Prieto2, J. Ortiga2, K. Gommed3
1: Tecnalia Research & Innovation, Thermal Energy Area, Energy and Environment
Division. Area Anardi 5, 20730 Azpeitia (Gipuzkoa), Spain
2: Universitat Rovira i Virgili, CREVER, Mechanical Engineering Department. Avda Països
Catalans 26, 43007 Tarragona, Spain
3: Technion Israel Institute of Technology, Faculty of Mechanical Engineering. Haifa
32000 Israel
ABSTRACT
In this paper the design, simulation and testing of a hybrid liquid desiccant system for a case
study in Taiwan is presented, in the scope of the European project nanoCOOL. The designed
system has been constructed and is being tested at laboratory scale. After the set-up of the
system, it will be sent to Taipei for demonstration in real scale conditions, in the Taiwan
Building Technology Center. The demonstration site comprises two locker rooms in a
swimming pool of the university, with high internal humidity generation, low sensible heat
ratio, and high external humidity levels due to sub-tropical humid climate present in Taiwan.
Keywords: hybrid liquid desiccant system, simulation, temperature and humidity control
INTRODUCTION
The treatment of humidity on HVAC systems is crucial when a satisfactory indoor air quality
needs to be reached. Traditional HVAC systems meet the latent cooling load by reducing the
air temperature until its dew point, heating subsequently the air in order to reach the supply
temperature for user comfort, with the energy waste this entails. Hybrid liquid desiccant
systems (HLDS) combine the liquid desiccant technology for dehumidification of air with
conventional compression cycle technology for cooling.
The demo site comprises two locker rooms in a swimming pool of the National Taiwan
University of Science and Technology, with high internal humidity generation, low sensible
heat ratio, and high external humidity levels due to sub-tropical humid climate present in
Taiwan. Based on the design conditions, the internal sensible and latent heat generation, and
the ventilation requirements according to international standards; the design conditions and
the cooling and dehumidification loads have been calculated:
Outdoor design conditions
30ºC / 21,5 g/kg dry air
Comfort design conditions
24ºC / 7,5 g/kg dry air
Ventilation rate
2.500 m3/h
Internal sensible heat load
3,5 kW
Ventilation sensible heat load
5 kW
Total sensible heat load
8,5 kW
Internal latent heat load
8,8 kW
Ventilation latent heat load
21,3 kW
Total latent heat load
30,1 kW
Table 1: Design conditions and cooling/dehumidification loads
CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland
333
DESIGN AND SIMULATION
The designed hybrid liquid desiccant system (HLDS) is comprised by a liquid desiccant
system (LDS) whose main components are the absorber, regenerator and liquid-liquid heat
exchanger; and a conventional Air Handling Unit (AHU) with a cooling coil and a cross-plate
heat exchanger for ventilation heat recovery. A polyvalent unit able to simultaneously provide
cooling and heating, feeds the absorber and the cooling coil with cold water at 15ºC, and the
regenerator with hot water at 55ºC.
Models of the individual components of the Liquid Desiccant System (absorber, regenerator
and liquid-liquid heat exchanger) and the Air Handling Unit (cooling coil, air-air plates heat
exchanger) have been implemented in EES (Engineering Equation Solver) [1]. By combining
the separate models of each component, a versatile mathematical model has been developed,
which can be used for both the design and simulation of the HLDS.
Liquid Desiccant System
The absorber and the regenerator are falling film type, internally cooled and heated
respectively. They are comprised by a polypropylene tube bundle, a liquid distributor system
with spray nozzles, and a demister inside a fiber glass tower. Tube bundles are formed of
individual modules of 97 tubes, which are linked horizontally in threes, and then vertically to
form 14 passes, with a total area of 53,2 m2. A proper wettability of the tubes is key for
obtaining good performance in the liquid desiccant cycle, for that reason the polypropylene
tubes have received a plasma treatment in order to improve their wettability [2]. The air flows
from bottom to top getting in contact with the descendent lithium chloride solution, which
forms a falling film outside the tubes. The rich LiCl solution absorbs humidity from air in the
absorber, and the poor LiCl solution desorbs humidity enriching the solution in the
regenerator. Cold water (15ºC) and hot water (55ºC) flows inside the tubes, cooling the
solution and the air in the absorber, and heating the solution and the air in the regenerator.
Models for the absorber and regenerator are based on the theoretical model described by
Gommed and Grossman [3]. Water and salt mass balances, energy balances, thermodynamic
equilibrium and heat and mass transfer equations are included in the model.
