Experimental study of a bubble basin intended for ...

2 downloads 0 Views 869KB Size Report
One new technique that can increase the water evaporation rate in the solar still is to generate air bubble through its base. The aim of this research work is to ...
DES-13033; No of Pages 6 Desalination xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Desalination journal homepage: www.elsevier.com/locate/desal

Experimental study of a bubble basin intended for water desalination system Hanen Ben Halima ⁎, Nader Frikha, Slimane Gabsi Research Unit of Environment, Catalysis and Process Analysis, National Engineering School of Gabes, Omar Ibn El Khattab Avenue, Gabes 6029, Tunisia

H I G H L I G H T S • The influence of bubbling air on water evaporation rate in a simple basin has been investigated experimentally. • The heat and mass transfer coefficients between air and water are calculated and fitted in forms of empirical correlations. • The results show a good agreement between the proposed model and the experimental measurements.

a r t i c l e

i n f o

Article history: Received 14 March 2016 Received in revised form 20 July 2016 Accepted 1 August 2016 Available online xxxx Keywords: Solar still Air bubbling Operating conditions Heat and mass transfer Desalination

a b s t r a c t One new technique that can increase the water evaporation rate in the solar still is to generate air bubble through its base. The aim of this research work is to perform a design basis for a modified solar still using air bubbling through water layer and to determine the heat and mass transfer coefficients between air bubbles and water in the basin. An experimental test set-up was fabricated and assembled. The process consists of air bubbles passing through the base of the basin to the seawater under some experimental conditions of airflow rate, inlet air temperature and humidity, and water temperature and depth. Detailed experiments were carried out at various operating conditions. Within the studies ranges, the results indicate that the water vapor content difference is moderately affected by the water temperature and airflow rate but slightly affected by the water level. The heat and mass transfer coefficients were obtained experimentally and fitted in forms of empirical correlations. The statistical study has shown a good agreement between the model and the experimental measurements, with an average relative error not exceeding 20%. © 2016 Elsevier B.V. All rights reserved.

1. Introduction The scarcity of fresh water is a highly important issue as most of the world population suffers from clean water shortage. Although water is available throughout the earth, only 1% of it is potable water [1]. One potential solution to tackle this issue is to develop a reliable, efficient and cost effective decentralized water desalination system to make clean water accessible for most of the world population. Countries having inadequate available water supplies have obtained fresh water from the sea using fossil fuels for a long time. Over the years, people realized that the use of fossil fuels is not a sustainable way as it damages environment. Nowadays, solar desalination has become a very affordable solution to cope with fresh water shortage, especially in remote areas, where solar radiation is available abundantly but with a bad water quality. The conventional solar still is one of the simplest and most promising techniques used to distill water. However, the major disadvantage of solar still is the low productivity. Actually, even in areas of relatively high solar radiation levels, its annual performance is limited to an average of about 3 L/m2·day [2]. This problem has motivated researchers to ⁎ Corresponding author. E-mail address: [email protected] (H. Ben Halima).

investigate various methods that would improve the conventional still productivity. Among the methods used are those concerned with cooling the glass cover. As the water vapor condenses on the glass cover, its latent heat is released to the glass cover, which increases the cover temperature and lowers the temperature difference between the water in the basin and the glass cover, thus reducing the driving force for water evaporation. Other approaches have been used to increase basin water temperature, evaporation and condensation surface areas. Y. H. Zurigat et al. [2] have proposed a regenerative solar still that consists of two basins (effects). In their proposed still, the condensation latent heat released to the first glass cover is utilized to produce additional fresh water from a second effect. The second effect may be arranged in such a way that it would have either a flowing water film or a stationary one of larger thickness. The results have confirmed that the regenerative still gives N20% higher productivity in comparison to the conventional still. A. Ghazy et al. [3] have undertaken an analytical study of a direct solar distillation system that combined solar still with an air heating humidification-dehumidification sub-system. Various procedures have been employed to improve the thermal performance of the integrated system by recovering heat losses from one component in another component of the system. Simulations have been carried out for the

http://dx.doi.org/10.1016/j.desal.2016.08.003 0011-9164/© 2016 Elsevier B.V. All rights reserved.

