- Email: [email protected]
Simultaneous Removal of Organic Compounds from Wastewater Using Reverse Osmosis Process: Modelling, Simulation, and Optimisation
Mario R. Eden, Marianthi Ierapetritou and Gavin P. Towler (Editors) Proceedings of the 13th International Symposium on Process Systems Engineering – PSE 2018 July 1-5, 2018, San Diego, California, USA © 2018 Elsevier B.V. All rights reserved. https://doi.org/10.1016/B978-0-444-64241-7.50306-2
Simultaneous Removal of Organic Compounds from Wastewater Using Reverse Osmosis Process: Modelling, Simulation, and Optimisation Mudhar A. Al-Obaidia,b, Chakib Kara-Zaïtria and Iqbal M. Mujtabaa* a
Chemical Engineering, School of Engineering, Faculty of Engineering and Informatics. University of Bradford. Bradford, West Yorkshire BD7 1DP, UK b Middle Technical University, Iraq – Baghdad [email protected]
Abstract The modern industrialized world is generating a huge amount of wastewater containing a variety of micro-pollutants which is discharged into rivers and oceans leading to disruption of the biological ecosystem. The reverse osmosis (RO) process is one of the most promising technologies to produce high-quality recycling water at a reasonable cost. In this work we develop a new mathematical model to predict the performance of the RO process for the removal of several organic and non-organic compounds from wastewater simultaneously. The realistic operating conditions ensuring high rejection of multicompounds is explored via the simulation of the RO process first. The model is then used for optimising the operating conditions of the process while maximising the rejection and permeate recovery. Keywords: RO Process, Wastewater Treatment; Modelling; Simulation; Optimisation; Multi-compound rejection.
1. Introduction The main concern of any industrial wastewater is the existence of several organic and non-organic compounds which are harmful to human beings and marine life. Despite the availability of several wastewater treatment methods such as adsorption, biological, oxidation, and electrochemical process; the Reverse Osmosis (RO) process has demonstrated its competitive performance at a reasonable cost for the removal of these compounds. Several previous studies modelled the spiral wound RO process considering the removal of a single organic contaminant from wastewater (Sundaramoorthy et al., 2011; Al-Obaidi et al., 2017). However, only a few attempted the modelling of the spiral wound RO process for the removal of several organic and non-organic compounds from waste water. Al-Bastaki (2004) developed a lumped model to study the performance of a spiral wound RO process for removing Na2SO4 and methyl orange dye from wastewater. Note however, the development of a distributed model for the spiral wound RO process for the removal of multi-compounds simultaneously from wastewater has not been considered yet and is the main focus of this work. Also note, in the absence of experimental data in the literature for multi-compounds removal from wastewater, the model developed in this work has been validated against experimental data available for the removal of a single compound (chlorophenol) from wastewater. The model is then used to examine the rejection of organic and non-organic compounds under various operating parameters which in turn evaluates the process performance. This is followed
M. A. Al-Obaidi et al.
1868
by embedding the model in a multi-objective optimisation framework to simultaneously maximise the rejection and the total permeate recovery.
2. RO Process Model In this work RO filtration system of Sundaramoorthy et al. (2011) is used to investigate the simultaneous removal of multi-compounds from wastewater. Table 1 shows the new one-dimensional model equations for the process based on solution-diffusion model and the film theory. See Al-Obaidi et al. (2017) for a similar model but for single compound rejection. Due to the multi-compounds nature of the wastewater, the contribution of all of them to the osmotic pressure was considered in the new model. The model is validated using experimental data of chlorophenol removal from wastewater (Sundaramoorthy et al., 2011) which shows a very good agreement in terms of rejection and recovery of water (Fig. 1). The process model shown in Table 1 can be written in the compact form: f(x, u, v) = 0, where x is the set of all algebraic variables, u is the set of decision variables and v denotes the constant parameters of the process. The physical properties of seawater were used to calculate the diffusivity (ܦ ), viscosity (ρୠ ǡ ρ୮ ) and density (ɏୠ ǡ ɏ୮ ) due to low concentration of wastewater. The characteristics of the membranes used are given in Table 2. Five organic compounds and three inorganic species are assumed to be in the wastewater. The solute transport parameters (ୱ ) of the selected compounds were gathered from the literature and given in Table 3.