ABSORBER
Inlet air
Outlet air
Cold/hot water
to absorber/regenerator
Rich/poor solution
to absorber/regenerator
Recirculation
from LLHX
Liquid distributor
Demister
Solution falling film
Tube bundle
Recirculation
to LLHX
Figure 1: Scheme of the absorber/regenerator units, and the tube bundle layout.
Hot/cold
water inlet
Hot/cold
water outlet
334
CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland
Heat and mass transfer coefficients involved in the modelling are the global heat transfer
coefficient between external medium and the solution (U), the heat and mass transfer
coefficients between air-solution interface and the air stream (a, a), and the heat and mass
transfer coefficients between air-solution interface and the solution stream (s, s).
The heat transfer coefficient between external medium and solution stream is calculated as:




(1)
The water-tube heat transfer coefficient (hwt) has been calculated depending on the flow
regime by using correlations described by Nellis & Klein [4]; using the following correlation
for Re<2300,
 󰇣
 󰇤
󰇟󰇠
(2)
The Gnielinski equation for fully developed flow for 2300<Re<10000,
 
󰇛󰇜
󰇛 󰇜
(3)

󰇟󰇛󰇜󰇠
(4)
And the Dittus-Boelter equation for Re>10000,
N=0,4 for heating of the fluid (absorber)
N=0,3 for cooling of the fluid (regenerator)
(5)
The tube-solution heat transfer coefficient (hts) and the mass transfer coefficient between air-
solution interface and the air stream (a) have been obtained from experimental results of the
field tests developed in Technion (Israel Institute of Technology) in Haifa [5]. In the field
tests a Liquid Desiccant System was experimentally tested using different types of tubes in
absorber mode. The performance with plastic tubes, which showed poor wettability, was
lower than with the titanium tubes, which presented good wettability. The designed HLDS
will be using polypropylene tubes which have received a plasma treatment in order to increase
their wettability [2]. Therefore, the experimental values used for the modelling correspond to
the experiments carried out with titanium tubes.
Assumption of a Lewis factor equal to one is considered, which is applicable for air.


(6)
The liquid-liquid heat exchanger is a plate heat exchanger made of a polymeric matrix
composite including graphene nanoparticles, from SGL company. It is used to precool the
solution going to the absorber, and to preheat the solution going to the regenerator. Its model
is based on the ε-NTU method with correlations for heat transfer in plate heat exchangers
described by H. Martin [6]. The model has been contrasted with the design data from SGL.
Heat exchanger effectiveness for design working conditions is 0,85.
CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland
335
Air Handling Unit and polyvalent unit
The air handling unit comprises the air plate heat exchanger for heat recovery of ventilation,
the cooling coil, and the corresponding fans, dampers and filters for the system. The casing
and the arrangement of the different elements is specially designed for its optimal connection
with the LDS.
The air heat exchanger is a compact plate heat exchanger, with cross-flow configuration,
made of aluminum and with internal fins to increase the heat transfer between both air
streams. The use of such equipment enables considerable savings to be achieved in the
operating costs of air-conditioning plants, and thus the saving of energy that would otherwise
be lost. In the HLDS it is used for preheating the air entering the regenerator. It is modelled
by using the ε-NTU method.
The cooling coil has been modelled by using the model described in [7]. This model is based
on epsilon-NTU and LMTD and LMHD equations and uses Braun's hypothesis to simulate
the behaviour of the cooling coil in an AHU (air handling unit). Normally, cooling coils in Air
Handling Units deal with sensible and latent cooling loads, condensing water from air in the
surface of the tubes. In this case, the cooling coil serves for dealing only with sensible cooling
loads; no condensation will occur in the surface of the tubes, so in theory the dry cooling coil
approach could be used for the modelling. However, the model includes also the wet coil
approach, in case the cooling coil may be used for dehumidification as well. In that case, the
cooling coil should be oversized respect to the actual size. For the air side the correlation
proposed by Wang et al.[8] is selected and the HTC validated against a reference HX
calculation software; on the other hand for the calculation of the tube side coefficient the
correlation proposed by Gnielinski [9] for turbulent region and the described on Nellis &
Klein [4] for the laminar one as in the case of the absorber and regenerator have been used.
The polyvalent unit is a heat pump able to provide heating and cooling simultaneously, by
recovering heat from a water condenser when the machine is working on dual mode, and by
condensing with air when the machine is working on cooling mode. Therefore, it will be used
in cooling mode when the LDS is only dehumidifying and cooling (regenerator off), or in dual
mode when the system is regenerating LiCl solution as well, with no need from another
external source of heat.
