TY - JOUR
AU1 - Oyieke, Andrew Y, A
AU2 - Inambao, Freddie, L
AB - Abstract Coupled heat and mass transfer performance of an adiabatic solar-powered liquid desiccant dehumidification and regeneration scheme using lithium bromide(LiBr) solution has been conducted experimentally as well as numerically under subtropical climatic conditions. The application of a vacuum insulated photovoltaic and thermal module to provide desiccant regeneration heat as well as electrical power to drive the air fans and liquid pumps have been explored. A square channelled ceramic cordierite packing with a varying channel density of 20–80 m|$^2$|/m|$^3$| has been used to establish the optimum direct air-LiBr contact ratio for maximum effectiveness. The aggregate crammed vertical dehumidifier and regenerator operational indices featured were effectiveness, moisture removal rate (MRR), heat and mass transfer constants and Lewis number. The influence of solar radiation, humidity and L/G ratios, air–desiccant flow rates and concentration on the indices have been scrutinized in details. A 3D predictive numerical thermal model based on falling liquid stream with constant thickness in counter-flow configuration has been developed and solved by a combination of separative appraisal and stepwise iterative technique. The heat and mass exchange coefficients significantly increased with the increase in Lewis number, air and desiccant flow rates for both the dehumidifier and regenerator vessels. The predicted results of heat and mass transfer coefficients, effectiveness and MRRs have been validated with experimental measurements within a general acceptable conformity of less than |$\pm $|10%. 1 Introduction Liquid desiccant technology has emerged as the most effective and preferred state-of-the-art vapour extraction technique from the processed air within the various dehumidifier and regenerator configurations. The preference is due to the flexibility in operation, elimination of organic and inorganic contaminants and low operating temperatures that favours the use of renewable energy from the sun [1]. The solar radiation supplies the low-grade regeneration heat to the desiccant solution, which loses water particles to the atmospheric air during the direct-contact interaction within the regenerator vessel. The air–desiccant interaction results in water vapour extraction and dispersion from the process air and diluted liquid desiccant, respectively. The water particles exchange process creates a combined thermal and mass transmission phenomenon at the fluid–air interaction phase. Figure 1 Open in new tabDownload slide The diagrammatic representation of the experimental setup. Figure 1 Open in new tabDownload slide The diagrammatic representation of the experimental setup. The experimental tests on solar-driven combined thermal and mass transmission in liquid desiccant regeneration and absorption technology date back to the late sixties [2]. Since then, various modifications of solar regeneration systems have emerged in the forms of experimental evaluations and validations on glazed and unglazed flat plate modules [3], enclosed and open zones, forced flow arrangements [4], psychometrics, feasibility and actualization [5–7]. The heat and mass exchange correlation constants found from investigational assessments of an open-cycle forced convection solar regenerator are obtainable in [8–13]. Combinations of theoretical examination and experimental justification of heat and mass exchange factors during chemical dehumidification and regeneration in different configurations using various desiccant solutions such as lithium chloride (LiCl) [14–17], triethylene glycol (TEG) solution [18], H2O/KCOOH [19], lithium bromide (LiBr) [20, 21] have been explored in details in packed columns. The performance comparison of structured and randomly packed dehumidifier/regenerator [22, 23] under transient conditions [24, 25] have also yielded satisfactory outcomes. Most of the theoretical studies in the literature are based on the adiabatic conditions, unidirectional and uniform thermal and mass exchange areas [26]. Conversely, the presumption that constant thermal and mass transfer zone is equivalent to the explicit expanse of packing, ignores the effects of liquid surface tension; hence, mass transfer is reduced due to insufficient wetting of the packing material surface. Consequently, the impact of the mean solution heat on unitless parameters is unaccounted for as well. The solution to these problems is often obtained by logarithmic and arithmetic temperature difference methods. However, these techniques do not adequately consider the contribution of individual parameters like flow rate, moisture content, temperature and humidity ratio (HR) on the heat and mass exchange constants. Therefore, this work seeks to examine the heat and mass transfer constants and correlations hinged upon the experimental data obtained under local settings and establish the impact of the aforementioned individual factors on the overall operation of the liquid desiccant system in terms of effectiveness and moisture removal rate (MRR) of the dehumidifier/regenerator packed with square channel ceramic cordierite (CC). 2 Experimental setup and procedure The configuration and arrangement of the experiment shown in Figure 1 comprises of four major segments: the airline, desiccant loop, vacuum insulated photovoltaic and thermal (VPV/T) module, dehumidification chamber and regenerator vessel. In the dehumidification chamber, the incoming air temperature and humidity were initially raised to a predetermined level for the packed bed dehumidifier vessel. Air enters the vessel’s lower end and exit from the top while LiBr desiccant solution comes in at the top and exits at the bottom. Within the vertical column, the air–desiccant mixture experiences a counter-flow interaction over a crammed section during which, heat and mass exchange occurs. During the dehumidification process, air losses water vapour to the desiccant solution because of the vapour pressure difference. The chemical reaction between highly concentrated LiBr and water particles is given in Equation 1. The reaction is in two stages; firstly, the endothermic breakage of the ionic bond between Li|$^+$| and Br|$^-$|⁠. Secondly, the exothermic bonding (Li|$^+$|⁠..H|$_2$|O) and (Br|$^-$|⁠..H|$_2$|O) during dissolution; thus, sensible heat is released due to temperature difference. At the same time, at the phase change of water vapour to liquid requires latent heat of condensation, and as a consequence, the air solution mixture temperature is increased. $$\begin{equation} \underbrace{Li^+Br^-}_{ionic\quad bond}(s) + H_2O(l) \leftrightarrows \underbrace{Li^+ (aq) + Br^- (aq)}_ {dilute \quad solution} .\end{equation}$$(1) The reverse happens in the regeneration process as atmospheric air engrosses water particles out of the heated LiBr liquid as a result of vapour pressure gradient. Latent heat of vaporization ensues when water changes phase to vapour and escapes with air while sensible heat results from air–desiccant temperature gradient as Li|$^+$| and Br|$^-$| fuse in an exothermic reaction to become LiBr solution. Due to the reactions, the solution’s concentration increases while temperature reduces and the opposite is exact for exhaust air. Therefore, the heat and mass transfer processes transpire together as a couple and cannot stay de-linked. The dehumidifier and regeneration columns are made of cylindrical vessels of diameter and height of 450 mm and 650 mm, respectively, with dome-shaped ends for solution spray and collection. The air duct was a 100 mm diameter circular cross-section PVC pipe while the desiccant distribution channel was a 19 mm diameter pipe. These geometrical dimensions were determined using the strategies provided in [25]. However, the adverse effects of the dimensional characteristics on the permanence of the whole set up were out of the scope for the present study. CC honey-comb packing shaped into the perforated square channel of sides 1 |$\pm $| 0.5 cm and specific surface area of 180 m|$^2$|/m|$^3$| was used to provide heat and mass exchange surface. The CC exhibits pronounced absorbency, excellent liquescent and even dispersal, extraordinary heat and biochemical oxidization resistance, low pressure-drop, outstanding separation efficiency and highly adoptive to regeneration conditions as well as a small thermal coefficient. The properties mentioned above informed the choice and consideration of CC a packing material for this study. The geometry of the channels, as shown in Figure 2, was designed to allow pressure to drop to create partial vapour pressure conducive for heat and mass transmission occurrence in the course of dehumidification and regeneration procedures. Figure 2 Open in new tabDownload slide The CC honey comb packing. Figure 2 Open in new tabDownload slide The CC honey comb packing. The gas-phase mixture flows in a straight channel while the liquid-phase film flows on the packing material surface under gravitational influence. The specific thermophysical properties of the CC packing is provided in Table 1. Two solution storage tanks were connected to the setup; the strong solution tank (SST) contained the concentrated solution supplied to the dehumidifier for water vapour removal, and the weak solution tank (WST) contained the weak/diluted solution emanating from the dehumidifier to be propelled towards the regenerator unit for improved concentration to near initial levels. The solution temperature from the SST to the dehumidifier was controlled at sub-minimum (below ambient). The regenerated fluid was then passed to the concentrated solution tank and the cycle continues. Table 1 The thermophysical properties of packing material. Property . Value . Parking Honeycomb ceramic substrate Material Ceramic cordierite Nominal surface to unit volume ratio 180 m|$^2$|/m|$^3$| Cross-sectional area 0.164 m|$^2$| Void fraction 0.86 Equivalent diameter 200 mm Column height 650 mm Wall thickness 1.5 mm Cell density (cell per square inch) 100 Porosity 85% Max working temperature 1400|$^o$|C Average aparture 7-15 Cell dimensions 10 mm x 10 mm Chemical components |$Al_2O_3$| (35.4%), |$SiO_2$| (50.4%) and |$MgO$| (13.5%) Property . Value . Parking Honeycomb ceramic substrate Material Ceramic cordierite Nominal surface to unit volume ratio 180 m|$^2$|/m|$^3$| Cross-sectional area 0.164 m|$^2$| Void fraction 0.86 Equivalent diameter 200 mm Column height 650 mm Wall thickness 1.5 mm Cell density (cell per square inch) 100 Porosity 85% Max working temperature 1400|$^o$|C Average aparture 7-15 Cell dimensions 10 mm x 10 mm Chemical components |$Al_2O_3$| (35.4%), |$SiO_2$| (50.4%) and |$MgO$| (13.5%) Open in new tab Table 1 The thermophysical properties of packing material. Property . Value . Parking Honeycomb ceramic substrate Material Ceramic cordierite Nominal surface to unit volume ratio 180 m|$^2$|/m|$^3$| Cross-sectional area 0.164 m|$^2$| Void fraction 0.86 Equivalent diameter 200 mm Column height 650 mm Wall thickness 1.5 mm Cell density (cell per square inch) 100 Porosity 85% Max working temperature 1400|$^o$|C Average aparture 7-15 Cell dimensions 10 mm x 10 mm Chemical components |$Al_2O_3$| (35.4%), |$SiO_2$| (50.4%) and |$MgO$| (13.5%) Property . Value . Parking Honeycomb ceramic substrate Material Ceramic cordierite Nominal surface to unit volume ratio 180 m|$^2$|/m|$^3$| Cross-sectional area 0.164 m|$^2$| Void fraction 0.86 Equivalent diameter 200 mm Column height 650 mm Wall thickness 1.5 mm Cell density (cell per square inch) 100 Porosity 85% Max working temperature 1400|$^o$|C Average aparture 7-15 Cell dimensions 10 mm x 10 mm Chemical components |$Al_2O_3$| (35.4%), |$SiO_2$| (50.4%) and |$MgO$| (13.5%) Open in new tab The photovoltaic/thermal solar collector consists of the photovoltaic cells mounted on a heat-absorbing copper plate. Beneath the copper plate, there is a serpentine-shaped copper tubing for fluid circulation, thereby, absorbing the excess heat from the plate. The primary purpose of the VPV/T collector was to provide regeneration heat as well as electrical energy to drive the circulation pumps and fans. The vacuum insulation provided more heat retention and absorption by the desiccant solution flowing in the pipes beneath the absorber plate. The fluid circulation pump and air fans are operated by direct current power from the VPV/T. The electrical energy produced was stored in two regulated sealed valve 12 V, 100 Ah batteries (B), from which the power was connected to the air blowers and fluid circulation pumps (P|$_{ws}$| and P|$_{ss}$|⁠), thus giving the setup a self-powering characteristic. The heat transfer and power generation characteristics of the VIP/T had previously been evaluated in [27] and therefore, will not be detailed in the present study. The dimensions and specifications of the PVT are given in Table 2 Table 2 VPV/T collector specifications. Collector property . Thermal . Collector property . Electrical . Collector dimensions 1.640 m|$\times $| 0.87 m|$\times $| 0.105 m PV module power 180 W Collector area 1.42 m|$^2$| PV cell type Mono-crystalline Collector slope 35|$^{o}$| PV cell dimension 0.125 m|$\times $| 0.125 m Absorber type Sheet and tube No. of cells 72 Absorber plate thickness 0.001 m PV cell encapsulation Non-encapsulated Absorber plate material Copper Packing factor 1.0 Internal piping Copper Top insulation medium Vacuum Riser tube diameter 0.012 m Bottom insulation material Fibre wool Riser tube thickness 0.007 m Insulation thickness 0.05 m Header tube diameter 0.022 m Glass cover (tempered) 0.004 m low iron Header tube thickness 0.008 m Number of tubes and spacing 14 and 120 mm Collector property . Thermal . Collector property . Electrical . Collector dimensions 1.640 m|$\times $| 0.87 m|$\times $| 0.105 m PV module power 180 W Collector area 1.42 m|$^2$| PV cell type Mono-crystalline Collector slope 35|$^{o}$| PV cell dimension 0.125 m|$\times $| 0.125 m Absorber type Sheet and tube No. of cells 72 Absorber plate thickness 0.001 m PV cell encapsulation Non-encapsulated Absorber plate material Copper Packing factor 1.0 Internal piping Copper Top insulation medium Vacuum Riser tube diameter 0.012 m Bottom insulation material Fibre wool Riser tube thickness 0.007 m Insulation thickness 0.05 m Header tube diameter 0.022 m Glass cover (tempered) 0.004 m low iron Header tube thickness 0.008 m Number of tubes and spacing 14 and 120 mm Open in new tab Table 2 VPV/T collector specifications. Collector property . Thermal . Collector property . Electrical . Collector dimensions 1.640 m|$\times $| 0.87 m|$\times $| 0.105 m PV module power 180 W Collector area 1.42 m|$^2$| PV cell type Mono-crystalline Collector slope 35|$^{o}$| PV cell dimension 0.125 m|$\times $| 0.125 m Absorber type Sheet and tube No. of cells 72 Absorber plate thickness 0.001 m PV cell encapsulation Non-encapsulated Absorber plate material Copper Packing factor 1.0 Internal piping Copper Top insulation medium Vacuum Riser tube diameter 0.012 m Bottom insulation material Fibre wool Riser tube thickness 0.007 m Insulation thickness 0.05 m Header tube diameter 0.022 m Glass cover (tempered) 0.004 m low iron Header tube thickness 0.008 m Number of tubes and spacing 14 and 120 mm Collector property . Thermal . Collector property . Electrical . Collector dimensions 1.640 m|$\times $| 0.87 m|$\times $| 0.105 m PV module power 180 W Collector area 1.42 m|$^2$| PV cell type Mono-crystalline Collector slope 35|$^{o}$| PV cell dimension 0.125 m|$\times $| 0.125 m Absorber type Sheet and tube No. of cells 72 Absorber plate thickness 0.001 m PV cell encapsulation Non-encapsulated Absorber plate material Copper Packing factor 1.0 Internal piping Copper Top insulation medium Vacuum Riser tube diameter 0.012 m Bottom insulation material Fibre wool Riser tube thickness 0.007 m Insulation thickness 0.05 m Header tube diameter 0.022 m Glass cover (tempered) 0.004 m low iron Header tube thickness 0.008 m Number of tubes and spacing 14 and 120 mm Open in new tab Table 3 Measuring instruments and specifications. Instrument . Uncertainty . Range . Parameter . Thermocouple (T-type) 0.1|$^o$|C 0–80|$^o$|C Temperature Thermocouple (K-type) 0.2|$^o$|C -50–+50|$^o$|C Dew point temperature Ultrasonic flow meter 0.1% 0–1600 kg/h Mass flow rate Density meter 0.1 kg/m|$^3$| 1–9999 kg/m|$^3$| Density Barometer 0.08% fs 0–10 mbars Pressure Digital flow meter |$\pm $| 1% fsd 0–800 m|$^3$|/h Fluid volume flow rate Pressure transducer 0.1% fs 0–10 mbars Pressure drop Instrument . Uncertainty . Range . Parameter . Thermocouple (T-type) 0.1|$^o$|C 0–80|$^o$|C Temperature Thermocouple (K-type) 0.2|$^o$|C -50–+50|$^o$|C Dew point temperature Ultrasonic flow meter 0.1% 0–1600 kg/h Mass flow rate Density meter 0.1 kg/m|$^3$| 1–9999 kg/m|$^3$| Density Barometer 0.08% fs 0–10 mbars Pressure Digital flow meter |$\pm $| 1% fsd 0–800 m|$^3$|/h Fluid volume flow rate Pressure transducer 0.1% fs 0–10 mbars Pressure drop Open in new tab Table 3 Measuring instruments and specifications. Instrument . Uncertainty . Range . Parameter . Thermocouple (T-type) 0.1|$^o$|C 0–80|$^o$|C Temperature Thermocouple (K-type) 0.2|$^o$|C -50–+50|$^o$|C Dew point temperature Ultrasonic flow meter 0.1% 0–1600 kg/h Mass flow rate Density meter 0.1 kg/m|$^3$| 1–9999 kg/m|$^3$| Density Barometer 0.08% fs 0–10 mbars Pressure Digital flow meter |$\pm $| 1% fsd 0–800 m|$^3$|/h Fluid volume flow rate Pressure transducer 0.1% fs 0–10 mbars Pressure drop Instrument . Uncertainty . Range . Parameter . Thermocouple (T-type) 0.1|$^o$|C 0–80|$^o$|C Temperature Thermocouple (K-type) 0.2|$^o$|C -50–+50|$^o$|C Dew point temperature Ultrasonic flow meter 0.1% 0–1600 kg/h Mass flow rate Density meter 0.1 kg/m|$^3$| 1–9999 kg/m|$^3$| Density Barometer 0.08% fs 0–10 mbars Pressure Digital flow meter |$\pm $| 1% fsd 0–800 m|$^3$|/h Fluid volume flow rate Pressure transducer 0.1% fs 0–10 mbars Pressure drop Open in new tab There was no need for external evaporator in the whole setup as presented since the desorption system was incorporated. The pressure was atmospheric throughout the setup; thus, the expensive pressure vessel components were eliminated. Apart from the use of low-grade heat energy from the sun, the system was self-powered as the electrical and thermal power from the VPV/T module was utilized. However, an auxiliary heater was incorporated to augment the regeneration heat during periods of no sunshine. The desiccant heat and concentration in the feeder tank were kept unchanged at pre-set conditions. The strong solution pump (P|$_{ss}$|⁠) transferred the solution to the spray chamber at the upper part of the dehumidification column where it was dispensed as a mist to the top of the packed bed and allowed to flow freely by gravity. The air and desiccant interaction took place in the packing chamber where thermal and mass exchange transpired as water vapour in air was transferred to the desiccant. The water absorption process enhanced the moisture content in the liquid exiting at the bottom of the packed bed as a diluted solution which then flowed to the WST. From the WST, the weak solution was pumped to the bottom of the VPV/T module through which it steadily flowed as it was heated by solar energy depending on the solar radiation intensity. A monitored heated solution stream was then transferred to the regenerator in a similar configuration to the dehumidifier while the returning solution was collected and directed to the SST. At raised temperature, the weak solution readily lost water vapour to the air in a counter flow configuration through a packed bed in the regenerator vessel. 2.1 Instrumentation Various parameters were monitored throughout the entire duration of the experiment in order to establish their various effects and dependability on other parameters and factors. The parameters of interest that were monitored and measured includes flow rates and temperatures of both air and desiccant at inlets and outlets of the vessels, the HR of air, desiccant solution density among others. The various measurements were achieved with the following instruments listed in Table 3. In the air loop, measurements of temperature (T|$_a$|⁠), humidity ratio (H) and flow rate (F|$_a$|⁠) were performed both on the dehumidifier and regenerator inlet and outlets. There was two temperature, and dew point plugs restrained by T-type and K-type thermocouples respectively installed at suitable positions in the airway to sample the temperatures. The strain-gauge differential pressure transducer was used for pressure-drop (⁠|$\vartriangle $|P) measurement along the height of the vertical column. On the other hand, the airflow rate was measured by an ultrasonic flow meter in combination with a strain gauge pressure transducer while the absolute pressure was monitored by use of a barometer. In the desiccant loop, temperatures and flow rates were always monitored both at the inlet and outlet conditions for the dehumidifier and regenerator vessels, respectively. At least two probe ports equipped with K-type thermocouples (T|$_d$|⁠) were installed on either side of inlet and outlet. Similarly, the desiccant flow rate was monitored using a digital flow-tech meter (EMFM-9) (F|$_d$|⁠) with inbuilt data acquisition at inlet and outlet of each vessel. In order to determine the solution concentration, the density was sampled by use of a DS7800 density meter. Before each run, both the absorption and desorption column assemblies were cleansed using clean, freshwater and blown dry by a stream of warm air. The readings were logged at an interval of 15 seconds while the densities were fed into a computer algorithm to determine the concentration Before regenerator measurements, the weak solution was stored continuously in the storage tank to equalize the conditions and for uniform distribution of the temperature and concentration. The desiccant was made to flow at meagre flow rates while the airflow rate was at maximum. The flow rates were then fixed at predetermined figures, then temperature, humidity and vapour pressure values were recorded at steady states. The process was repeated until the desiccant was restored to near its initial concentration. The data collected were then applied in a computer algorithm to determine the heat and mass exchange constants along the columns, change in the moisture content of the air as well as the change in temperatures between the inlet and outlet conditions corresponding to various flow rates. 