3. Process simulation: Effect of operating parameters Fig. 2 shows the variation of feed pressure, osmotic pressure, and water flux along the membrane length (x-axis). The feed pressure decreases along the x-axis due to pressure drop caused by the friction. This in turn reduces the water flux as a result of decreasing diving force. An increase of total osmotic pressure along the x-axis is noticed due to increasing accumulated concentration of the solutes at the membrane wall. The variation of inlet feed concentration of the pollutants found in wastewater is expected. Therefore, the RO process performance is investigated here using a range (350 to 500 ppm) of each pollutant concentration carried out at fixed feed flow rate, pressure, and temperature. Fig. 3 shows the effect of increasing feed concentration of each component on the rejection and total permeate recovery. In the selected range of 350 – 500 ppm of solute feed concentration, there was no considerable effect on the rejection. However, increasing the feed concentration of all the components causes a continuous reduction in permeate recovery due to increased osmotic pressure. 0.89 R² = 0.9493
0.83
Rec (The.)
Rej (The.)
0.86
0.8 0.77 0.74 0.72 0.74 0.76 0.78 0.8 0.82 0.84 Rej (Exp.)
29 25 21 17 13 9 5
R² = 0.9879
8
11
14
Fig. 1. The model validation results
17
20
23
Rec (Exp.)
26
29
Simultaneous Removal of Organic Compounds from Wastewater
1869
Table 1. Model equations of a spiral wound RO for multi-compounds wastewater Model Equations Specifications
no.
୵ሺ୶ሻ ൌ ୵ ൜οୠሺ୶ሻ െ ౭ሺ౮ሻ େ౦ሺሻሺ౮ሻ
ʹ͵Ǥͳͷሻ൬
౩ሺሻ
σ୬୧ ሺሺୠ
൰൨ൠ ǡ are the solute under
consideration and total number of solutes ୢౘሺ౮ሻ ୢ୶ ୢౘሺ౮ሻ ୢ୶
ൌ െ ୵ሺ୶ሻ ୠሺ୶ሻ ൌ ୢ౦ሺ౮ሻ
ൌെ
୲
ୢ୶
