Duplicating Freshwater Productivity of Adsorption Desalination System Using Aluminum Metal Filings

Abstract

In this paper, improving the overall heat transfer coefficient by adding aluminum species to silica gel has been studied theoretically. An adsorption desalination system is proposed, and a lumped theoretical model conducted to investigate employing the metal additives within the adsorbent bed with and without a heat recovery between condenser and evaporator. A 30% of the total mass of the adsorbent bed contents was considered to be replaced by aluminum species. According to this, the overall heat transfer coefficient has been increased by 260%, which shows a good impact on the performance of the adsorption system. Daily water productivity was increased by 70% at the worst-case, reaching up to 17 m³/day/ton of silica gel without heat recovery. By employing heat recovery with the metal filing, the daily water productivity reached 42 m³/day/ton of silica gel which is four times the productivity of the classic silica gel-based adsorption desalination system.

1. Introduction

In light of the climate and water challenges that the world faces today, it has become necessary to find solutions for the survival of mankind. Accordingly, searching for systems that provide drinking water using environmentally friendly approaches has become inevitable, regardless of its financial cost. One of these systems with a bright future is the adsorption desalination system (ADS), which proves day by day its promising capabilities as a drinking water desalination system. Although the adsorption desalination system represents a solution, its performance is still not satisfactory, as its efficiency and productivity are still low. Therefore, many researchers are trying to develop this system depending on alternative or waste energies to attenuate salinity from water. This development has taken several aspects, including presenting new adsorbent materials, improving the heat transfer coefficient, or improving the design configurations.

The ADS works mainly on two processes: (i) adsorption–evaporation and (ii) desorption–condensation. In the first process, the adsorbent adsorbs vapors from the evaporator that is filled by seawater, producing a cooling effect at the evaporator along with desalination at condenser [1,2,3,4,5]. A growing number of studies have been devoted to the field of adsorption desalination systems. For instance, Askalany et al. [6] presented a new combination between the ADS and the ejector, reaching a specific daily water production (SDWP) of 23 m³/day/ton of adsorbent. Wang et al. [7] and El-Sharkawy et al. [8] investigated an ADS experimentally with 2 and 4 beds filled with silica gel, where mass recovery was accomplished, and the SDWP reached as high as 8.2 m³/ton/day. Amirfakhraei et al. [9] performed a modeling study on ADS with mass recovery reaching an SDWP and specific cooling power (SCP) of 9.58 m³/ton/day and 51.18 R-ton/ton, respectively. The coefficient of performance (COP) of 0.58 was achieved, which was significantly high compared to the previous studies. Olkis et al. [10] fabricated a lab-scale ADS employing heat recovery where the SDWP reached 10 m³/ton/day [11,12] with a COP of 0.83. Ali et al. [13] studied an ADS with two evaporators and four adsorber beds where the evaporator worked at 10 °C. The reported SDWP was 8.09 m³/ton of silica gel per day. Effect of brine salinity on ADS performance had been conducted by Woo et al. [14]. The study demonstrated that the water salinity has no significant effect on the ADS performance. The highest SDWP achieved employing silica gel was 6.76 m³/ton per day. Ali et al. [15] presented an innovative merge between ADS and salt hydrate water desalination system. The system achieved an SDWP of 40 m³/ton of silica gel per day at 95 °C driving temperature. An ADS was modeled and experimentally verified by Bai et al. [16] that considered a mass recovery process and salinity effect. At a desorption temperature of 85 °C and evaporation temperature of 14 °C, the SCP and SDWP were 490 W/kg and 18 m³/ton/day.

Olkis et al. [17] analyzed the feasibility of a solar-powered ADS at two distinct European locations using advanced ionogel materials with regeneration temperatures as low as 25 °C. 1 m³/ton/day required 140 kg of ionogel and 200 m² area of solar collectors. Amirfakhraei et al. [18] presented an advanced mass and heat recovery design for the ADS. During this recovery cycle, a heat exchange directly occurred between the adsorption beds and evaporator, resulting in an improvement in SDWP by 31%. Elsheniti et al. [19] investigated the influence of employing metal–organic framework (AlFum) within solar powdered ADS with conventional silica gel. The silica gel showed better performance than the AlFum due to the low thermal diffusivity of Al-Fum. Average solar COPs of the silica gel system over different months were higher by 83, 43, and 22% at inlet chilled water temperatures of 15, 20, and 25 °C, respectively.