2.2 Experimental data uncertainties The purpose of experimental uncertainty analysis was to scrutinize the calculated parameters, focusing on the inconsistencies in the measured parameters, applied in form of mathematical formulation. The theoretical relationships used to transform the measurements into the derived quantity are prone to bias and unavoidable indiscriminate disparities resulting from repetitive measurements. For reliability, accuracy and precision of the measured parameters such as temperature (air and desiccant), flow rate (mass and volume), density and pressure, the propagation of the errors caused by the bias and disparities into the derived quantities must be avoided. Therefore, an approximation technique that offers consistent and valuable outcome is required. The uncertainty analysis of derived determinant parameters such as moisture removal rate (MRR,|$\epsilon $| ), effectiveness (⁠|$\varepsilon $|⁠) and enthalpy (⁠|$h$|⁠) was carried out for normalization using the equations adapted from [28] whose general form is given in Equation 2. The respective estimated values of uncertainties were |$\pm $| 3.5%(⁠|$\epsilon $|⁠), |$\pm $| 2.8% (⁠|$\varepsilon $|⁠) and |$\pm $| 3.8% (⁠|$h$|⁠) while those of desiccant concentration, desiccant temperature, air HR, air velocity and temperature were estimated to be |$\pm $| 0.15%, |$\pm $| 0.2|$^o$|C, |$\pm $| 0.2%, |$\pm $| 0.1 m/s and |$\pm $| 0.2|$^o$|C, respectively. $$\begin{align}& \Delta y =\nonumber\\ &\!\biggl\lbrace\! \biggl(\!\frac{\partial y}{\partial x_1}\Delta x_1\!\biggr)^2 \!\!+\!\! \biggl(\!\frac{\partial y}{\partial x_2}\Delta x_2\!\biggr)^2 \!\!+ \!\!\biggl(\!\frac{\partial y}{\partial x_3}\Delta x_3\!\biggr)^2 \!\!+ \!\ldots\ldots +\!\! \biggl(\!\frac{\partial y}{\partial x_n}\Delta x_n\!\biggr)^2\biggr\rbrace^2 \!,\end{align}$$(2) where |$x_1$|⁠, |$x_2$|⁠, |$x_3$|⁠......|$x_n$| are various contributing parameters and y is the determined parameter. The respective individual expressions for enthalpy, effectiveness and MRR are given in Table 4. Table 4 Uncertainties of experimental data. Parameter . Uncertainty equation . Enthalpy of air (⁠|$h_a$|⁠) |$\Delta h_a = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial h_a}{\partial \omega }\biggr )^2 + 2\biggl (\Delta T_a\frac{\partial h_a}{\partial T_a}\biggr )^2\biggr \rbrace ^{1/2}$| Enthalpy of desiccant (⁠|$h_d$|⁠) |$\Delta h_d = \biggl \lbrace 2\biggl (\Delta \chi \frac{\partial h_a}{\partial \chi }\biggr )^2 + 2\biggl (\Delta T_d\frac{\partial h_a}{\partial T_d}\biggr )^2\biggr \rbrace ^{1/2}$| Moisture removal rate (⁠|$\epsilon $|⁠) |$\Delta \epsilon = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial \epsilon }{\partial \omega }\biggr )^2 + \biggl (\Delta v\frac{\partial \epsilon }{\partial v}\biggr )^2 + \biggl (\Delta d\frac{\partial \epsilon }{\partial d}\biggr )^2\biggr \rbrace ^{1/2}$| Effectiveness (⁠|$\varepsilon $|⁠) |$\Delta \varepsilon = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial \varepsilon }{\partial \omega }\biggr )^2 \biggr \rbrace ^{1/2}$| Parameter . Uncertainty equation . Enthalpy of air (⁠|$h_a$|⁠) |$\Delta h_a = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial h_a}{\partial \omega }\biggr )^2 + 2\biggl (\Delta T_a\frac{\partial h_a}{\partial T_a}\biggr )^2\biggr \rbrace ^{1/2}$| Enthalpy of desiccant (⁠|$h_d$|⁠) |$\Delta h_d = \biggl \lbrace 2\biggl (\Delta \chi \frac{\partial h_a}{\partial \chi }\biggr )^2 + 2\biggl (\Delta T_d\frac{\partial h_a}{\partial T_d}\biggr )^2\biggr \rbrace ^{1/2}$| Moisture removal rate (⁠|$\epsilon $|⁠) |$\Delta \epsilon = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial \epsilon }{\partial \omega }\biggr )^2 + \biggl (\Delta v\frac{\partial \epsilon }{\partial v}\biggr )^2 + \biggl (\Delta d\frac{\partial \epsilon }{\partial d}\biggr )^2\biggr \rbrace ^{1/2}$| Effectiveness (⁠|$\varepsilon $|⁠) |$\Delta \varepsilon = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial \varepsilon }{\partial \omega }\biggr )^2 \biggr \rbrace ^{1/2}$| Open in new tab Table 4 Uncertainties of experimental data. Parameter . Uncertainty equation . Enthalpy of air (⁠|$h_a$|⁠) |$\Delta h_a = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial h_a}{\partial \omega }\biggr )^2 + 2\biggl (\Delta T_a\frac{\partial h_a}{\partial T_a}\biggr )^2\biggr \rbrace ^{1/2}$| Enthalpy of desiccant (⁠|$h_d$|⁠) |$\Delta h_d = \biggl \lbrace 2\biggl (\Delta \chi \frac{\partial h_a}{\partial \chi }\biggr )^2 + 2\biggl (\Delta T_d\frac{\partial h_a}{\partial T_d}\biggr )^2\biggr \rbrace ^{1/2}$| Moisture removal rate (⁠|$\epsilon $|⁠) |$\Delta \epsilon = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial \epsilon }{\partial \omega }\biggr )^2 + \biggl (\Delta v\frac{\partial \epsilon }{\partial v}\biggr )^2 + \biggl (\Delta d\frac{\partial \epsilon }{\partial d}\biggr )^2\biggr \rbrace ^{1/2}$| Effectiveness (⁠|$\varepsilon $|⁠) |$\Delta \varepsilon = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial \varepsilon }{\partial \omega }\biggr )^2 \biggr \rbrace ^{1/2}$| Parameter . Uncertainty equation . Enthalpy of air (⁠|$h_a$|⁠) |$\Delta h_a = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial h_a}{\partial \omega }\biggr )^2 + 2\biggl (\Delta T_a\frac{\partial h_a}{\partial T_a}\biggr )^2\biggr \rbrace ^{1/2}$| Enthalpy of desiccant (⁠|$h_d$|⁠) |$\Delta h_d = \biggl \lbrace 2\biggl (\Delta \chi \frac{\partial h_a}{\partial \chi }\biggr )^2 + 2\biggl (\Delta T_d\frac{\partial h_a}{\partial T_d}\biggr )^2\biggr \rbrace ^{1/2}$| Moisture removal rate (⁠|$\epsilon $|⁠) |$\Delta \epsilon = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial \epsilon }{\partial \omega }\biggr )^2 + \biggl (\Delta v\frac{\partial \epsilon }{\partial v}\biggr )^2 + \biggl (\Delta d\frac{\partial \epsilon }{\partial d}\biggr )^2\biggr \rbrace ^{1/2}$| Effectiveness (⁠|$\varepsilon $|⁠) |$\Delta \varepsilon = \biggl \lbrace 2\biggl (\Delta \omega \frac{\partial \varepsilon }{\partial \omega }\biggr )^2 \biggr \rbrace ^{1/2}$| Open in new tab 2.3 Performance indicators The overall dehumidifier and regenerator performance were assessed using effectiveness, MRR and enthalpy. The effectiveness is given as a percentage ratio of the tangible air humidity ration difference to the extreme possible as expressed in Equations 3 and 4 for the dehumidifier and regenerator. $$\begin{equation} \varepsilon_{deh} = \biggl(\frac{\omega_i - \omega_o}{\omega_i - \omega_e}\biggr)\times 100\% \end{equation}$$(3) $$\begin{equation} \varepsilon_{reg} = \biggl(\frac{\omega_o - \omega_i}{\omega_e - \omega_i}\biggr)\times 100\% ,\end{equation}$$(4) where, |$\omega $| is the HR in |$kg/kg_{dry air}$| while subscripts |$R$|⁠, |$D$|⁠, |$i$|⁠, |$o$| and |$e$| are an inlet, outlet and equilibrium conditions, respectively, and |$\omega _e$| is stated in relation to the inlet desiccant temperature |$T_d$| and the atmospheric pressure |$P_a$| as follows: $$\begin{equation} \omega_e = 0.622\left[\frac{0.6107\exp{\biggl(\frac{17.27T_d}{T_d - 237.3}\biggr)}}{p_a - 0.6107\exp\biggl(\frac{17.27T_d}{T_d - 237.3}\biggr)}\right] .\end{equation}$$(5) The extreme range of HR was accomplished when air outlet partial vapour pressure was equivalent to the inlet solution’s saturation pressure to the column. The results of the experiment are captured for the solution concentration, regeneration effectiveness and pressure-drop on the air-loop moisture removal rate |$\epsilon $| is by definition, is directly proportional to the HR difference at outlet and inlet conditions so long as the mass flow rate |$\dot m_a$| and |$\dot m_d$| are constants. Equations 6 and 7 show the mathematical formulations in terms of desiccant |$\chi $| concentration as well. $$\begin{equation} \epsilon_{deh} = \dot m_a(\omega_i - \omega_o) = \dot m_d\biggl(\frac{\chi_{i}}{\chi_{o}} - 1\biggr) \Rightarrow \chi_{i}>\chi_{o} \end{equation}$$(6) and $$\begin{equation} \epsilon_{reg} = \dot m_a(\omega_o - \omega_i) = \dot m_d\biggl(\frac{\chi_{o}}{\chi_{i}} - 1\biggr) \Rightarrow \chi_{o}>\chi_{i} ,\end{equation}$$(7) where the subscripts |$a$| and |$d$| represent states of air and desiccant, respectively. The desiccant’s concentration defines the proportion and quantity of vapour expended to or engrossed from the air. The expressions in Equations 8 and 9 can be applied for the dehumidifier and regenerator respectively to determine the outlet concentration |$\chi $|⁠, $$\begin{equation} \chi_{o,deh} = \frac{\chi_{d,i}}{1+\biggl(\frac{\varphi}{\dot{M}_a}\biggr)} \end{equation}$$(8) $$\begin{equation} \chi_{o,reg} = \frac{\chi_{d,i}}{1-\biggl(\frac{\varphi}{\dot{M}_a}\biggr)} .\end{equation}$$(9) It was assumed that the concentration of the desiccant at the dehumidifier outlet was the same as that of the regenerator inlet. Therefore, the heat energy balance for the VPV/T becomes $$\begin{equation} G_cdx = \dot m_adh_a + \dot m_ddh_d +U_L(T_d - T_{amb}) + mh_{fg} ,\end{equation}$$(10) where |$h_a$| and |$h_d$| are the air and desiccant enthalpies in J/kg, |$m$| represents the amount of vapour removed in kg, |$h_{fg}$| is latent energy of vaporization, |$U_L$| is the total heat-loss constant, |$T_d$| and |$T_{amb}$| are the solutions and ambient temperatures. |$G_c$| is the radiation available to the solution from the VPV/T calculated as a function of the total solar radiation |$G_T$|⁠, electrical efficiency |$\eta _{el}$| and collector thermal properties as adapted from [29]. $$\begin{equation} G_c = G_T[(1-\rho)\tau\alpha-(\tau\alpha-\tau_{pv}\eta_{el})], \end{equation}$$(11) where |$\tau $|⁠, |$\alpha $| and |$\rho $| are transmittance, absorptance and reflectance of the PV cell. The interfacial equilibrium energy balance during dehumidification and regeneration is given by Equations 12 and 13. $$\begin{equation} Q_{e,deh} = -\dot{m_d}dh_d - \dot{m_a}dh_a \end{equation}$$(12) $$\begin{equation} Q_{e,reg} = \dot{m_a}dh_a - \dot{m_d}dh_d, \end{equation}$$(13) where the difference in enthalpies of air and desiccant are given in Equations 14 and 15. $$\begin{equation} dh_a = cp_a(T-T_{ref}) + Cp_d(T_d-T_{ref}) + Q_e d\omega. \end{equation}$$(14) The change in solution enthalpy is given in terms of the regeneration heat and total heat gain at saturation point |$T_{ref}$|⁠. $$\begin{equation} dh_d = cp_d(T_d-T_{ref}) + Q_{sol}d\omega_{sat}, \end{equation}$$(15) where |$\omega _{sat}$| is the air HR at full vapour capacity point, which depends on the desiccant concentration and temperature difference. Considering the interfacial interaction of air and desiccant solution, we can obtain the latent heat ratio of the dehumidifier and regenerator as a fraction of latent heat generated to the overall heat transferred among the air and desiccant liquid stated as follows [28]. $$\begin{equation} \xi = \frac{\dot{m}_a\delta d\omega}{\dot{m}_acp_adT_a + \dot{m}_a\delta d\omega} = \frac{\dot{m}_d\delta d\omega_{sat}}{\dot{m}_dcp_ddT_d + \dot{m}_d\delta d\omega_{sat}}. \end{equation}$$(16) 3 Experimental results and discussion The main determinant parameter upon which the performances of the dehumidifier and regenerator are based on the air HR. It is therefore very essential to ascertain the effect of humidity levels on the enthalpy changes of both desiccant and air. The HRs were extracted from using the psychometric principle integrated into engineers equation solver