ቀ౭ሺ౮ሻ େ౦ሺሻሺ౮ሻቁ ୲
ౘሺ౮ሻ ୢେౘሺሻሺ౮ሻ ୲
ୢౘሺ౮ሻ
୲
ୢ୶
ൌ െ ୠሺ୶ሻ
οୠሺ୶ሻ ൌ ୠሺ୶ሻ െ ୮
ୢ୶
େౘሺሻሺ౮ሻ ୢౘሺ౮ሻ
ౘሺ౮ሻ
ୢ୶
ୢ
ൌ
ୢ୶
ቂୠሺ୧ሻሺ୶ሻ
ୢେౘሺሻሺ౮ሻ ୢ୶
ቃെ
ቀ౭ሺ౮ሻ େౘሺሻሺ౮ሻቁ ౘሺ౮ሻ ୈమ ౘሺሻሺ౮ሻ ୲
൰Ǥଷଷଷ
Ǥ଼ଷ Ǥଵଶଽ ሺ୧ሻሺ୶ሻ ሺʹݐ ሻ ൌ ʹͶǤͻୠሺ୧ሻሺ୶ሻ Ǥଵଵ ሺ୶ሻ ୮ሺ୶ሻ ୫ሺ୧ሻሺ୶ሻ Ǥଷଽ Ǥଵଷହ ሺ୧ሻሺ୶ሻ ሺʹ ሻ ൌ ͳͶǤͶୠሺ୧ሻሺ୶ሻ Ǥଵଷ ሺ୶ሻ ୮ሺ୶ሻ ୫ሺ୧ሻሺ୶ሻ
୮ሺ୶ሻ ൌ
Feed solute concentration (kmol/m³)
1
2,3, 4,5
6,7
୲
ሺ୧ሻሺ୶ሻ ൌ ͳǤͳ ൬
୫ሺ୧ሻሺ୶ሻ ൌ
Permeate flux (m/s) at any point along the xaxis. The second term is total osmotic pressure (atm) Feed flow rate (m³/s), velocity (m/s), pressure (atm) and permeate flow rate (m³/s) and operating pressure (atm)
େౘሺሻሺ౮ሻ
ሺ୶ሻ ൌ
౭
ଶౘሺ౬ሻሺ౮ሻౘሺ౮ሻ ρౘሺ౬ሻሺ౮ሻ
ଶ౦ሺ౬ሻሺ౮ሻ ௧ ౭ሺ౮ሻ ρ౦ሺ౬ሻሺ౮ሻ
Mass transfer coefficient (m/s) of solute (Wankat, 1990) Dimethylphenol (Srinivasan et al., 2011) Chlorophenol (Sundaramoorthy et al., 2011) Dimensionless solute concentration and Reynolds number at feed and permeate channels. ɏ୵ ǣ The molal density of water (55.56 kmol/m3)
8 9 10
11, 12, 13
ె౭ሺబሻ
୮ሺ୧ሻሺሻ ൌ
౩ሺሻ େౘሺሻሺబሻ ୣౡሺሻሺబሻ ె౭ሺబሻ
౭ሺబሻ ା౩ሺሻ ୣౡሺሻሺబሻ
୮ሺ୧ሻሺሻ ൌ
౩ሺሻ େౘሺǡైሻ ౭ሺైሻ ା౩ሺሻ
ቀେ౭ሺሻሺ౮ሻ ିେ౦ሺሻሺ౬ሻቁ ቀେౘሺሻሺ౮ሻିେ౦ሺሻሺ౬ሻ ቁ
ె౭ሺైሻ ୣౡሺሻሺైሻ
ె౭ሺైሻ ୣౡሺሻሺైሻ
ൌ ൬
୮ሺ୧ሻሺୟ୴ሻ ൌ
౭ሺ౮ሻ
୩ሺሻሺ౮ሻ
େ౦ሺሻሺబሻ ାେ౦ሺሻሺైሻ
౦ሺైሻ ౘሺబሻ
ͳͲͲ
ሺ୧ሻ ൌ
൰
େౘሺሻሺైሻିେ౦ሺሻሺ౬ሻ େౘሺሻሺైሻ
14, 15
Wall solute concentration (kmol/m³) and solute flux (kmol/m² s)
16, 17
Recovery and rejection
18, 19
ଶ
ୱሺ୧ሻሺ୶ሻ ൌ ୱሺ୧ሻ ሺ୵ሺ୧ሻሺ୶ሻ െ ୮ሺ୧ሻሺୟ୴ሻ ሻ ൌ
Permeate solute concentration (kmol/m³) at x=0 and x=L and average one
ͳͲͲ
Table 2. Membrane specifications and geometry (Ion Exchange, India) Parameter Value Membrane material TFC Polyamide Feed and permeate spacers thickness ( ǡ ୮ ) 0.8E-3 m, 0.5E-3 m 0.934 m, 8.4 m Module length and width (ǡ ) Water Permeability Constant ሺ୵ ሻ and friction ୫ ୟ୲୫ୱ 9.5188E-7ቀ ቁ, 8529.45ቀ ర ቁ ୟ୲୫ୱ ୫ factor ሺሻ