Please cite this article as: H. Ben Halima, et al., Experimental study of a bubble basin intended for water desalination system, Desalination (2016), http://dx.doi.org/10.1016/j.desal.2016.08.003

2

H. Ben Halima et al. / Desalination xxx (2016) xxx–xxx

performance of the Still-HDH system under different weather conditions. A comparison has been held between the Still-HDH system and a conventional solar still of the same size and under the same operating conditions. They found that the total water production from the StillHDH system is about 6.5 L/m2/day as compared to 4.2 L/m2/day from the conventional still, under the same conditions. In a previous work, H. Ben Halima et al. [4] developed an analytical model of a simple solar still coupled to a compression heat pump. The compression heat pump is made up of a condenser immersed in the water basin, an evaporator located below the upper region of the glass cover, a compressor and an expander. The condenser contributes to heat the basin water, and thus allows its evaporation thanks to the refrigerant flow through the heat pump. On the other hand, the evaporator condenses a large part of the water vapor. The results have proven that the productivity of this type of solar still is 75% higher than that of the conventional one. The daily production of the still reached 13.5 kg/m2 in June 21. A. E. Kabeel et al. [5] investigated an experimental study of a double passes solar air collector–coupled modified solar still, with Phase Change Material (PCM). A comparison between modified still and PCM, forced hot air injection and conventional still was conducted to evaluate the development in the freshwater productivity under the same atmospheric conditions. The experimental results have revealed that the daily freshwater productivity for modified still is higher than that of conventional still. The freshwater productivity reached approximately 9.36 L/m2·day for modified still while its value was 4.5 L/m2·day for the conventional one, with percentage of increase of 108%. R. Sathyamurthy et al. [6] presented an experimental analysis of a portable solar still with evaporation and condensation chambers. The phase change material (PCM) is used in order to divide a single slope portable solar still into evaporating and condensing chambers. The result shows that the accumulated yield obtained with PCM is 52% more important than accumulated yield obtained for still without PCM. The still continues to produce fresh water after the sunset. In conventional solar still, the covering glass serves two purposes: a solar radiation transmitter and a condenser. However, since it is exposed to radiation and because it relies on passive cooling by natural air convection, its condensation capacity is more limited. Moreover, solar radiation might re-evaporate some of the formed condensate. One of the possible ways to increase the capacity and thus the productivity of a solar still is to add a separate condenser. Ayman G.M. Ibrahim et al. [7] conducted an experimental study of a modified basin type solar still equipped with an air-cooled condenser. The experimental results showed an enhancement of 16.2% and 29.7% in productivity and thermal efficiency, respectively, compared with the conventional solar still. P. Refalo et al. [8] used a solar chimney and condensers to enhance the productivity of a solar still. Condensers in solar stills typically consist of seawater flowing through a bank of tubes. However, in their proposed configuration, water vapor was passed through a number of ducts immersed in seawater. They found that the externally watercooled condensers coupled to the solar chimney improved condensation by separating and shifting the condensation process from the evaporation chamber to the condensers. Moreover, when comparing the efficiency based on the actual basin area, it was noted that the solar still with the solar chimney and condensers performed 8.8% better. One new technique that can increase the water evaporation rate in the solar still is to generate air bubble through its base. The aim of this work is to perform a basis design for a modified solar still using air bubbling through water layer and to determine the heat and mass transfer coefficients between air bubbles and water in the basin. To this end, a laboratory scale basin with air diffuser was built. The experiments were carried out by bubbling air in the hot water in order to investigate the influence of the operating conditions such as the airflow rate and the hot water temperature and depth on the water evaporation rate. To simulate the solar energy, electric water heater was used.