software for the points corresponding to the temperatures and humidities obtained from the experiment. The change in enthalpies was evaluated using Equations 14 and 15 and the ensuing values used to analyse the effects of air HR. From the experimental results obtained under South African subtropical climate in the coastal city Durban, the change in HR of inlet air was carefully monitored within the ranges included in Table 5 and plotted against the changes in the respective enthalpies of air and LiBr both at entrance and exit condition as shown in Figures 3 and 4. Table 5 Typical quantities and series of working and reference parameters. Parameter . Symbol . Unit . Dehumidifier . Regenerator . Range . Reference . Range . Reference . Air inlet temperature |$T_a$| |$^{o}C$| 21.02–37.47 31 21.02–37.47 31 Air flow rate |$\dot{m}_a$| |$kg/s$| 0.51–3.2 2.2 0.51–3.2 2.2 Air inlet humidity ratio |$\omega _i$| |$kg_w/kg_a$| 0.0157–0.0347 0.0214 0.0157–0.0347 0.0214 Desiccant temperature |$T_d$| |$^{o}C$| 24–35 29 50–68 62 Desiccant flow rate |$\dot{m}_d$| |$kg/s$| 0.39–0.85 0.62 0.39–0.85 0.62 Desiccant concentration |$\chi $| |$\% wt$| 50–75 54 50–75 50 Parameter . Symbol . Unit . Dehumidifier . Regenerator . Range . Reference . Range . Reference . Air inlet temperature |$T_a$| |$^{o}C$| 21.02–37.47 31 21.02–37.47 31 Air flow rate |$\dot{m}_a$| |$kg/s$| 0.51–3.2 2.2 0.51–3.2 2.2 Air inlet humidity ratio |$\omega _i$| |$kg_w/kg_a$| 0.0157–0.0347 0.0214 0.0157–0.0347 0.0214 Desiccant temperature |$T_d$| |$^{o}C$| 24–35 29 50–68 62 Desiccant flow rate |$\dot{m}_d$| |$kg/s$| 0.39–0.85 0.62 0.39–0.85 0.62 Desiccant concentration |$\chi $| |$\% wt$| 50–75 54 50–75 50 Open in new tab Table 5 Typical quantities and series of working and reference parameters. Parameter . Symbol . Unit . Dehumidifier . Regenerator . Range . Reference . Range . Reference . Air inlet temperature |$T_a$| |$^{o}C$| 21.02–37.47 31 21.02–37.47 31 Air flow rate |$\dot{m}_a$| |$kg/s$| 0.51–3.2 2.2 0.51–3.2 2.2 Air inlet humidity ratio |$\omega _i$| |$kg_w/kg_a$| 0.0157–0.0347 0.0214 0.0157–0.0347 0.0214 Desiccant temperature |$T_d$| |$^{o}C$| 24–35 29 50–68 62 Desiccant flow rate |$\dot{m}_d$| |$kg/s$| 0.39–0.85 0.62 0.39–0.85 0.62 Desiccant concentration |$\chi $| |$\% wt$| 50–75 54 50–75 50 Parameter . Symbol . Unit . Dehumidifier . Regenerator . Range . Reference . Range . Reference . Air inlet temperature |$T_a$| |$^{o}C$| 21.02–37.47 31 21.02–37.47 31 Air flow rate |$\dot{m}_a$| |$kg/s$| 0.51–3.2 2.2 0.51–3.2 2.2 Air inlet humidity ratio |$\omega _i$| |$kg_w/kg_a$| 0.0157–0.0347 0.0214 0.0157–0.0347 0.0214 Desiccant temperature |$T_d$| |$^{o}C$| 24–35 29 50–68 62 Desiccant flow rate |$\dot{m}_d$| |$kg/s$| 0.39–0.85 0.62 0.39–0.85 0.62 Desiccant concentration |$\chi $| |$\% wt$| 50–75 54 50–75 50 Open in new tab Figure 3 Open in new tabDownload slide The consequence of inlet air HR on (a) air and (b) desiccant enthalpy of the dehumidifier. Figure 3 Open in new tabDownload slide The consequence of inlet air HR on (a) air and (b) desiccant enthalpy of the dehumidifier. Figure 4 Open in new tabDownload slide The consequence of inlet air HR on (a) air and (b) desiccant enthalpy of the regenerator. Figure 4 Open in new tabDownload slide The consequence of inlet air HR on (a) air and (b) desiccant enthalpy of the regenerator. During the dehumidification process, it was detected that the upsurge in air HR resulted in increased enthalpy of inlet air while the outlet air enthalpy significantly reduced, as shown in Figure 3a. The reduction in enthalpy was attributed to the proportionate increase in air vapour pressure due to higher HR. At higher vapour pressures, the magnitude of latent energy of condensation liberated to the air becomes larger, thus more water vapour is engrossed by the LiBr solution. However, for an increase of HR from 0.015 to 0.035 |$kg_w/kg_a$|⁠, there was an increase in enthalpy gradient of 4.11% to 27.59% along the dehumidifier height and 16% overall reduction in air enthalpy between the inlet and outlet. On the other hand, the inlet solution enthalpy showed less sensitivity to the escalation in air inlet HR while the exit solution enthalpy considerably increased, as shown in Figure 3b. The overall increase in the inlet to outlet solution enthalpy difference was 21.55%, gradually varying from 14.08% to 29.01% within the HR range. Due to the increase in HR, there is a high likelihood of moisture absorption by LiBr from the air, hence the upsurge of interfacial heat transfer potential. During regeneration, a similar increase in air HR increased inlet air enthalpy whereas, the outlet air enthalpy reduced as shown in Figure 4a. There was an overall reduction of enthalpy of 37.26% between inlet and outlet conditions. A progressive reduction in enthalpy difference of 46.15% to 27.21% was realized. This reduction was significantly higher than the dehumidification process due to the higher air-desiccant temperature gradient during regeneration. Subsequently, the interfacial vapour pressure variation and the prospective latent heat exchange reduces along the height of the regenerator vessel. Similarly, from Figure 4b, the inlet and outlet solution enthalpy changed at approximately 9.29% with a gradual increase of 8.49% to 10.09% along the height of the regenerator vessel. The increase in HR resulted in increased outlet and inlet solution enthalpies. The higher vapour pressure as a result of high HR gave rise to low moisture removal from the desiccant. Consequently, less interfacial heat transfer is experienced and hence, the slight prospect of moisture extraction from the air by the desiccant. The general observation is that, as the HR increases, the solution enthalpy gradient from the packed vessel inlet increases towards the outlet through the dehumidifier height while, in the regenerator, the enthalpy gradient reduces. Within the specified array of inlet conditions, with the increased air HR, the LiBr concentration reduces considerably in the dehumidifier as a result of more water vapour desorption from the conditioned air. On the other hand, the regeneration vessel experiences an increased outlet desiccant concentration due to vaporization of water particle into the atmosphere. It, therefore, means that, as the HR increases, the magnitude of the change in desiccant concentration from the inlet to exit of the crammed vessel reduces and increases in the dehumidifier and regenerator, respectively. A summary of the analysed outcomes is presented in Table 6. Concerning the outlet air HR for both dehumidification and regeneration vessels, as the incoming air HR increases, the outgoing air HR reduces and increases considerably in the dehumidifier and regenerator respectively due to increased water vapour in the atmospheric air. The LiBr solution is, therefore, able to absorb as much water as possible to the saturation levels in the dehumidifier. However, in the regenerator, the air is not able to accommodate more water vapour from the LiBr solution due to the truncated vapour-pressure variance. The air–desiccant interface, therefore, becomes saturated inhibiting further heat and mass exchange. Table 6 Effects of air HR on air and desiccant enthalpy. Parameter . Dehumidifier . Regenerator . |$\omega _i$| . Inlet to outlet % (⁠|$\uparrow /\downarrow $|⁠) . |$\omega _i$| . Inlet to outlet% (⁠|$\uparrow /\downarrow $|⁠) . |$\omega _{i_{(min)}}(\uparrow )$| . |$\omega _{i_{(max)}}(\uparrow )$| . |$\omega _{i_{(min)}}(\uparrow )$| . |$\omega _{i_{(max)}}(\uparrow )$| . Air enthalpy (⁠|$kJ/kg$|⁠) 16.1%(⁠|$\downarrow $|⁠) 4.11%(⁠|$\uparrow $|⁠) 27.6%(⁠|$\uparrow $|⁠) 37.3%(⁠|$\uparrow $|⁠) 27.21%(⁠|$\uparrow $|⁠) 46.14%(⁠|$\uparrow $|⁠) Desiccant enthalpy (⁠|$kJ/kg$|⁠) 21.6%(⁠|$\uparrow $|⁠) 14.08%(⁠|$\uparrow $|⁠) 29.01%(⁠|$\uparrow $|⁠) 9.3%(⁠|$\uparrow $|⁠) 8.49%(⁠|$\downarrow $|⁠) 10.09%(⁠|$\downarrow $|⁠) Desiccant concentration (⁠|$kg_d/kg_s$|⁠) 4%(⁠|$\downarrow $|⁠) 1.3% (⁠|$\downarrow $|⁠) 5.8%(⁠|$\downarrow $|⁠) 9%(⁠|$\downarrow $|⁠) 12.8%(⁠|$\downarrow $|⁠) 4.9%(⁠|$\downarrow $|⁠) Air humidity ratio (⁠|$kg_w/kg_a$|⁠) 8.72%(⁠|$\downarrow $|⁠) 5.76%(⁠|$\uparrow $|⁠) 38.2%(⁠|$\uparrow $|⁠) 8%(⁠|$\downarrow $|⁠) 64.5%(⁠|$\downarrow $|⁠) 44.3%(⁠|$\downarrow $|⁠) Parameter . Dehumidifier . Regenerator . |$\omega _i$| . Inlet to outlet % (⁠|$\uparrow /\downarrow $|⁠) . |$\omega _i$| . Inlet to outlet% (⁠|$\uparrow /\downarrow $|⁠) . |$\omega _{i_{(min)}}(\uparrow )$| . |$\omega _{i_{(max)}}(\uparrow )$| . |$\omega _{i_{(min)}}(\uparrow )$| . |$\omega _{i_{(max)}}(\uparrow )$| . Air enthalpy (⁠|$kJ/kg$|⁠) 16.1%(⁠|$\downarrow $|⁠) 4.11%(⁠|$\uparrow $|⁠) 27.6%(⁠|$\uparrow $|⁠) 37.3%(⁠|$\uparrow $|⁠) 27.21%(⁠|$\uparrow $|⁠) 46.14%(⁠|$\uparrow $|⁠) Desiccant enthalpy (⁠|$kJ/kg$|⁠) 21.6%(⁠|$\uparrow $|⁠) 14.08%(⁠|$\uparrow $|⁠) 29.01%(⁠|$\uparrow $|⁠) 9.3%(⁠|$\uparrow $|⁠) 8.49%(⁠|$\downarrow $|⁠) 10.09%(⁠|$\downarrow $|⁠) Desiccant concentration (⁠|$kg_d/kg_s$|⁠) 4%(⁠|$\downarrow $|⁠) 1.3% (⁠|$\downarrow $|⁠) 5.8%(⁠|$\downarrow $|⁠) 9%(⁠|$\downarrow $|⁠) 12.8%(⁠|$\downarrow $|⁠) 4.9%(⁠|$\downarrow $|⁠) Air humidity ratio (⁠|$kg_w/kg_a$|⁠) 8.72%(⁠|$\downarrow $|⁠) 5.76%(⁠|$\uparrow $|⁠) 38.2%(⁠|$\uparrow $|⁠) 8%(⁠|$\downarrow $|⁠) 64.5%(⁠|$\downarrow $|⁠) 44.3%(⁠|$\downarrow $|⁠) Open in new tab Table 6 Effects of air HR on air and desiccant enthalpy. Parameter . Dehumidifier . Regenerator . |$\omega _i$| . Inlet to outlet % (⁠|$\uparrow /\downarrow $|⁠) . |$\omega _i$| . Inlet to outlet% (⁠|$\uparrow /\downarrow $|⁠) . |$\omega _{i_{(min)}}(\uparrow )$| . |$\omega _{i_{(max)}}(\uparrow )$| . |$\omega _{i_{(min)}}(\uparrow )$| . |$\omega _{i_{(max)}}(\uparrow )$| . Air enthalpy (⁠|$kJ/kg$|⁠) 16.1%(⁠|$\downarrow $|⁠) 4.11%(⁠|$\uparrow $|⁠) 27.6%(⁠|$\uparrow $|⁠) 37.3%(⁠|$\uparrow $|⁠) 27.21%(⁠|$\uparrow $|⁠) 46.14%(⁠|$\uparrow $|⁠) Desiccant enthalpy (⁠|$kJ/kg$|⁠) 21.6%(⁠|$\uparrow $|⁠) 14.08%(⁠|$\uparrow $|⁠) 29.01%(⁠|$\uparrow $|⁠) 9.3%(⁠|$\uparrow $|⁠) 8.49%(⁠|$\downarrow $|⁠) 10.09%(⁠|$\downarrow $|⁠) Desiccant concentration (⁠|$kg_d/kg_s$|⁠) 4%(⁠|$\downarrow $|⁠) 1.3% (⁠|$\downarrow $|⁠) 5.8%(⁠|$\downarrow $|⁠) 9%(⁠|$\downarrow $|⁠) 12.8%(⁠|$\downarrow $|⁠) 4.9%(⁠|$\downarrow $|⁠) Air humidity ratio (⁠|$kg_w/kg_a$|⁠) 8.72%(⁠|$\downarrow $|⁠) 5.76%(⁠|$\uparrow $|⁠) 38.2%(⁠|$\uparrow $|⁠) 8%(⁠|$\downarrow $|⁠) 64.5%(⁠|$\downarrow $|⁠) 44.3%(⁠|$\downarrow $|⁠) Parameter . Dehumidifier . Regenerator . |$\omega _i$| . Inlet to outlet % (⁠|$\uparrow /\downarrow $|⁠) . |$\omega _i$| . Inlet to outlet% (⁠|$\uparrow /\downarrow $|⁠) . |$\omega _{i_{(min)}}(\uparrow )$| . |$\omega _{i_{(max)}}(\uparrow )$| . |$\omega _{i_{(min)}}(\uparrow )$| . |$\omega _{i_{(max)}}(\uparrow )$| . Air enthalpy (⁠|$kJ/kg$|⁠) 16.1%(⁠|$\downarrow $|⁠) 4.11%(⁠|$\uparrow $|⁠) 27.6%(⁠|$\uparrow $|⁠) 37.3%(⁠|$\uparrow $|⁠) 27.21%(⁠|$\uparrow $|⁠) 46.14%(⁠|$\uparrow $|⁠) Desiccant enthalpy (⁠|$kJ/kg$|⁠) 21.6%(⁠|$\uparrow $|⁠) 14.08%(⁠|$\uparrow $|⁠) 29.01%(⁠|$\uparrow $|⁠) 9.3%(⁠|$\uparrow $|⁠) 8.49%(⁠|$\downarrow $|⁠) 10.09%(⁠|$\downarrow $|⁠) Desiccant concentration (⁠|$kg_d/kg_s$|⁠) 4%(⁠|$\downarrow $|⁠) 1.3% (⁠|$\downarrow $|⁠) 5.8%(⁠|$\downarrow $|⁠) 9%(⁠|$\downarrow $|⁠) 12.8%(⁠|$\downarrow $|⁠) 4.9%(⁠|$\downarrow $|⁠) Air humidity ratio (⁠|$kg_w/kg_a$|⁠) 8.72%(⁠|$\downarrow $|⁠) 5.76%(⁠|$\uparrow $|⁠) 