M. A. Al-Obaidi et al.
1870
Table 3. Physical and transport parameters of the eight selected organic compounds Compound Bs (m/s) Membrane Reference Ion Exchange, India
Chlorophenol
8.4680E-8
Ion Exchange, India
Phenol Methyl orange dye Aniline Ammonium Cyanide Sulphate
6.5367E-7 3.2000E-9 4.1900E-6 1.1696E-7 2.1861E-6 3.9869E-8
Permionics, India FilmTec SW30 DESAL-3B SEPA-SSIC DESAL-3 SEPA-SSIC
12 10 8 6 4 2 0
Total osmotic pressure
0
0.2
Srinivasan et al. (2011) Sundaramoorthy et al. (2011) Srinivasan et al. (2010) Al-Bastaki (2004) Hidalgo et al. (2014) Bódalo et al. (2005) Bódalo et al. (2004) Bódalo et al. (2003)
4.60E-06 4.40E-06 4.20E-06 4.00E-06 3.80E-06 3.60E-06 3.40E-06
Feed pressure, Pb
0.4 0.6 0.8 Membrane Length, L (m)
Water Flux, Jw (m/s)
1.5876E-8
Pressure, Pb (atm)
Dimethylphenol
1
12.5
Dimethylphenol
12
Chlrophenol
11.5 11
325
375
425
475
Feed Concentration, Cb(0) (ppm)
525
%Rec
%Rej(i)
Fig. 2. The variation of operating parameters along the membrane length 100 90 80 70 60 50 40 30
Phenol Methyl orange dye Aniline
10.5
Ammonium
10
Cyanide Sulphate
Fig. 3. Effect of compound concentration on rejection and recovery rate (operating conditions: 2.583E-4 m³/s, 10 atm, 30 °C) Fig. 4 shows the impact of temperature on the removal of all compounds and permeate recovery. Increasing temperature causes more flexibility of membrane chains resulting in increasing convective transport by elevating the water flux. Also, it is noticed that diffusion transport increases due to increase in temperature, which is accompanied by a continuous reduction of average density and viscosity of the mixture. The effect of inlet feed pressure is shown in Fig. 5. Increasing the pressure from 10 to 20 atm causes an increase in the rejection due to increase in water flux, which dilutes the permeate. However, it seems that there is an optimum pressure which would maximise rejection of some of compounds. The impact of feed flow rate is shown in Fig. 6. The increase of feed flow rate from 2E-4 to 2.583E-4 m³/s causes a little increase in the rejection parameter but a remarkable decrease in permeate recovery. Increasing the feed flow rate causes a reduction in the osmotic pressure as a result of decreasing the membrane wall concentration. Increasing the feed flow rate from 2.583E-4 to 3E-4 m³/s causes a little
Simultaneous Removal of Organic Compounds from Wastewater
1871
decrease in rejection and steady decrease in the permeate recovery. The reduction of permeate flux is due to reduction of wastewater residence time inside the module. Dimethylphenol
100 90 80 70 60 50 40 30
15
Chlrophenol
%Rej(i)
14
30
32
34
36
38
40
Phenol
13
Methyl orange dye Aniline
12.5
Ammonium
12
Cyanide
13.5
28
%Rec
14.5
42
Sulphate
Operating Temperature,Tb (°C)
Fig. 4. Effect of operating temperature on rejection and recovery rate (operating conditions: 2.583E-4 m³/s, 10 atm, 350 ppm) 28
Dimethylphenol
25
Chlrophenol
22
Phenol
19 16
9
11
13
15
17
19
%Rec
%Rej(i)
100 90 80 70 60 50 40 30
Methyl orange dye Aniline
13
Ammonium
10
Cyanide
21
Sulphate
Feed Pressure, Pb(0) (atm)
Fig. 5. Operating pressure verses rejection and recovery rate (operating conditions: 2.583E-4 m³/s, 30 °C, 350 ppm) 16
Dimethylphenol
15
Chlrophenol
14
Phenol
13
Methyl orange dye Aniline
12 11 2.20E-04
2.50E-04
2.80E-04
Feed Flow rate, Fb(0) (m³/s)
10 3.10E-04
%Rec
%Rej(i)
100 90 80 70 60 50 40 30 1.90E-04
Ammonium Cyanide Sulphate
Fig. 6. Effect of operating feed flow rate on rejection and recovery rate (operating conditions: 10 atm, 30 °C, 350 ppm)