WHOLE CYCLE MODELLING AND RESULTS
The individual models of the LDS and the AHU have been combined to develop the complete
model of the HLDS. A split system is included at the solution outlet of the absorber and
regenerator, so that the amount of solution which is recirculated to the absorber and
regenerator from the bottom of each tower is set to 0,9 of the outlet solution from each
component; sending 10% of the solution through the liquid-liquid heat exchanger. The figure
below shows the results from the modelling of the whole HLDS.
Calculated impulsion air conditions are 19,8ºC and a humidity ratio of 6,8 g/kg dry air. In
order to achieve these conditions, the required capacity of the absorber and regenerator is 40,3
kW and 40,8 kW respectively, with a 13,5 kW capacity liquid-liquid heat exchanger. The air
plate heat exchanger has a capacity of 13,3 kW. The cooling coil needs to provide 1,4 kW in
the design conditions, however for the construction of the prototype it will be oversized in
order to be able to deliver the necessary cooling to meet the sensible cooling load, which leads
to 8,5 kW. Considering the electrical consumption of the polyvalent unit, the HLDS has a
COP of 2,7.
336
CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland
Figure 2: Results from simulation of the HLDS
COP of the HLDS and conventional system are calculated as:
 

(7)
 

 
(8)
A representation of the HLDS process in a psychrometric chart is shown below, as well as the
corresponding process to the treatment of air with a conventional system. In the conventional
system, the air needs to be cooled down to a high extent in the cooling coil to reach the
required dehumidification. In order to achieve comfort indoor conditions, the air needs to be
reheated in a reheating coil. With the HLDS process, the need for reheating is eliminated,
which leads to important energy savings. Taking into account the reheating consumption in
the conventional system, an improvement of around 45% in COP is achieved.
Figure 3: Air treatment processes in the HLDS and in a conventional cooling system
represented in the psychrometric chart
0 5 10 15 20 25 30 35
0,000
0,005
0,010
0,015
0,020
0,025
0,030
0,035
0,040
0,045
0,050
T [°C]
Humidity Ratio
Pressure = 101,3 [kPa]
10°C
15°C
20°C
25°C
30°C
35°C
0,2
0,4
0,6
0,8
0,8
0,825
0,85 m3/kg
0,875
0,9
AirH2O
HLDS process
Conventional process
CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland
337
CONCLUSION
Based on the cooling loads calculation, a hybrid liquid desiccant system has been designed
and simulated for a case study in Taiwan, sizing every component to match the sensible and
latent cooling loads. The system achieves a reduction in energy consumption; as unlike
conventional systems it does not need to reach the dew point to remove humidity, with the
consequent improvement in the COP. The designed system has been constructed and is at the
setting-up phase in the laboratory, in order to define the best control strategies for the system.
Future work focuses on its performance monitoring in the real-site demonstration in Taiwan.
Figure 4: Nanocool system
REFERENCES
1. Engineering Equation Solver. F-Chart Software. http://www.fchart.com
2. A. Fina, A. Guerriero, S. Colonna, F. Carosio and G. Saracco. Polymer-based materials
for the application in Liquid Dessicant heat exchangers. V Congreso Iberoamericano de
Ciencias y Técnicas del frío. Tarragona, June 18-20 (2014).
3. K. Gommed, G. Grossman. Improved Simulation Model for Air-Liquid Contactors in
Open Absorption Air Conditioning Systems. Proceedings, the International Sorption Heat
Pump Conference, College Park, MD, USA, March 31-April 3 (2014).
4. G. Nellis, S. Klein. Heat Transfer. Cambridge University Press (2009), 661-671.
5. K. Gommed, G. Grossman, J. Prieto, J. Ortiga, A. Coronas. Experimental Comparison
between Internally and Externally Cooled Air-Solution Contactors. Proceedings, the
International Sorption Heat Pump Conference, College Park, MD, USA, March 31-April
3 (2014).
6. Holger Martin. Pressure Drop and Heat Transfer in Plate Heat Exchangers. VDI Heat
Atlas (2010), 1515-1521.
7. Air-Braun, J.E., Klein, S.A., Mitchell, P.E., Effectiveness models for cooling towers and
cooling coils. ASHRAE Transactions Vol.92 part 2 (1989), p.164-174.
8. C. Wang, Y. Hsieh, Y. Lin. Performance of Plate finned Tube Heat Exchangers Under
Dehumidifying Conditions, Journal of Heat Transfer. Vol.119 nº1 (1997), p109-117
9. V. Gnielinski, New equation for heat and mass transfer in turbulent pipe and channel flow,
International chemical engineering, vol. 16 (1976) pp.359-368.