38.2%(⁠|$\uparrow $|⁠) 8%(⁠|$\downarrow $|⁠) 64.5%(⁠|$\downarrow $|⁠) 44.3%(⁠|$\downarrow $|⁠) Open in new tab Figure 5 Open in new tabDownload slide The effect of (a) inlet air HR and (b) L/G ratio on the effectiveness |$\varepsilon $| and MRR, |$\epsilon $| of the dehumidifier and regenerator. Figure 5 Open in new tabDownload slide The effect of (a) inlet air HR and (b) L/G ratio on the effectiveness |$\varepsilon $| and MRR, |$\epsilon $| of the dehumidifier and regenerator. The effects of inlet air HR and L/G ratio on the effectiveness |$\varepsilon $| and MRR, |$\epsilon $| of the dehumidifier and regenerator vessels are indicated in Figure 5. The influence of inlet air HR of the dehumidifier and regenerator on the effectiveness and MRR was assessed and plotted as shown in Figure 5a. A significant reduction in regenerator effectiveness from 60% to 12% was experienced as the inlet air HR increased from 0.0158 kg|$_w$|/kg|$_a$| to 0.0367 kg|$_w$|/kg|$_a$|⁠. The reduction was due to the fact that, as the air HR increased, the moisture-holding capacity decreased and hence less water vapour could be expelled from the desiccant, and therefore, the mass conveyance possibility is significantly reduced. On the other hand, the increase in inlet air HR resulted in increased dehumidifier effectiveness. Within the same HR range, the dehumidifier effectiveness increased from 22% to 36%. The gradual increase was due to increased interfacial mass transfer potential of air and LiBr fed into the dehumidifier. At desiccant-air flow-rate mix of 0.5 kg/h and 0.8 kg/s respectively, the water vapour absorption was limited to the desorption capacity of the desiccant. At these conditions, the thermal-mass exchange was optimized. Considering the same range of HR values from 0.0158 kg|$_w$|/kg|$_a$|–0.0367 kg|$_w$|/kg|$_a$|⁠, it was noticed that the regenerator MRR decreased considerably from 1.9 kg/s to 0.7 kg/s. The reduction was due to improved vapour pressure, thus, inhibiting mass transfer at the air–desiccant interface. With higher inlet air HR, there is less capacity to suspend more water vapour from the regenerated desiccant. There was an increased dehumidifier MRR with the increase in air HR from 0.3 kg/s to 2.2 kg/s within the same HR range. This increase was occasioned by the high affinity for the concentrated desiccant to absorb more water vapour. The lower vapour pressure in the desiccant necessitated the mass transfer process; hence, more water vapour was drawn from the process air. The air flow-rate possess a significant consequence on the operation of the liquid desiccant dehumidification and regeneration schemes as opposed to the desiccant flow rates [28]. Therefore, the L/G ratio was altered by adjusting the airflow rate, while the desiccant flow rate remained unchanged. The effectiveness and MRR of the dehumidifier and regenerator were drawn alongside the L/G ratio, as shown in Figure 5b. The rise in L/G ratio from 0.06 to 0.24 resulted in a sharp increase in dehumidifier MRR for 0.4 kg/s to 2.15 kg/s while the regenerator MRR reduces from 2.0 kg/s to 0.8 kg/s as presented in the figure. These changes were experienced since by reducing the air mass flow rate; the L/G ratio increased; as a result, the interfacial mass transfer potential was diminished, hence, the reduction in regenerator MRR. In contrast to the MRR, the dehumidifier effectiveness reduced from 50% to 10% while that of the regenerator increased from 29% to 41% within the same L/G ratio range of 0.06–0.24. The reduction of air mass flow meant that there was a prolonged resident time in the vessel that resulted in decelerated interfacial interaction. As a result, the dehumidifier exhibited the characteristics of a regenerator while the regenerator performance was adequately satisfactory. Since raised LiBr solution temperatures initiate the regeneration process, it was necessary to include the VPV/T collector to aid in solar radiation conversion and conduction of heat into the circulation solution. The LiBr temperature at the exit of the VPV/T was therefore monitored and recorded for sunshine hours. The variation of outlet desiccant temperature from the collector at different flow rates is presented in Figure 6. The flow-rate was changed from 0.5–15 kg/h. It was observed that, the greater the flow-rates, the lesser the temperature increase and vice versa. At 0.5 kg/h, the temperature varied from an inlet state of 20|$^o$|C to a maximum of 68.14|$^o$|C at 12:30 hours. Alternatively, the highest flow rate considered was 15 kg/h, which resulted in a temperature rise from 20|$^o$|C to a maximum of 35.14|$^o$|C. Figure 6 Open in new tabDownload slide The VPV/T collector outlet temperature variation at different flow rates with time of the day. Figure 6 Open in new tabDownload slide The VPV/T collector outlet temperature variation at different flow rates with time of the day. The implication was that at lower flow-rates, the desiccant solution gathered more heat from the collector due to more retention time. However, the risk of high stagnation temperature of the VPV/T collector was avoided by adjusting the flow rates accordingly. Therefore, the flow-rate of 0.5 kg/h through the collector was selected and maintained throughout the experiment. The effect of solar radiation on the MRR and effectiveness of the regenerator and dehumidifier was studied under the following conditions: |$\chi $| = 60 kg|$_w$|/kg|$_a$|⁠, |$\dot{m}_a$| = 0.8 kg/ms and T|$_{sol,in}$| = 21 |$^o$|C, plotted as presented in Figure 7. The regenerator MRR and effectiveness were both observed to improve with the rise in solar radiation, implying that the mass and heat transmission potential was enriched. Figure 7 Open in new tabDownload slide The effect of solar radiation on MRR and effectiveness of the dehumidifier and regenerator. Figure 7 Open in new tabDownload slide The effect of solar radiation on MRR and effectiveness of the dehumidifier and regenerator. The higher the temperature of the desiccant solution, the less the capability to hold water particles hence evaporation ensued, resulting in HR difference and desiccant re-concentration to near initial condition. However, both the dehumidifier MRR and effectiveness dipped with increased desiccant temperature as a consequence of higher solar radiation. Since the dehumidifier experiences high mass transfer rates, as opposed to the high heat transfer rates in the regenerator, the temperature rise due to solar radiation worked negatively for the dehumidifier. The desiccant effectively absorbs water vapour at low temperatures, but as temperature increases, more water vapour is likely to be released from the desiccant to the air, which is not the dehumidification process objective. In order to achieve optimal performance levels of the vessels in terms of effectiveness and MRR, the air and desiccant flow rates were varied during the experiment, and the corresponding measurements were taken. Subsequently, the results were plotted, as shown in Figure 8, from which the consequence of air and desiccant mass flow rates on the effectiveness and MRR of both the dehumidifier and regenerator was studied. The air mass flow rate was changed progressively from 0.5 kg/ms to 1.1 kg/ms. Within this range, it was observed that the regenerator effectiveness increased steadily from 24% to 39% while that of the dehumidifier reduced from 56% to 13% as depicted in Figure 8a. The increase in MRR was because of the improve vapour pressure dispersion from the weak LiBr at elevated temperature, which exhibited weak moisture-holding capacity and hence increased mass transfer potential. On the other hand, the decline realized on the dehumidifier was primarily due to reduced resident time and minimized interfacial interaction of air-desiccant in the vessel hence lowering the mass transfer capability. Figure 8 Open in new tabDownload slide The influence of (a) airflow rate and (b) desiccant flow rate on the MRR and effectiveness of the dehumidifier and regenerator. Figure 8 Open in new tabDownload slide The influence of (a) airflow rate and (b) desiccant flow rate on the MRR and effectiveness of the dehumidifier and regenerator. Similarly, the MRR showed a slight upward trend for the regenerator from 0.3 kg/s to 0.4 kg/s. There was a slight adjustment compared to that of air due to the hastened interfacial air–desiccant interaction, which hindered the mass transmission rate and resulted in low MRR. On the contrary, the dehumidifier MRR decreased significantly with the upsurge in air mass flow–rate. The MRR changed from 2.5 kg/s to 0.75 kg/s within the considered range of airflow rates. Again the mass transfer capability was lowered, which negatively affected MRR. From Figure 8b, the influence of LiBr flow rate on MRR and effectiveness is analysed for both the dehumidifier and regenerator. The MRR curves follow the same trend as those of air; however, the only difference is the rate of change. It was observed that the regenerator MRR reduced significantly from 1.9 kg/s to 0.8 kg/s within the desiccant flow rate range of 0.04 kg/ms to 0.12 kg/ms. On the other hand, the dehumidifier MRR increased from 0.3 kg/s to 0.4 kg/s as in the previous case depicting that MRR generally shows low sensitivity to the air and desiccant flow rates. Increased desiccant flow rate of similar range reduced the dehumidifier effectiveness from 40% to 8%, a difference of 8% while the effectiveness increased by 15% from 22% to 37%. The influence of incoming LiBr concentration on MRR and effectiveness of both the regenerator and dehumidifier is displayed in Figure 9. The LiBr concentration significantly affects the dehumidifier MRR and effectiveness, as shown in Figure 9a. As the solution concentration increases from 93% to 98%, the MRR considerably decreased from 5.8 kg/s to 1.8 kg/s. This upward change was attributed to the upright geometry of the packing profile, which enabled the solution to flow through with limited air contact, which inhibited mass transfer possibility. The effectiveness profile curve exhibited an increase from 12.5% to 23% with an increase of concentration up to 95 kg|$_w$|/kg|$_a$|⁠. Further increase in concentration up to 98 kg|$_w$|/kg|$_a$| resulted in a decrease of effectiveness up to 3% due to the variation in HR as correlated in [22]. Figure 9 Open in new tabDownload slide The influence of inlet LiBr concentration on MRR |$\epsilon $| and effectiveness |$\varepsilon $| of (a) dehumidifier and (b) regenerator. Figure 9 Open in new tabDownload slide The influence of inlet LiBr concentration on MRR |$\epsilon $| and effectiveness |$\varepsilon $| of (a) dehumidifier and (b) regenerator. Similarly, the impact of inlet solution concentration on the regenerator effectiveness and MRR is presented in Figure 9b. It was observed that the MRR decreased from 1.7 kg/s to 1.1 kg/s with an increase in concentration from 64 kg|$_w$|/kg|$_a$| to 80 kg|$_w$|/kg|$_a$|⁠. In contrast, the effectiveness increased from 18% to 23% within the same range of concentration. This increase was due to the low moisture-holding capacity of desiccant solution at high temperature caused by the hybrid VPV/T as well as the cooling effect of air. The surface texture and geometrical configuration of the packing in terms of smoothness and straight vertical profiles meant that less solution was in contact with the surface, resulting in low contact time due to hastened gravitational flow. For higher MRR, it would require a corresponding increase in air mass flow in line with the findings of [30]. 