2. Experimental setup and procedure 2.1. Experimental setup A schematic diagram of the experimental apparatus is presented in Fig. 1, showing that the system consists mainly of an air compressor, a simple basin equipped with an air diffuser, an air flow meter and an electric water heater. The air is delivered from the ambient milieu by the air compressor to the basin water through a perforated plate installed at the bottom of the basin which distribute the air and generate bubble. So, air bubbles will be charged by water vapor when passing through the hot water in the basin then leaves from 10 cm outlet vent. The basin was made of steel with 47 cm in length, 30 cm in width and 25 cm in height. It was insulated through the side and bottom by a 3-cm glass wool layer. The system temperatures were checked before the implementation of any heating loads to guarantee a uniform temperature environment. Every run is accomplished when a steady state condition is achieved. At this steady state condition, all measuring variables fluctuate within their uncertainty tolerances, and continuously carried out for new experimental set conditions. The air temperature and humidity at the basin inlet and on the water surface were controlled by 2 thermo-hygrometers. The water level in the basin was controlled by a graduate level and an electric heater used to heat the water. The temperature of the water was adjusted to the desired degree through a digital reading and a thermo-regulator controller. The supply of airflow was adjusted to the desired flow through the control valve. The heat side and bottom losses were found to be negligible. The experimental data were obtained under the steady state conditions of heat and mass transfer. Once equilibrium was reached, the measurements of airflow rate, water temperature and air temperature and humidity at the basin inlet and on the water surface were taken. The experimental conditions are as follows: Water temperature: Tw = 40–70 °C, Water level: Z = 3–13 cm, Dry air mass flow rate: G = 3.8*10−3 − 15.4*10−3 kg/m2·s, Inlet air temperature: 27–30 °C. 2.2. Experimentation error analysis The measurements of the parametric variables, air flow rate, water temperature, water level and relative air humidity and temperature at the basin inlet and water surface, were taken during the experiments. The air flow rate was measured using the rotameter (3) with a range of 5–2000 L/h and an uncertainty of 4.6%. The water temperature in the basin was measured using the thermometer-Pt100 (class B) (7) which works in the range from −20 to +260C with an uncertainty of 2.6%. The relative humidity and temperature of air streams was measured using 2 thermo-hygrometers (4, 10) which work in the range from 0 to 100% RH and from −40 to +120C and its uncertainty is 1.4%. 3. Theoretical analysis 3.1. Air diffuser design The air diffuser consists of steel rectangular box with 47 cm in length, 30 cm in width and 2 cm in height. A rubber plate covers this box and consists of 1168 orifices with 1 mm in diameter each (Fig.2). The air diffuser is an important accessory in such a system, since it provides the bubbles distribution. In order to obtain a homogenous distribution of bubble in the basin, analytical correlations were used for designing the perforated plate. The number of holes in the perforated plate was calculated by using the following relation: n¼

GAt U b A0

ð1Þ

Please cite this article as: H. Ben Halima, et al., Experimental study of a bubble basin intended for water desalination system, Desalination (2016), http://dx.doi.org/10.1016/j.desal.2016.08.003

H. Ben Halima et al. / Desalination xxx (2016) xxx–xxx

3

Bubbler humidifier Inlet air

Humid air (9) (8)

(13) (11)

(10)

(12) (6)

(7)

(1)

(3)

(4)

(2)

(5) Fig. 1. Experimental setup: (1) – Compressor, (2) – Control valve, (3) – Airflow meter, (4, 10) – Thermo-hygrometer (5) – Air diffuser (6) – Electric heater, (7) – Thermometer-Pt100, (8) – Graduate level, (9) – Transparent cover, (11) – Water level, (12) – Air stream, (13) –Air diffuser orifices.

whereUb is the bubble velocity from orifice. It is calculated from [9]: Ub ¼

gρw 2 d 18μ w b

ð2Þ

db is the size of bubbles produced at an orifice. It is given by [10]:  db ¼

6σ d0 ðρw −ρair Þg

1=3 ð3Þ

3.3. Heat and mass transfer coefficients calculation The heat and mass transfer coefficients between air and water in the bubbler basin are calculated from the experimental data and heat and mass transfer balances. The volumetric heat transfer coefficient of sensible-heat transfer between the air and water is calculated using the following relation [11]:   G:C Pair : T air;out −T air;in hG :as ¼ Z:LMTD