338
CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland
ResearchGate has not been able to resolve any citations for this publication.
Article
Air-conditioning demand, especially in hot and humid climates, causes most of the energy consumption in buildings. In these applications, the latent load often constitutes a significant part of the total load. Conventional air-conditioning systems deal with the required dehumidification by cooling the moist air below its dew point and reheat it later to increase its temperature to comfortable supply conditions. By using solid/liquid desiccant systems, which absorb humidity from the air, it becomes unnecessary to reach the dew point to dehumidify, thereby reducing energy consumption. In liquid desiccant systems, dehumidification occurs in the absorber, where the humidity of the air is absorbed by direct contact with a desiccant. Most of the previous liquid desiccant system studies have used adiabatic absorbers, where the solution must be cooled externally before being delivered into the air-solution contactor. However, based on experience with closed-cycle absorption systems, internally cooled absorbers are expected to yield better performance considering that a cool solution temperature throughout the absorber will maintain a constant and low vapor pressure. This article presents an experimental study of two different internally cooled air-solution contactors using falling-film tubular heat exchangers: The tubes in one are made of titanium and in the other of high-density poly ethylene. The experiments have been conducted in a liquid desiccant system under summer conditions in the Mediterranean city of Haifa, Israel. A performance comparison is also carried out with an externally cooled air-solution contactor, using results obtained from previous studies.
Article
This textbook provides engineers with the capability, tools and confidence to solve real-world heat transfer problems. It includes many advanced topics, such as Bessel functions, Laplace transforms, separation of variables, Duhamel's theorem and complex combination, as well as high order explicit and implicit numerical integration algorithms. These analytical and numerical solution methods are applied to topics not considered in most textbooks. Examples include heat exchangers involving fluids with varying specific heats or phase changes; heat exchangers in which axial conduction is a concern; and regenerators. Derivations of important results are presented completely, without skipping steps, which reduces student frustration and improves readability and retention. The examples are not trivial 'textbook' exercises; they are rather complex and timely real-world problems that are inherently interesting. This book integrates the computational software packages Maple, MATLAB®, FEHT and Engineering Equation Solver (EES) directly with the heat transfer material.
Article
Systematic studies of continuous fin-and-tube tube heat exchangers under dehumidifying conditions are reported in the present study. The heat exchangers consist of nine fin-and-tube heat exchangers having plane fins. The effects of fin spacing, the number of tube row, and inlet conditions are investigated. Data are presented in terms of j factors and friction factors f. It is found that the inconsistencies in the open literature may be associated with the wet fin efficiency. A correlation is proposed for the present plate fin configuration; this correlation can describe 92 percent of j, and 91 percent of the f data within +/- 10 percent.
Conference Paper
This paper presents a simple, yet mechanistic method for modeling the performance of cooling towers a cooling coils. Through the introduction of an air saturation specific heat, effectiveness relationships are developed. The advantages of this approach are its simplicity, accuracy, and consistency with the methods for analyzing sensible heat exchangers. The accuracy of the effectiveness models is as good or better than that associated with standard methods presented in the literature and these models have significantly less computational requirements.
Polymer-based materials for the application in Liquid Dessicant heat exchangers. V Congreso Iberoamericano de Ciencias y Técnicas del frío
  • A Fina
  • A Guerriero
  • S Colonna
  • F Carosio
  • G Saracco
A. Fina, A. Guerriero, S. Colonna, F. Carosio and G. Saracco. Polymer-based materials for the application in Liquid Dessicant heat exchangers. V Congreso Iberoamericano de Ciencias y Técnicas del frío. Tarragona, June 18-20 (2014).
Improved Simulation Model for Air-Liquid Contactors in Open Absorption Air Conditioning Systems
  • K Gommed
  • G Grossman
K. Gommed, G. Grossman. Improved Simulation Model for Air-Liquid Contactors in Open Absorption Air Conditioning Systems. Proceedings, the International Sorption Heat Pump Conference, College Park, MD, USA, March 31-April 3 (2014).
Experimental Comparison between Internally and Externally Cooled Air-Solution Contactors
  • K Gommed
  • G Grossman
  • J Prieto
  • J Ortiga
  • A Coronas
K. Gommed, G. Grossman, J. Prieto, J. Ortiga, A. Coronas. Experimental Comparison between Internally and Externally Cooled Air-Solution Contactors. Proceedings, the International Sorption Heat Pump Conference, College Park, MD, USA, March 31-April 3 (2014).