Askalany et al. [20] studied employing filings of different metals (iron, copper, and aluminum) with mass concentrations from 10 to 30% within the adsorption cooling system. The study included an experimental determination of the thermal conductivity, where 30% of Aluminum filings increased the SCP by 100%. Demir et al. [21] enhanced the heat transfer rate by adding metallic additives (copper, brass, and aluminum) to silica gel. It had been noticed that that the thermal conductivity of silica gel could be improved by 242% if 15 wt% of aluminum pieces were added to the silica gel. Rezk et al. [22] used aluminum additives in improving the performance of an adsorption cooling system. The performance of the system was improved by 58.2% when 15 wt% of aluminum pieces were consumed. Askalany et al. [23] optimized the switching and cycle times of the ADS experimentally and theoretically. The metal–organic framework (CPO-27(Ni)) had been used as an adsorbent. SDWP of 9.5 m³/ton/day had been recorded. A field analysis had been conducted for an ADS finned-flat bed employing a three-dimensional computational fluid dynamics model by Li et al. [24]. 12.5 m³/ton/day SDWP had been recorded with a SCP of 363 W/kg.

Based on the presented literature survey, there are a lot of studies that have been focused on improving the adsorption systems either for cooling and or desalination applications. One of the approaches to improve the thermal performance of the system is adding metallic additives to enhance the system performance. As a continuation of this way and following the same approach, this work has been focused on improving the performance of desalination system through the adsorption approach by raising the heat transfer coefficient using aluminum filings.

2. Theoretical Model

2.1. System Description

An adsorption desalination system was considered containing two adsorption beds where every bed had 7 kg of silica gel. A schematic diagram for the adsorption system is presented in Figure 1. The system contained an evaporator, condenser, and different water tanks for feeding saltwater, collecting pure water, and brine. The system worked in two modes, with and without heat recovery, where, in the heat recovery mode, the heat exchangers of both the condenser and the evaporator were directly connected. This raises the evaporator temperature and decreases the condenser temperature, which helps in improving the system productivity with no chance for using any cooling effect of the system.

2.2. Mathematical Model

A set of equations were used to model the prosed system containing mass and energy balances of the different parts of the system. In addition, adsorption isotherms and kinetics of the water adsorption onto silica gel were considered within the model. SDWP, SCP, and COP equations were also included in the theoretical model, as shown in the following section.

Equilibrium water uptake

The D-A model for water adsorption onto silica gel, adsorption isotherms [15].

$$C=C_{o}exp\left\{-{\left(\frac{RT}{E}ln\left(\frac{P_{sat}}{P}\right)\right)}^n\right\}$$

(1)

Mass balance equations for saltwater and condensate water mass balance

$$\frac{d{M}_{sw,evap}}{dt}={\dot{m}}_{sw,in}-{\dot{m}}_{p,cond}-{\dot{m}}_b$$

(2)

Evaporator and salt mass balance

$$\frac{d{M}_{sw,evap}}{dt}={\dot{m}}_{sw,in}-{\dot{m}}_b-\left(\frac{d_{Cads}}{dt}\right)M_{ads}$$

(3)

$$M_{sw,eva}\frac{d{X}_{sw,evap}}{dt}=X_{sw,in}{\dot{m}}_{sw,in}-X_{sw,in}{\dot{m}}_b-X_D\left(\frac{d_{Cads}}{dt}\right)M_{ads}$$

(4)