ð7Þ

This air diffuser consists of 1168 orifices with 1 mm in diameter each (Fig. 2).

where LMTD is the logarithmic mean temperature difference between the air and water, which is:

3.2. 3.1. Data acquisition system

LMTD ¼

The water vapor content difference in the humid air at the water surface is defined as: ΔY ¼ Yout −Y in

ð4Þ



Yout −Y in  100 Yout;sat −Y in

ð5Þ

where Yout , sat is the outlet saturation humidity. The value of Yout, sat is calculated assuming that the outlet air is saturated at the water temperature. The water evaporation rate is calculated using the following formula: P ¼ ΔYG3600

ð6Þ

ð8Þ

The volumetric mass transfer coefficient between the air and water is calculated using the following relation [11]: kG :as ¼

The bubbler basin efficiency is given by:

ðT airout −T airin Þ ðT w −T airin Þ ln ðT w −T airout Þ

G:ðY out −Y in Þ Z:LMHD

ð9Þ

where LMHD is the logarithmic mean humidity difference between the water and the air, which is: LMHD ¼

ðY out −Y in Þ   Y out;sat −Y in  ln  Y out;sat −Y out

ð10Þ

4. Experimental results and discussion 4.1. Parametric study

Fig. 2. Diffuser plate structure.

The experimental measurements were conducted for the bubbler basin under different operating conditions. The system was operated under steady state conditions. Once equilibrium was reached, the measurements of the airflow rate, water temperature and air temperature and humidity at the basin inlet and water surface were taken. The heat and mass transfer coefficients were calculated using the previously mentioned relations. The obtained results show the influence of the design parameters on the vapor content difference, air humidification efficiency and evaporated water flow rate. In fact, Figs. 3 and 4 present the effect of the airflow rate on the vapor content difference and air humidification efficiency at different hot water temperatures. It is seen from this figures that the vapor content difference increases with the increase of the air flow rate. However, this is more important at high water temperature than at low one. Thus, a low airflow rate value is recommended for low water temperatures so as to reduce the air compressor power

Please cite this article as: H. Ben Halima, et al., Experimental study of a bubble basin intended for water desalination system, Desalination (2016), http://dx.doi.org/10.1016/j.desal.2016.08.003

4

H. Ben Halima et al. / Desalination xxx (2016) xxx–xxx

0,3

0,25

Z=10 cm

Tw=40 °C

Y

0,25

Vapor content difference, (kgw /kg dryair )

Tw=40 °C

0,2

Tw=60 °c

Tw=60 °C 0,2

G=6.65*10 -3 kg/m 2/s

Tw=50 °C

Tw=50 °C

Tw=70 °C

Tw=70 °C 0,15

0,15 0,1

0,1 0,05

0,05

0

0

0

0,005

0,01

Air flow rate,

0,015

0

0,02

2

4

6

8

10

12

14

Water level (cm)

G (kg/m 2/s)

Fig. 3. Effect of airflow rate and water temperature on the vapor content difference.

consumption. The increase of the water temperatures enhances the vapor content difference. Therefore, the ability of air to carry the water vapor increases with the increase in water temperature. It is clear from Fig. 4 that the air humidification efficiency is proportional to the air flow rate and water temperature for the already set ranges. For high water temperature values and airflow rates, the humidification efficiency can reach almost 100%. Fig. 5 indicates that the vapor content difference is slightly affected by the water level for different water temperatures. It is remarkable that the effect of water temperature is more important than that of water level. However, it should be taken into account that when the thickness of water increases, the mass of water in the basin increases, which requires a larger amount of energy to heat. On the other hand, the pressure drop in the basin was found to increase with the increase of water level. Therefore, in order to ensure a high humidification capacity, the water level needs to be low, which would result in a low pressure drop and less energy consumption. The evaporated water rate is presented in Fig. 6 which shows that the water vapor amount influx in the air stream is more important for high airflow rate and water temperature values. Therefore, the maximum evaporated water rate reached was14 kg/m2/h of water at 70 °C water temperature and 15.4 ∗ 10−3 kg/m2·s airflow rate.