4 The theoretical model The internal structure of the dehumidifier/regenerator vessel is configured to a counter-flow mode where the desiccant flows downwards from the top countering the air stream flowing upwards from the bottom of the vessel. The vessels are packed with square channelled ceramic cordierite. Although the LiBr liquid is sprayed on the upper side of the packing, the fluid settles and flows continuously as a thin film along the square packing walls. This continuous flow is lamina and is taken to be within the range of |$0 < Re_d < 4$|⁠. For descending fluid stream within this |$Re$| range, the prediction of fluid stream thickness and velocity characteristics is best estimated using Nusselt number analysis [31]. The ensuing expression for the stream breadth is given as: $$\begin{equation} \varrho = \sqrt[3]{\frac{3\dot{m}_d \mu_d}{g{\mu_d}^2}}, \end{equation}$$(17) where |$\dot{m}_d$| is the desiccant flow rate per unit length of the packing in kg/ms, |$\mu _d$| is the liquid desiccant viscosity and g is the gravitational acceleration. Taking a small elemental particle of interfacial area shown in Figure 10 in which the heat and mass transmission occurs in x, y and z planes, the following assumptions are considered in formulating the fundamental equations: (i) the dehumidification/regeneration process is adiabatic, (ii) static heat and mass transfer constants throughout the process, (iii) no heat loss caused by latent heat of condensation and evaporation and (iv) the fluid flow is continuous and laminar. Figure 10 Open in new tabDownload slide The flow schematic of an elemental particle of interfacial area. Figure 10 Open in new tabDownload slide The flow schematic of an elemental particle of interfacial area. Since the water vapour absorption/desorption rate is far much lower than the LiBr solution flow rates, keeping the desiccant flow rate unchanged yields a relatively stream thickness and velocity of similar invariable nature. The invariability implies that there is no interfacial velocity gradient, and therefore the velocity contour is wholly established at the beginning of the interaction span. Due to the low absorption/desorption rates, the mass flow rate towards the |$y$|-direction and the accompanying partial velocity |$v$| are significantly dominant. Therefore, the underlying expressions for the interfacial thermal and mass exchange can thus be developed [31]. Let the subscripts |$a$| and |$d$| represent air desiccant solution respectively, the fluid momentum in the |$x$|-plane is given as $$\begin{equation} \mu_d \frac{\partial^2u}{\partial{y_d}^2} + \rho_dg = 0. \end{equation}$$(18) The energy balance expression is $$\begin{equation} v_d \frac{\partial T_d}{\partial x} = \gamma_a\frac{\partial^2T_d}{\partial{y_d}^2}. \end{equation}$$(19) And mass diffusion equation becomes: $$\begin{equation} v_d \frac{\partial \chi_d}{\partial x} = \delta_a\frac{\partial^2\chi_d}{\partial{y_d}^2}. \end{equation}$$(20) The air flows through a vertical rectangular channel of the packing whose length is considered to be extremely large compared to the width. Additionally, the depth of the channel varies form minimum at inlet to maximum at outlet, hence, the fluid flow contour is fully developed and laminar. Then the |$z$| momentum equation is given as: $$\begin{equation} \mu_a \frac{\partial^2v_a}{\partial{y_d}^2} - \frac{\partial p_w}{\partial z} = 0. \end{equation}$$(21) The energy balance expression is $$\begin{equation} v_a \frac{\partial T_a}{\partial z} = \gamma_a\frac{\partial^2T_a}{\partial{y_d}^2}. \end{equation}$$(22) Mass diffusion equation becomes: $$\begin{equation} v_a \frac{\partial \chi_a}{\partial z} = \delta_a\frac{\partial^2\chi_a}{\partial{y_d}^2}. \end{equation}$$(23) The equilibrium mass flow expression is formulated as $$\begin{equation} \varphi_aC_{p_w}\frac{\partial T_a}{\partial x} - \gamma a_dA_p\frac{\partial^2 T_a}{\partial y^2} = 0 \end{equation}$$(24) $$\begin{equation} \varphi_a\frac{\partial \omega}{\partial x} - \beta a_dA_p\frac{\partial^2 \omega}{\partial y^2} = 0, \end{equation}$$(25) where |$\varphi $| is the mass flux in kg/ms, |$\gamma $| and |$\beta $| are the heat and mass exchange constants, respectively, |$C_{p_w}$| is the specific heat capacity of water vapour at stagnated pressure, |$\delta $| is the diffusion coefficient, |$A_p$| is the void space area of the packing and |$P_w$| is the vapour pressure. Considering the energy conservation on the liquid side $$\begin{equation} \varphi_dC_{p_d}\frac{\partial T_d}{\partial x} - \gamma a_dA_p\frac{\partial^2 T_d}{\partial y^2} = 0 \end{equation}$$(26) $$\begin{equation} \varphi_d\frac{\partial \chi}{\partial x} - \beta a_dA_p\frac{\partial^2 \omega}{\partial y^2} = 0. \end{equation}$$(27) The interfacial characteristic equations can therefore be developed as follows: $$\begin{equation} \varphi_a\biggl(C_{p_w}\frac{\partial T_a}{\partial y} - \xi \frac{\partial \omega}{\partial y}\biggr) + \gamma a_d(T_a - T_d) = 0 \end{equation}$$(28) $$\begin{equation} \varphi_d\biggl(C_{p_d}\frac{\partial T_d}{\partial y} - \xi \frac{\partial \omega}{\partial y}\biggr) + \gamma a_d(T_d - T_a) = 0 \end{equation}$$(29) $$\begin{equation} \varphi_d\frac{\partial \chi}{\partial y} + \varphi_a \frac{\partial \omega}{\partial y} = 0 \end{equation}$$(30) $$\begin{equation} \varphi_a\frac{\partial \omega}{\partial y} + \beta a_d(\omega-\omega_e) = 0, \end{equation}$$(31) where |$\xi $| is the specific heat of dilution, the length and width-wise desiccant solution temperature variations are represented by |$\frac{\partial T_d}{\partial x}$| and |$\frac{\partial ^2 T_d}{\partial y^2}$|⁠, respectively, while lengthwise and height-wise variation of solution concentration within the vessel are given by |$\frac{\partial \chi }{\partial x}$| and |$\frac{\partial ^2 \chi }{\partial y^2}$|⁠. The latent heat produced during air dehumidification/regeneration processes and water vapour absorption/desorption is represented by |$\xi \frac{\partial \omega }{\partial y}$| and |$\xi \frac{\partial \chi }{\partial y}$|⁠, respectively. The generated interfacial heat |$\gamma a_d(T_a - T_d)$| and mass transfer |$\beta a_d(\omega -\omega _e)$| driven by vapour pressure difference are also considered. Since dehumidification and regeneration processes are exothermic and endothermic respectively, the interfacial absorption/desorption heat can thus be found as a function of enthalpy of condensation/evaporation and dilution as follows: $$\begin{equation} \xi = h_{fg}(T_a, \omega) + \Delta h(T_a, P, \beta). \end{equation}$$(32) Due to the change in air and desiccant temperatures, the change in specific enthalpy is inevitable. Hence, the variation in specific enthalpy of the humid air can be found by $$\begin{equation} \partial h_a =Cp_a\partial T_a + \partial\omega(Cp_w(T_a - T_{amb}) + \lambda), \end{equation}$$(33) where |$\lambda $| is the latent heat of condensation in kJ/kg. Therefore, the conditions at the boundary layer of x-y planes in the direction of z can thus be expressed as $$\begin{align} x = 0; T_d = T_{d,i}; \chi = \chi_i \end{align}$$(34) $$\begin{align} x = H; T_a = T_{da,i}; \omega = \omega_i \end{align}$$(35) $$\begin{align} y = 0; T_d = T_a; \frac{\partial\chi}{\partial y}=0; u = 0 \end{align}$$(36) $$\begin{align} y = \frac{y}{2}=\dfrac{\omega}{2n}; T_d = T_a; \omega_a = \omega_e ; \frac{\partial T_a}{\partial y}=\frac{\partial\chi}{\partial y}=\frac{\partial v_a}{\partial y}=0 \end{align}$$(37) $$\begin{align} y = y_p=\dfrac{w}{n}; \frac{\partial T_a}{\partial y} = 0; \frac{\partial \omega_a}{\partial y} = 0 \end{align}$$(38) $$\begin{align} y = \varrho; \frac{\partial u}{\partial y}=0 \end{align}$$(39) $$\begin{align} y = h; v_a = 0. \end{align}$$(40) Considering the above defined boundary modalities, the corresponding interfacial scenarios specified at |$y$| = |$\varrho $|⁠, implies that there is no temperature gradient, hence |$T_d$| = |$T_a$|⁠. However, for thermal balance of a perfect air-vapour blend, the interfacial air and solution vapour pressures are equal, thus, the interfacial concentration can be formulated in terms of molecular weight of water |$M_w$| and air |$M_a$| and vapour pressure |$P_w$| as follows: $$\begin{equation} \chi_i = \frac{M_wP_w}{M_a(P_{mix} - P_w) + M_wP_w}. \end{equation}$$(41) The vapour pressure of water in LiBr solution can be derived in terms of temperature and concentration from the thermal properties as $$\begin{equation} P_w= f(T_{di},\chi_{wi}). \end{equation}$$(42) The interfacial mass balance is expressed as $$\begin{equation} \rho_d\delta_d\frac{\partial \chi_d}{\partial y} = -\rho_a \frac{\delta_a\delta \chi_w}{\partial y}. \end{equation}$$(43) Similarly, the interfacial energy balance becomes $$\begin{equation} -k_d\frac{\partial T_d}{\partial y_d} = k_a\frac{\partial T_a}{\partial y_a}+\rho_a \delta_a\frac{\delta \chi_w}{\partial y}h_{fg}. \end{equation}$$(44) The component |$h_{fg}$| is the latent heat of vaporization and |$\delta $| is the diffusion coefficient. The value of the respective coefficients often determines the degree of heat and mass exchange. These constants can thus be established using the interfacial expression obtained from [29]. $$\begin{equation} \gamma = \frac{\varphi_aCp_{wv}(T_{a,o} - T_{a,i})}{\frac{1}{2}\lambda a_d\biggl((T_{a,o} - T_{a,i})-(T_{d,o} - T_{d,i})\biggr)}, \end{equation}$$(45) where |$\lambda $| is the volume per unit surface area of one segment of the packing material. But the packed vessel thermal and overall efficiencies are given as a function of the air and desiccant temperature difference ratios [28]. $$\begin{equation} \eta_t = \frac{(T_{a,o} - T_{a,i})}{(T_{d,i} - T_{a,i})} = \frac{(T_{a,i} - T_{a,o})}{(T_{a,i} - T_{d,i})} \end{equation}$$(46) and $$\begin{equation} \eta_o = \frac{(T_{d,o} - T_{d,i})}{(T_{i} - T_{d,i})} = \frac{(T_{a,o} - T_{a,i})}{(T_{i} - T_{a,i})} \end{equation}$$(47) Equations 46 and 47 can then be substituted in Equation 45 to yield $$\begin{align} \gamma = \frac{2\varphi_aCp_{wv}(T_{d,i} - T_{a,i})\eta_t}{\lambda a_d\biggl((T_{d,i} - T_{a,i})\eta_t -(T_{i} - T_{d,i})\eta_o\biggr)}. \end{align}$$(48) Similarly, the interfacial mass transfer coefficient is found by $$\begin{equation} \beta = \frac{\varphi_a}{a_d\lambda}\ln \biggl(\frac{1}{1-\varepsilon}\biggr). \end{equation}$$(49) In general, since the heat and mass exchange phenomenon occurs simultaneously and therefore combined, their coefficients can be correlated using Lewis number (Le), and number of thermal units (NTU) determined by $$\begin{equation} Le = \frac{\gamma}{\beta Cp_w} \end{equation}$$(50) $$\begin{equation} NTU = \frac{\beta A_pV}{\varphi_a}. \end{equation}$$(51) The Equations 50 and 51 are correlated as function of the corresponding changes in enthalpy in the z direction and HR as follows: $$\begin{equation} \frac{\partial h_a}{\partial z} = \frac{NTU.Le} {h_a}\biggl[(h_e - h_a) + \lambda\biggl(\frac{1}{Le} - 1\biggr)(\omega_e - \omega_a)\biggr]. \end{equation}$$(52) From the numerical analysis, it can be established that the interfacial energy exchange between air and LiBr liquid desiccant stream is majorly subjected to the gas side temperature variation giving rise to useful heat and mass transfer rates and consequently causing potential heat gain. The coupling nature of heat and mass transfer obscures the problem and makes the solution a challenging task. Equations 44 and 52 can be conveniently solved by a combination of step by step iterative and separative evaluation methods to obtain the heat and mass transfer characteristic results that can then be compared to the experimental data. The model solution procedure involves the application of algebraic conversions of fundamental equations to group the segments at the inlet section to account for enormous variations in temperature and concentration. 