Energy balance equations employed for adsorption beds

$${\left[{\left(M{c}_p\right)}_{cu}+{\left(M{c}_p\right)}_{al}+{\left(M{c}_p\right)}_{ads}+M_{ads}{c}_p{}_{v}C\right]}^{bed}\frac{d{T}_{bed}}{dt}=M_{ads}{H}_{st}\frac{dC}{dt}-{\dot{m}}_w{c}_p{}_w\left(T_{w,out}-T_{w,in}\right)$$

(5)

Heat of adsorption

$$H_{st}=h_{fg}+E\left[ln{\left(\frac{C_o}{C}\right)}^{\frac{1}{n}}\right]+\frac{ET\alpha }{n}{\left[ln\left(\frac{C_o}{C}\right)\right]}^{\frac{1-n}{n}}$$

(6)

Condenser energy balance

$$\begin{array}{l}\left[{\left(M{c}_p\right)}_{cu}+{\left(M{c}_p\right)}_{iron}+{\left(M{c}_p\right)}_w\right] \frac{d{T}_{cond}}{dt}\\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}=h_f\left(T_{cond}\right)\frac{dM}{dt}+h_{fg}\left(T_{cond}\right)\frac{d{C}_{des}}{dt}{M}_{ads}+{\dot{m}}_w{c}_p{}_w\left(T_{cond}\right){\left(T_{w,in}-T_{w,out}\right)}^{cond}{\dot{m}}_b\end{array}$$

(7)

Evaporator energy balance

$$\begin{array}{ll}[M_{s,evap}{c}_p{}_s(T_{evap},& X_{s,evap})+M_{cu,evap}{c}_p{}_{cu,evap}] \frac{d{T}_{eva}}{dt}\\ & =h_f\left(T_{evap},X_{s,evap}\right){\dot{m}}_{s,in}-h_{fg}\left(T_{evap}\right)\frac{d{C}_{des}}{dt}{M}_{sg}+{\dot{m}}_{ch}{c}_p{}_{ch}\left(T_{ch,in}-T_{ch,out}\right)\\ & -h_f\left(T_{evap},X_{s,evap}\right){\dot{m}}_b\end{array}$$

(8)

The temperature of the outlet water of the heat exchanger is expressed as;

$$T_{w,out}=T_{hex}+\left(T_{w,in}-T_{hex}\right)exp\left(\frac{-U{A}_{hex}}{{\left(\dot{m}{c}_p\right)}_w{}_{ }}\right)$$

(9)

Cooling load, desorption, and condensation energy are given by;

$$Q_{eva} ={{\displaystyle \int }}_0^{t_{cycle}}{\dot{m}}_{ch}{c}_p{}_{ch}\left(T_{ch,in}-T_{ch,out}\right)dt$$

(10)

$$Q_{des} ={{\displaystyle \int }}_0^{t_{cycle}}{\dot{m}}_{hw}{c}_p{}_w\left(T_{hw,in}-T_{hw,out}\right)dt$$

(11)

$$Q_{cond} ={{\displaystyle \int }}_0^{t_{cycle}}{\dot{m}}_w{c}_p{}_w\left(T_{cw,in}-T_{cw,out}\right)dt$$

(12)

Cycle performance parameters

$$SDWP=\frac{24\times 60\times 60}{t_{cycle}}{{\displaystyle \int }}_0^{t_{cycle}}\frac{{\dot{m}}_w{c}_p{}_w\left(T_{cw,out}-T_{cw,in}\right)}{h_{fg}{M}_{sg}}dt$$

(13)

$$SCP ={{\displaystyle \int }}_0^{t_{cycle}}\frac{{\dot{m}}_{ch}{c}_p{}_{ch}\left(T_{ch,in}-T_{ch,out}\right)}{M_{sg}}dt$$

(14)

Coefficient of performance (COP)

$$COP ={{\displaystyle \int }}_0^{t_{cycle}}\frac{{\dot{m}}_{ch}{c}_p{}_{ch}\left(T_{ch,in}-T_{ch,out}\right)}{{\dot{m}}_{hw}{c}_p{}_w\left(T_{hw,in}-T_{hw,out}\right)}dt$$

(15)

The differential algebraic system of equations were solved in FORTRAN. The used parameters are indicated in

Table 1. Model parameters.