Fig. 5. Effect of water level and temperature on the vapor content difference.

4.2. Experimental heat and mass transfer coefficients correlations In order to evaluate the system performance accurately and to obtain the heat and mass transfer coefficients, the following procedure has been adopted. The energy and the mass lost by hot water and gained by air were obtained using the measured flow, temperature and humidity of air inlet and exit and water temperature. The logarithmic mean temperature and humidity difference were calculated and the overall heat and mass transfer coefficients were obtained. Several experiments were conducted to obtain a large data set for the regression analysis. A computer program was constructed to perform the multiple nonlinear curve fitting using least squares method by MATLAB software. The measured data were provided as the program input, and correlations were obtained in the form of: F ¼ αGβ T γw Z θ

ð11Þ

where β, γ, θ are dimensionless constants and α has dimensions such that both sides of the equations are of the same units. This form was used as suggested in previous publications [12,13]. 4.2.1. Heat and mass transfer identification The predefined ‘fminsearch’ function was used to identify the parameters (α, β, γ and θ). The average relative error was used as the objective function to identify the parameter values for the performed 2,5

100

Tw=50 °C

2

80

Tw=60 °C

70

Productivity, P (kg w /h)

Humidification efficiency, E

Z=10 cm

Tw=40 °C

Z=10 cm

90

60 50 40 Tw=40 °C

30

Tw=70 °C

1,5

1

Tw=50 °C

20

Tw=60 °C

10

Tw=70 °C

0,5

0

0 0

0,005

0,01

0,015

0,02

Air flow rate, G (kg/m2/s) Fig. 4. Effect of airflow rate and temperature on the humidification efficiency.

0

0,005

0,01

0,015

0,02

Air flow rate, G (kg/m 2/s ) Fig. 6. The estimate productivity of the bubbler basin.

Please cite this article as: H. Ben Halima, et al., Experimental study of a bubble basin intended for water desalination system, Desalination (2016), http://dx.doi.org/10.1016/j.desal.2016.08.003

H. Ben Halima et al. / Desalination xxx (2016) xxx–xxx

experiments. The aim of the procedure is to obtain the corresponding α, β, γ and θ values to the minimum average relative error value. relative error ¼

 2 !1=2 1 n F expi −F cali ∑ n i¼1 F expi

5

with tα/2,(n-m) is the Student value, we have taken a confidence level of 95% (α = 0.05). The uncertainty can be defined by:

ð12Þ σ ai ¼

t α=2;ðn;−mÞ Sai 100 ai

ð19Þ

where n is the number of experimental and calculated points. The average relative error between the calculated and measured values was minimized using the subroutine ‘fminsearch’. The values of the identified parameters correspond to the minimum relative error. The regression procedure which is based on the simplex method and using the Nelder-Mead simplex algorithm is used by the ‘fminsearch’ function. The simplex Method is an iterative procedure that allows the improvement of the solution at each step. This procedure is finished when the improvement of the solution is not possible. The NelderMead simplex algorithm is designed to solve the classical unconstrained optimization problem of minimizing a given nonlinear function.

 n  ∑i¼1 F exp;i F mod;i −F exp F mod r ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi r¼   ffi 2 2 n n ∑i¼1 F exp;i 2 −F exp ∑i¼1 F mod;i 2 −F mod

4.2.2. Parameters uncertainties determination To perform a comprehensive statistical analysis of all experiments, we define a function D corresponding to the sum of squared deviations calculated over all the performed experiments.

4.2.3. Heat and mass transfer coefficients correlations The relation between the total heat transfer coefficient and operating conditions was obtained by the following equation:



2 1 n  ∑ F expi −F cali n i¼1

ð13Þ

In order to establish the Information Matrix, we first determined the

We can also calculate a correlation coefficient given by the following equation [16]:

ð20Þ

with F exp and F mod representing the mean value of the experimental temperature and calculated by the model.