4.1 The model solution procedure The general outline of the solution procedure is as follows: START Input initial guess values for |$\varphi $|⁠, |$a$|⁠, |$V$|⁠, |$A$|⁠, |$\xi $|⁠, |$T_{a,i}$|⁠, |$T_{d,i}$|⁠, |$Cp$|⁠, T, |$\eta _T$|⁠, |$\delta $|⁠, |$\omega _e$|⁠, |$\omega _i$| and |$\eta $| for the air and LiBr. Initialize the values of input parameters at inlet and borderline settings and guess the interfacial temperature |$T_i$|⁠. Input the initial guess for the interfacial LiBr concentration and use thermodynamic equilibrium properties to obtain the concentration of vapour in air. Compute the wall temperature using the heat resistance analogy. Compute the liquid stream thickness |$\varrho $|⁠. Evaluate the ensuing energy balance matrix emanating from equation. Compute the interfacial diffusive characteristics using Equations 20 and 23. Using separative evaluation technique, compute |$\omega $|⁠. Solve Equations 28, 29 and 30 to obtain |$T_a$|⁠, |$T_d$| and |$\chi $|⁠. Test if the interfacial energy balance Equation 43, is valid, if NO, recompute the interfacial temperature using Equation 44 and revert to steps 3–10. If YES, continue to the subsequent step. Compute |$\gamma $| and |$\beta $| using Equations 48 and 49. Solve Equations 50 and 51 and 52 to obtain |$Le$| and |$NTU$|⁠. STOP. Figure 11 Open in new tabDownload slide The effect of (a) air and (b) LiBr flow rates on the heat transfer coefficient. Figure 11 Open in new tabDownload slide The effect of (a) air and (b) LiBr flow rates on the heat transfer coefficient. Figure 12 Open in new tabDownload slide The effect of air and LiBr flow rates on the mass transfer coefficient. Figure 12 Open in new tabDownload slide The effect of air and LiBr flow rates on the mass transfer coefficient. 4.2 The numerical model results Figure 11 shows the influence of air and LiBr flow rates on the heat transfer coefficients. During dehumidification and regeneration processes, an increase in airflow rate resulted in increased heat transfer coefficient but with varying magnitudes. For the given inlet conditions, when airflow rate per unit length of the vessel was increased from 1.48 kg/ms to 4.9 kg/ms, the heat transfer coefficient increased from 39.7 W/m|$^2$|K to 43.22 W/m|$^2$|K during dehumidification, whereas, in the regenerator, increased mass flow rate from 0.7 kg/ms to 3.0 kg/ms gave a corresponding increase in heat transfer from 54.80 W/m|$^2$|K to 69.44 W/m|$^2$|K as shown in Figure 11a. The impact of the LiBr flow rate on the heat transfer coefficient is presented in Figure 11b. For the same range of solution flow rates from 0.039 kg/m|$^2$|s to 0.012 kg/m|$^2$|s, it was observed that the heat transfer coefficient increased steadily both in the dehumidifier and regenerator vessels. The magnitudes of the corresponding increases were from 39.16 W/m|$^2$|K to 43.45 W/m|$^2$|K and 55.05 W/m|$^2$|K to 69.41 W/m|$^2$|K, respectively. Therefore, in order to obtain optimum heat transfer coefficient, the vessels should be subjected to gradually increasing air and LiBr flow rates taking into account the risk of carry-over. Figure 13 Open in new tabDownload slide The effect of Lewis number on the heat and mass transfer coefficients. Figure 13 Open in new tabDownload slide The effect of Lewis number on the heat and mass transfer coefficients. Figure 12 shows the influence of air and LiBr flow rates on the mass exchange coefficients of the dehumidifier and regenerator vessels. For both vessels, it was generally observed that the respective increases in air and LiBr flow rates produced varying increases in the mass exchange coefficients. Regarding Figure 12a, when the airflow rate was varied between 1.992 kg/ms to 6.012 kg/ms while keeping desiccant solution flow rate constant, a rapid increase in mass transfer coefficient from 2.29 kg/m|$^2$|s to 8.78 kg/m|$^2$|s was realized during dehumidification process. Similarly, for the regeneration process, the same air mass flow rate range gave a gradual proliferation of the mass transferal coefficient from 0.87 kg/m|$^2$|s to 3.39 kg/m|$^2$|s. In another scenario illustrated in Figure 12b, when the air mass flow rate per unit length was kept constant while varying the desiccant solution flow rate per unit surface area from 0.102 kg/m|$^2$|s to 0.2536 kg/m|$^2$|s, steady increases in mass transfer coefficients were realized during both dehumidification and regeneration processes. Increases of mass transfer coefficients from 0.029 kg/m|$^2$|s to 8.35 kg/m|$^2$|s and 0.71 kg/m|$^2$|s to 3.55 kg/m|$^2$|s were realized for the dehumidifier and regenerator vessels, respectively. In both cases of varying air and desiccant flow rates alternately within the specified ranges, there were significant increases of above 70% in the course of dehumidification and regeneration procedures because of the low vapour pressure and geometry of the CC packing as well as improved interfacial mass transfer potential caused by improved MRR. The inlet parameters outlined in Table 5 were used to compute the Lewis number and plotted against the heat and mass exchange coefficients corresponding to the dehumidifier and regenerator vessels, as displayed in Figure 13. For both the dehumidifier and regenerator vessels, it was observed that as the Lewis number increased, both the heat and mass transfer constants expressively decreased as depicted in Figures 13a and 13b. These decreases were attributed to the fact that the Lewis number increases when the vapour absorption capability weakens due to reduced interfacial mass transfer. The resultant implication of the reduced mass transfer on the latent heat of vaporization and condensation is also significantly negative, and hence the heat transfer coefficient is also reduced significantly. For the dehumidification system, the increase in Lewis number from 1.07 to 4.24 caused a decline in heat and mass transfer coefficients from 43.72 W/m|$^2$|K to 39.25 W/m|$^2$|K and 9.74 kg/m|$^2$|s to 2.21 kg/m|$^2$|s, respectively. Similarly, for the regenerator, as Lewis number increased from 3.83 to 13.54, the heat and mass transfer coefficients reduced from 70.22 W/m|$^2$|K to 57.23 W/m|$^2$|K and 4.38 kg/m|$^2$|s to 1.01 kg/m|$^2$|s, respectively. These findings were in agreement with those of [28]. 5 Validation of the model To ascertain the validity of numerical model outcomes, the measured data, as well as experimental outcomes, were used as a benchmark. Comparisons were conducted at various levels of inputs and outputs of different dehumidifier and regenerator thermal and mass exchange performance characteristics and indices. Since the leading performance indices of the dehumidifier and regenerator units were the MRR and effectiveness, a side-by-side comparative evaluation was done concerning projected and experimental figures for the considered range of experimental conditions at different air and desiccant solution inlet temperatures along with inlet air HR. Figure 14 illustrates the degree of validation match between the predicted and experimental dehumidifier and regenerator vessels effectiveness. The variation of dehumidifier effectiveness was within |$\pm $|6.2%, as shown in Figure 14a, while that of the regenerator was |$\pm $|2.9%, as shown in Figure 14b. The lower variation experienced in the regenerator was due to the influence of solar radiation that had a powerful effect on the regeneration process. The temperature increase was significant in improving the regenerator effectiveness. On the contrary, the temperature increase worked against the definite increase in dehumidifier performance. Figure 14 Open in new tabDownload slide Experimental validation of predicted effectiveness |$\varepsilon $|⁠. Figure 14 Open in new tabDownload slide Experimental validation of predicted effectiveness |$\varepsilon $|⁠. Figure 15 Open in new tabDownload slide Experimental validation of predicted MRR |$\epsilon $|⁠. Figure 15 Open in new tabDownload slide Experimental validation of predicted MRR |$\epsilon $|⁠. Comparisons of the experimental and predicted MRR for the dehumidifier and regenerator is shown in Figure 15. Specifically, Figure 15a and b show the validation of predicted results with experimental data at near precise conformity within |$\pm $| 2% and |$\pm $|1.2% for dehumidifier and regenerator MRR, respectively. Again, the MRR of the desiccant solution was found to depend on the temperature variation provided by the solar radiation as the fluid exhibited weak moisture-holding capacity at a higher temperature than lower temperature. For the dehumidifier, the results were accurate at a lower temperature while the variation widened at a higher temperature. This occurrence of the LiBr temperature at the dehumidifier entry needed to be as low as possible for effective MRR. A comparison between the experimental and modelled heat transfer coefficients of the dehumidifier and regenerator is presented in Figure 16. The observed average deviations between the investigational and modelled thermal transfer coefficients were |$\pm $|9.7% and |$\pm $|2.8% for the dehumidifier and regenerator, respectively, as shown in Figure 16a and b. Similarly, from the comparisons between the investigational and modelled mass transfer constants in Figure 17, the average deviations were |$\pm $|3.5% and |$\pm $|8.2% during dehumidification and regeneration processes as displayed in Figure 17a and b, respectively. Figure 16 Open in new tabDownload slide Experimental validation of predicted heat transfer coefficient |$\gamma $|⁠. Figure 16 Open in new tabDownload slide Experimental validation of predicted heat transfer coefficient |$\gamma $|⁠. Figure 17 Open in new tabDownload slide Experimental validation of predicted mass transfer coefficient |$\beta $|⁠. Figure 17 Open in new tabDownload slide Experimental validation of predicted mass transfer coefficient |$\beta $|⁠. 