Parameters	Description	Value	Unit
$U{A}_{bed}$	Overall heat transfer coefficient of bed	600	W.K−1
$U{A}_{cond}$	Overall heat transfer coefficient of condenser	500	W.K−1
$U{A}_{eva}$	Overall heat transfer coefficient of evaporator	350	W.K−1
$M_{cu}$	Bed heat exchangers tube weight (Cu)	2.97	kg
$M_{al}$	Bed heat exchanger fin weight (Al)	0.72	kg
$M_{bed, iron}$	Bed heat exchanger cover weight (Iron)	15	kg
$M_{cond,cu}$	Condenser heat exchangers tube weight (Cu)	1.535	kg
$M_{eva, cu}$	Evaporator heat exchangers tube weight (Cu)	1.3	kg
$M_{w,eva}$	Liquid water inside evaporator initially	1	kg
$M_{ads}$	Weight of adsorbent in each bed	6.75	kg
${\dot{m}}_{hw}$	Heating water flow rate to the adsorber	0.2	kg/s
${\dot{m}}_{chw}$	Chilled water flow rate	0.025	kg/s
${\dot{m}}_{cw}$	Cooling water flow rate to the adsorber	0.3	kg/s
C0	Saturation uptake	0.36	kg/kg
E	Characteristic energy	167.74	kJ/kg
n	Dubinin-Astakhov fitting parameter	1.68
$C{p}_{cu}$	Copper specific heat	0.386	J/kg.K
$C{p}_{al}$	Aluminum specific heat	0.905	J/kg.K
$C{p}_{ads}$	Adsorbent specific heat	0.924	J/kg.K
$C{p}_w$	Water specific heat in liquid phase	4.18	J/kg.K
$C{p}_{ch}$	Chilled water-specific energy in the vapor phase	4.20	J/kg.K
$C{p}_v$	Water specific heat in vapor phase	1.89	J/kg.K
R	Universal gas constant	8.314	J/kg.K
Rp	Average radius of adsorbent particle	1.75	mm
Thw	Heating source temperature	95–75	°C
Tcw	Cooling source temperature	30	°C
Tch,in	Chilled water inlet temperature	30	°C
tcycle	Cycle time	650	s

[25]. Different operating conditions have been considered in the system modeling as the effect of changing driving temperature in the range of 65–95 °C has been considered. The used model has been presented and employed in a large number of studies, and its validation was proven accordingly.

3. Results and Discussions

ADS without Heat Recovery

To justify the improvement of the ADS, the most important three parameters SDWP, SCP and COP, have been determined and presented in the next figures, respectively. Figure 2 shows the effect of using aluminum species within the adsorbent silica gel on the SDWP of the system based on the driving water temperature. As expected, raising the driving temperature caused an increase in the SDWP since it increased the desorbed water vapor.

The figure also illustrates the significant effect of the aluminum filings on the productivity of the system. The SDWP was increased by 50% in the worst case, and could even be duplicated at certain driving temperatures. Using the aluminum filings causes an increase in the SDWP from 10 m³/day/ton to 15.5 m³/day/ton at a driving temperature of 95 °C. This is due to the improvement that occurs on the heat transfer coefficient of the adsorption bed because of the metallic filings, which decreases the gaps between the adsorbent’s particles with higher thermal conductivity particles. This increase in the thermal conductivity makes the heating process faster, which decreases the cycle time, causing an increase in the SDWP, according to Equation (13). There is no change in the trend of the SDWP with and without the filings, where raising the driving temperature increases the SDWP in both cases.

Change in the SCP is not less than the SDWP, and this is clearly visible in Figure 3. It is noticeable in this figure that it is possible to improve productivity from 271 W/kg to 432 W/kg by using aluminum filings at the same driving temperature. Since the SCP is also a time-dependent property, decreasing the cycle time that happens by raising the overall heat transfer coefficient has a good impact on it. The trend of SCP with raising the driving temperature has the same manner whether the aluminum filings are utilized or not.