−0:97 hG :as ¼ 593G1:45 T 0:91 w Z

ð21Þ

The resulting correlation of the mass transfer coefficient is given by the following equation:

2

matrix of the second partial derivative (∂∂aD2 ) [14,15]. This matrix can be calculated as follows:

−0:98 kG :as ¼ 4:3810−2 G2:15 T 2:21 w Z

ð22Þ

2

∂ D ¼ 2ΦT Φ ∂a2

ð14Þ

‘a’ denotes the parameters vector to be determined: a = {α, β, γ, θ}. α⁎, β⁎, γ⁎ and θ⁎are the determined average values of α, β, γ and θ. The matrix terms are numerically calculated by taking the approximation of derivatives by finite differences. For the numerical computation gradients, we have taken a value ofε = 10−3. By calculating the Information Matrix (ΦTΦ) inverse, we can determine the covariance matrix of parameters (Cov):  −1 Cov ¼ ΦT Φ ¼

!−1 2 1∂ D 2 ∂a2

ð15Þ

The statistical analysis is achieved relying on performed experiments. The variance values for all the parameters is in the range of 0.3% b σai b 2.86% and the correlations coefficients values are rhG . as = 0.9836 and rkG .as = 0.9815. The proposed model gives a clear description of the performed experiments. Indeed, the parameter uncertainties values are relatively low (0.3% b σai b 2.86%) and the correlation coefficient value is very close to 1 (rhG . as = 0.9836 and rkG . as = 0.9815). This shows the good agreement between the model and the experimental measurements. The predicted results and experimental data for heat and mass transfer are reported in Figs. 7 and 8. This comparison shows that our correlation successfully predicted 72 values measured in the bubbler basin. Indeed, the fit average relative error was 0.1946 (≈20%). Fig.7 shows that our model predicts the experimental kG·a values (72 points) very well.

The diagonal elements of the matrix Cov: Cov1,1, Cov2,2, Cov3,3 and Cov4,4 correspond to the covariance of parameters α, β, γ and θ, respectively. The variance of the identified parameters is given by the following equation: S2ai ¼ Covi;i S2y

ð16Þ

S2yis the variance estimator: S2y ¼

DðaÞ n−m

ð17Þ

n and m are the total number of points and the number of parameters (n = 72, m = 4), respectively. The confidence interval of the parameter ‘ai’ is given by: ai −t α=2;ðn−mÞ Sai bai bai þ t α=2;ðn−mÞ Sai Thus ai ¼ ai  t α=2;ðn−mÞ Sai

ð18Þ

Fig. 7. Comparison of predicted heat transfer coefficient with experimental results.

Please cite this article as: H. Ben Halima, et al., Experimental study of a bubble basin intended for water desalination system, Desalination (2016), http://dx.doi.org/10.1016/j.desal.2016.08.003

6

H. Ben Halima et al. / Desalination xxx (2016) xxx–xxx

airflow rate (kg/m2·s) gravitational acceleration (m/s2) volumetric heat transfer coefficient (W/m3·k) volumetric mass transfer coefficient (kg/m3·s) logarithmic mean humidity difference (kgw/kgdry air) logarithmic mean temperature difference (°C) orifice number productivity (kg/h) water temperature (°C) inlet air temperature (°C) outlet air temperature (°C) bubble velocity (m/s) inlet humidity (kgw/kgdryair) outlet humidity (kgw/kgdryair) outlet saturation humidity at water temperature (kgw/ kgdryair) Z water level (cm) Greek letters μ dynamic viscosity (kg/m·s) ρ density (kg/m3) σ surface tension (N/m)

G g hG·as kG·as LMHD LMTD n P Tw Tair,in Tair,out Ub Yin Yout Yout,sat

Fig. 8. Comparison of predicted mass transfer coefficient with experimental results.