6 Conclusions The heat and mass exchange performance of an adiabatic solar-powered liquid desiccant dehumidification and regeneration scheme using LiBr solution has been conducted experimentally as well as numerically. Based on the analysis procedures and computational approaches, the following significant findings have featured prominently. Based on the considered sub-tropical climatic conditions, the vacuum insulated solar photovoltaic and thermal module has shown great potential to provide desiccant regeneration heat and electrical energy to drive the system components. Additionally, the square channel structured CC has shown great potential for application as a packing material and catalyst for heat and mass transmission in the dehumidifier and regenerator vessels. For the given inlet conditions, increased inlet air humidity caused increases in inlet air enthalpies and reduction in outlet air enthalpies during both regeneration and dehumidification process. On the contrary, the desiccant solution enthalpies reduced at inlets and increased at outlets of both dehumidifier and regenerator vessels while the incoming air humidity increased. The increase in inlet air HR also significantly reduced regenerator effectiveness and MRR, while causing increased dehumidifier effectiveness and MRR. Alternatively, the increase in L/G ratio caused an increase in dehumidifier MRR and decrease in regenerator MRR. In contrast, the dehumidifier effectiveness is reduced while that of the regenerator is improved. Varying the air mass flow rate progressively upwards, improved the regenerator effectiveness by 15% while, that of the dehumidifier reduced by 43%. The MRR showed a slight upward trend for the regenerator of 0.1 kg/s and significantly increase of 1.85 kg/s in dehumidifier MRR. The MRR generally shows low sensitivity to the air and LiBr flow rates. For instance, increased LiBr solution flow rate within the rage of 0.04 kg/ms to 0.12 kg/ms caused a significant reduction in regenerator MRR by 0.8 kg/s while the dehumidifier MRR increased by 0.1 kg/s MRR. On the effectiveness, a similar margin of desiccant flow rate reduced the dehumidifier effectiveness by 32% while the generator effectiveness increased by 15%. The desiccant concentration significantly affected the dehumidifier MRR and effectiveness. As the solution concentration increased, the MRR decreased significantly by up to 4 kg/s. The effectiveness improved with increased LiBr concentration. Similarly, the regenerator, MRR decreased with increase in concentration while the effectiveness increased by up to 5% within the same range of concentration. The t3D predictive numerical thermal model based on falling liquid stream with constant thickness in counter-flow configuration was developed and solved by a combination of separative appraisal and stepwise iterative technique. The numerical model showed that during the dehumidification and regeneration processes, an increase in airflow rate per unit length and desiccant solution flow rate per unit area resulted in increased thermal and mass exchange coefficients but with varying proportions. As the Lewis number increased, both the heat and mass transfer constants decreased significantly for both the dehumidifier and regenerator vessels. A 74% increase in Lewis number caused a decrease in heat and mass transfer coefficients by 10% and 77%, respectively. Comparisons conducted at various levels of input and output of the experimental and predicted dehumidifier and regenerator MRR, effectiveness, heat and mass transfer coefficients revealed sublime conformity. The variation of dehumidifier effectiveness was within |$\pm $|6.2% while that of the regenerator was |$\pm $|2.9%. The MRR was within |$\pm $|2% and |$\pm $|1.2% conformity for dehumidifier and regenerator, respectively. The heat transfer coefficients were within |$\pm $|9.7% and |$\pm $|2.8% for the dehumidifier and regenerator, respectively. The average deviations of |$\pm $|3.5% and |$\pm $|8.2% were achieved during dehumidification and regeneration procedures. The underlying findings of this study, provides insights into the design and optimization, application of solar energy as well as CC in liquid desiccant air dehumidification and regeneration setups. Sets of reliable and consistent data upon which theoretical simulation models can be validated and empirical correlations developed have been provided. However, the biochemical reactions and biodegradability of LiBr and CC has not been considered and can be proposed for further investigations. References 1 Yonggao Y , Xiaosong Z, Chen Z. Experimental study on dehumidifier and regenerator of liquid desiccant cooling air conditioning system . Build Environ 2007 ; 42 : 2505 – 11 . Google Scholar Crossref Search ADS WorldCat 2 Kakabayev A , Khandurdyev A. Absorption solar regeneration unit with open regeneration of solution . Appl Sol Energy 1969 ; 5 : 69 – 72 . Google Scholar OpenURL Placeholder Text WorldCat 3 Kakabayev A , Klyschaeva O, Khandurdyev A. Refrigeration capacity of an absorption refrigeration plant with flat plate glazed solution regenerator . Appl Sol Energy 1972 ; 8 : 90 – 5 . Google Scholar OpenURL Placeholder Text WorldCat 4 Kaudinya JV , Kaushik SC. Experimental validation of theoretical studies on open and forced flow solar regeneration . Int J Sol Energy 1986 ; 4 : 13 – 33 . Google Scholar Crossref Search ADS WorldCat 5 Kaushik SC, Kumar Y, Kaundinya JV. Feasibility of an open cycle absorption solar cooling system with solution storage for continuous operation . Int J Ambient Energy 1991 ; 12 : 101 – 6 . Crossref Search ADS WorldCat 6 Hawlader MNA, Stack P , Wood BD. Performance evaluation of glazed and unglazed collector/regenerator in a liquid desiccant absorbent open-cycle absorption cooling system . Int J Sustain Energy 1992 ; 11 : 135 – 64 . Google Scholar OpenURL Placeholder Text WorldCat 7 Davangere BS , Sheriff SA, Goswami DY. Feasibility study of a solar air conditioning system - part 1: psychometric and analysis of air conditioned zone. Int J Energy Res 1999 ; 23 : 103 – 16 . 8 Ji LJ, Wood B. Performance enhancement studies of a solar collector/regenerator for open cycle liquid desiccant regeneration . Proceedings of the Annual ASES Conference , American Solar Energy Society (ASES) , 1993 , Inc (USA), ISSN: 1062 – 4910 . Google Scholar OpenURL Placeholder Text WorldCat 9 Yang R , Wang PL. Experimental study of a forced convection solar collector/regenerator for open cycle absorption cooling. J Sol Energy Eng 1994 ; 116 : 194 - 9 . Crossref Search ADS 10 Yang R , Wang PL. Experimental study of a glazed solar collector/regenerator operated under humid climate. Int J Solar Energy 1995 ; 16 : 185 – 201 . 11 Yang R , Wang PL. Experimental study of a double glazed solar collector/regenerator for open cycle absorption cooling . J Solar Energy Eng 1998 ; 120 : 253 – 9 . Google Scholar Crossref Search ADS WorldCat 12 Yutong Li, Hongxing Y . Experimental study on an open-cycle solar collector regenerator using liquid desiccant for air conditioning . Int J Green Energy 2010 ; 7 : 273 – 88 . Google Scholar Crossref Search ADS WorldCat 13 Alizadeh S , Saman WY . An experimental study of forced flow solar collector/regenerator using liquid desiccant. Solar Energy 2002 ; 73 : 345 – 62 . Google Scholar OpenURL Placeholder Text WorldCat 14 Lazzarin RM, Gasperella A , Longo GA. Chemical dehumidification by liquid desiccants; theory and experiments. J Refrigeration 1999 ; 22 : 334 – 47 . Crossref Search ADS 15 Hennings HM . Solar assisted air conditioning of buildings—an overview . Appl Therm Energy 2007 ; 27 : 1734 – 49 . Google Scholar Crossref Search ADS WorldCat 16 Zhang L , Hihara E, Matsuoka F, Dang C. Experimental analysis of mass transfer in adiabatic structured packing dehumidifier/regenerator with liquid desiccant . Int J Heat Mass Tran 2010 ; 53 : 2856 – 63 . Google Scholar Crossref Search ADS WorldCat 17 Gao WZ , Lui J, Cheng YP, Zhang XL. Experimental investigation on the heat and mass transfer between air and liquid desiccant in a cross-flow dehumidifier . Renew Energy 2012 ; 37 : 117 – 23 . Google Scholar Crossref Search ADS WorldCat 18 Chung TW , Wu H. Comparison between comparison between spray towers with and without fin coils for air dehumidification using triethylene glycol solutions and development of their heat and mass transfer correlations . Ind Eng Chem Res 2000 ; 39 : 2076 – 86 . Google Scholar Crossref Search ADS WorldCat 19 Longo GA , Gasparella A. Experimental and theoretical analysis of heat and mass transfer in a packed column dehumidifier/regenerator with liquid desiccant . Int J Heat Mass Tran 2005 ; 48 : 5240 – 54 . Google Scholar Crossref Search ADS WorldCat 20 Liu XH , Jiang Y, Chang XM, Yi XQ. Experimental investigation of the heat and mass transfer between air and liquid desiccant in a cross-flow regenerator . Renew Energy 2007 ; 32 : 1623 – 36 . Google Scholar Crossref Search ADS WorldCat 21 Longo GA , Gasparella A. Experimental analysis on desiccant regeneration in a packed column with structured and random packing . Solar Energy 2009 ; 83 : 511 – 21 . Google Scholar Crossref Search ADS WorldCat 22 Chung TW , Wu H. Comparison between random and structured packings for dehumidification of air by Licl solution in a packed bed and their heat and mass transfer correlations . Ind Eng Chem Res 1996 ; 35 : 192 – 8 . Google Scholar Crossref Search ADS WorldCat 23 Yin YG, Zhang XS , Chen ZQ. Experimental study on dehumidifier and regenerator of liquid desiccant cooling air conditioning system . Build Environ 2007 ; 42 : 2505 – 11 . Google Scholar Crossref Search ADS WorldCat 24 Peng SW , Pan ZM. Heat and mass transfer in liquid desiccant air conditioning process at low flow conditions . Commun Nonlinear Sci Numer Simul 2009 ; 14 : 3599 – 607 . Google Scholar Crossref Search ADS WorldCat 25 Wang L , Xiao F, Niu X, Gao D. Experimental study of dynamic characteristics of liquid desiccant dehumidification process . Sci Technol Built Environ 2016 ; 00 : 1 – 14 . Google Scholar OpenURL Placeholder Text WorldCat 26 Treybal , R. E. Mass Transfer Operations, 3rd ed. McGraw-Hill, New York, 1969 edition. 27 Oyieke AYA , Inambao FL. Performance characterisation of a hybrid flat-plate vacuum insulated photovoltaic/thermal solar power module in subtropical climate . Int J Photoenergy 2016 ; 1 : 1 – 15 . Google Scholar Crossref Search ADS WorldCat 28 Naik B K, Muthukumar P. Experimental investigation and parametric studieson structured packing based liquid desiccant dehumidification and regeneration system . Build Environ 2019 ; 149 : 330 – 49 . Google Scholar Crossref Search ADS WorldCat 29 Oyieke AY A, Inambao FL. Interfacial heat and mass transfer analysis in solar-powered packed-bed adiabatic liquid desiccant regeneration for air conditioning . Int J Low-Carbon Tec 2018 ; 13 : 277 – 85 . Google Scholar Crossref Search ADS WorldCat 30 Zurigat YH , Abu-Arabi MK, Abdul-Wahab SA. Air dehumidification by triethylene glycol desiccant in a packed column . Energy Conversion and Management 2004 ; 45 : 141 – 55 . Google Scholar Crossref Search ADS WorldCat 31 Park MS , Howell JR, Gray CV, Peterson J. Numerical and experimental results for coupled heat and mass transfer between a desiccant film and air in cross flow . Int J Heat Mass Tran 1994 ; 37 : 395 – 402 . Google Scholar Crossref Search ADS WorldCat © The Author(s) 2020. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
TI - Experimental assessment of heat and mass transfer characteristics of solar-powered adiabatic liquid desiccant dehumidifier and regenerator
JF - International Journal of Low-Carbon Technologies
DO - 10.1093/ijlct/ctaa013
DA - 2020-11-18
UR - https://www.deepdyve.com/lp/oxford-university-press/experimental-assessment-of-heat-and-mass-transfer-characteristics-of-gVQ5QglmNQ
SP - 477
EP - 495
VL - 15
IS - 4
DP - DeepDyve
ER -