COP of the ADS at different driving temperatures is also presented in Figure 3 with and without aluminum filings without a heat recovery system. On the contrary of the above, the effect on the COP is not significant. This can be attributed to the increase in the metal content of the bed, which means consuming a relatively larger amount of energy in heating this content and then cooling it down, which reduces the efficiency of the system. This explains the decrease in the value of COP at relatively higher temperatures than 85 °C compared to the COP of the ADS without aluminum filings. It can be said that the metallic filings have a quantity impact on the system with no significant impact on the quality of the system.

In the next part, the ADS was studied to work in heat recovery mode. In this mode, the condenser and the evaporator were connected together, where the condenser was cooled down by using the evaporator and the heat of the condenser was used to raise the pressure of the evaporator. This technique had a good impact on the performance of the ADS, as shown in the below figures.

The impact of employing the heat recovery along with using the aluminum filings on the SDWP at different driving temperatures is depicted in Figure 4. As in the case of no heat recovery, it became clear that the productivity of daily desalinated water could double in the case of heat recovery by adding aluminum filings, where 42 m³/day/ton SDWP could be achieved. The main reason for this was the increase that happened in the heat transfer coefficients by using the aluminum filings, which decreased the cycle time.

Figure 4 also shows the COP of the ADS in heat recovery mode with aluminum filings at different driving temperatures. The COP of the system was deteriorated by raising the driving temperature in the presence of aluminum filings due to raising the metallic load by adding the aluminum filings.

To conclude the different presented cases in this study, Figure 5 is conducted showing SDWP at 95 °C driving temperature with and without aluminum filings and heat recovery. It can be said that the SDWP could be increased four times by using heat recovery mode along with replacing 30% mass of the adsorbent with aluminum filings.

In order to show the achievement of this study, a comparison between the resulting SDWP and the previously presented studies was conducted and is shown in Figure 6. About 21 studies were considered, showing different achieved values of the SDWP, starting from 0.12 and reaching up to 25 cubic meters of pure water per ton of adsorbent mater per day. It is clear from the figure that by employing aluminum filings along with heat recovery, the system can reach unprecedented values of the SDWP. Compared with the highest presented values of the SDWP, the system can increase the productivity by at least 50%. This confirms that the presented idea is fruitful and it greatly improves the operation of the system, which leads to energy and cost savings.

It is important to validate the presented data experimentally, however, there is difficulty in building an experimental test rig and managing experimental work in this field. This work is a necessary introduction to prove the validity of the hypothesis of the efficiency of this system after improving the overall heat transfer coefficient. The theoretical model used in the study is approved in many previous studies in this field, and researchers have always used it, as it has been proven to be true in the laboratory in many research, and this has been mentioned in the context of the presented work. Knowing that it is recognized that improving the heat transfer coefficient will certainly improve the performance of the adsorption desalination system, and this is an axiom. Therefore, there is no dispute over the preference of the proposed system, but there may be a dispute over the amount of improvement, and this is what will be proved practically in the second part of the study, which will be worked on.

4. Conclusions and Future Work

In this paper, the effect of adding aluminum in the form of aluminum filings to the sorbent material on the performance of the ADS was studied. A mathematical model for an ADS employing silica gel as an adsorbent was considered. In total, 30% of the adsorbent silica gel was considered to be replaced by aluminum filings. The system was modeled to work at different driving temperatures in the range of 65–95 °C. Four different cases were tested, including the use of aluminum filings and/or the consideration of heat recovery.

The results demonstrate the positive impact of using the aluminum filings as the productivity of desalinated water is increased by at least half, with the possibility of doubling it. By using the heat recovery technique, the SDWP reached as high as 40 m³/ton/day at 95 °C driving temperature; SCP could be increased by about 60% by using the aluminum filings without heat recovery. The COP shows no significant effect by using the metallic filings. By comparing the presented results with previously presented results, a great improvement is shown.

Regarding future work, it is very important to continue working, based on the presented results, to build an adsorption desalination system powered by solar energy resulting in a relatively high rate of purified water.