5. Conclusion An experimental study was performed to investigate the influence of various operating conditions (water temperature and level and airflow rate) on the air humidification effectiveness in a simple structure bubbler basin. The results have shown that the bubbler basin has higher efficiency. In the studied range of experimental conditions, humidity can reach almost 100%. The effect of water temperature is significant on the water vapor content, which is slightly affected by the water level, and humidifier efficiency. Since the effect of airflow rate is negligible for low temperature, a low value of air flow rate is recommended for low water temperature in order to reduce the air compressor power consumption. In this study the heat and mass transfer coefficients between air bubble and water inside the simple bubbler basin are obtained experimentally. Empirical correlations are developed to express these coefficients in terms of the airflow rate and water temperature and level in the basin. The proposed model gives a very good description of the performed experiments. Indeed, parameter uncertainties values are relatively low and the correlation coefficient value is very close to 1. This study will serve as a basis design tool for the construction of a new desalination system. Nomenclature a parameters vector as specific gas-liquid interfacial area (m2/m3) A0 orifice section (m2) At basin cross section area (m2) Cp,air air specific heat(kJ/kg·°C) d0 orifice diameter (m) db bubble diameter (m) E efficiency

References [1] C. Yildirim, I. Solmus, A parametric study on a humidification–dehumidification (HDH) desalination unit powered by solar air and water heaters, Energy Convers. Manag. 86 (2014) 568–575. [2] Y.H. Zurigat, M.K. Abu-Arabi, Modelling and performance analysis of a regenerative solar desalination unit, Appl. Therm. Eng. 24 (2004) 1061–1072. [3] A. Ghazy, H.E.S. Fath, Solar desalination system of combined solar still and humidification–dehumidification unit, Heat Mass Transf. (2016), http://dx.doi.org/10.1007/ s00231-016-1761-1. [4] H. Ben Halima, N. Frikha, R. Ben Slama, Numerical investigation of a simple solar still coupled to a compression heat pump, Desalination 337 (2014) 60–66. [5] A.E. Kabeel, M. Abdelgaied, M. Mahgoub, The performance of a modified solar still using hot air injection and PCM, Desalination 379 (2016) 102–107. [6] R. Sathyamurthy, S.A. El-Agouz, V. Dharmaraj, Experimental analysis of a portable solar still with evaporation and condensation chambers, Desalination 367 (2015) 180–185. [7] A.G.M. Ibrahim, E.E. Allam, S.E. Elshamarka, A modified basin type solar still: Experimental performance and economic study, Energy 93 (2015) 335–342. [8] R. Paul, G. Robert, A. Stephen, The use of a solar chimney and condensers to enhance the productivity of a solar still, Desalin. Water Treat. 1–14 (2015). [9] N. Kantarcia, F. Borakb, K.O. Ulgena, Bubble column reactors, Process Biochem. 40 (2005) 2263–2283. [10] H. Liu, M.H. Sharqawy, Experimental performance of bubble column humidifier and dehumidifier under varying pressure, Int. J. Heat Mass Transf. 93 (2016) 934–944. [11] R.E. Treybal, Mass Transfer Operations, third ed. McGraw-Hill, NY, 1980. [12] E.H. Amer, H. Kotb, G.H. Mostafa, A.R. El-Ghalban, Theoretical and experimental investigation of humidification–dehumidification desalination unit, Desalination 249 (2009) 949–959. [13] J.C. Kloppers, D.G. Kröger, Refinement of the transfer characteristic correlation of wet-cooling tower fills, Heat Transfer Engineering 26 (4) (2005) 35–41. [14] J. P. Corriou, commande des procédés, Lavoisier Technique and Documentation, Paris 2012. [15] J.P. Corriou, Process Control-theory and Applications, Springer-Verlag, London, 2004. [16] E.B. Manoukian, Guide de statistique appliquée, Hermann, Paris, 1986.

Please cite this article as: H. Ben Halima, et al., Experimental study of a bubble basin intended for water desalination system, Desalination (2016), http://dx.doi.org/10.1016/j.desal.2016.08.003