M. A. Al-Obaidi et al.
1872
4. Process optimisation The multi-objective function is presented below: Max ୠሺሻ ǡ ୠሺሻ ǡ ୠ ሺ୧ሻ , Subject to: Equality constraints - Process Model f(x, u, v) = 0 Inequality constraints: Inlet pressure ͷ ୠሺሻ ʹͲ Inlet feed flow rate Operating temperature
୫య
ͳ െ Ͷ ୠሺሻ ͳ െ ͵ ୱ ʹͲι ୠ ͶͲι
୫య ୱ
The optimal values of feed flow rate, pressure, and temperature are 7.4515E-4 m³/s, 20 atm and 40 °C respectively. The maximum permeate recovery is found to be 13.54%, while the maximum rejection of the compounds are: = < 99.269, 94.922, 86.477, 99.923, 51.430, 97.253, 66.252, 99.045>. The optimisation approach leads to an increased rejection of all the selected components compared to what have been presented in Figs. 3- 6. However, the low value of permeate recovery is due to the impact of osmotic pressure of eight compounds.
5. Conclusions A new one-dimensional model is developed for a spiral-wound RO process which is used to evaluate the impact of operating conditions on the rejection of several organic and nonorganic compounds from wastewater and permeate recovery. The simulation results confirmed the importance of feed pressure and temperature to drive high performance of RO process. Finally, the multi-objective optimisation problem finds the maximum values of the rejection of all the compounds and permeate recovery.
References N. Al-Bastaki, 2004, Removal of methyl orange dye and Na2SO4 salt from synthetic waste water using reverse osmosis, Chemical Engineering and Processing, 43, 1561-1567. M.A. Al-Obaidi, J-P. Li, C. Kara-Zaïtri, I.M. Mujtaba, 2017, Optimisation of reverse osmosis based wastewater treatment system for the removal of chlorophenol using genetic algorithms, Chemical Engineering Journal, 316, 91-100. A. Bódalo, J.L. Gómez, E. Gómez, G., León, M., Tejera, 2003, Reduction of sulphate content in aqueous solutions by reverse osmosis using celloulose acetate membranes, Desalination, 162, 55-60. A. Bódalo, J.L. Gómez, E. Gómez, G. León, M. Tejera, 2004, Sulfonated polyethersulfone membrane in the desalination of aqueous solutions, Desalination, 168, 277-282. A. Bódalo, J.L. Gómez, E. Gómez, G. León, M. Tejera, 2005, Ammonium removal from aqueous solutions by reverse osmosis using cellulose acetate membranes, Desalination, 184, 149-155. A.M. Hidalgo, G. León, M. Gómez, M.D. Murcia, M.D. Bernal, S. Ortega, 2014, Polyamide nenofiltration membranes to remove aniline in aqueous solutions, Environmental Technology, 35(9), 1175-1181. G. Srinivasan, S. Sundaramoorthy, D.V.R. Murthy, 2010, Spiral wound reverse osmosis membranes for the recovery of phenol compounds – Experimental and parameter estimation studies, American Journal of Engineering and Applied Science, 3(1), 31-36. G. Srinivasan, S. Sundaramoorthy, D.V.R. Murthy, 2011, Validation of an analytical model for spiral wound reverse osmosis membrane module using experimental data on the removal of dimethylphenol, Desalination, 281, 199-208. S. Sundaramoorthy, G. Srinivasan, D.V.R. Murthy, 2011, An analytical model for spiral wound reverse osmosis membrane modules: Part II — Experimental validation, Desalination, 277(1–3), 257-264. P.C. Wankat, 1990, Rate-Controlled Separation. Ist. Edn., Springer, pp: 873.
Simultaneous Removal of Organic Compounds from Wastewater Using Reverse Osmosis Process: Modelling, Simulation, and Optimisation Mudhar A. Al-Obaidia,b, Chakib Kara-Zaïtria and Iqbal M. Mujtabaa* a
Chemical Engineering, School of Engineering, Faculty of Engineering and Informatics. University of Bradford. Bradford, West Yorkshire BD7 1DP, UK b Middle Technical University, Iraq – Baghdad [email protected]
Abstract The modern industrialized world is generating a huge amount of wastewater containing a variety of micro-pollutants which is discharged into rivers and oceans leading to disruption of the biological ecosystem. The reverse osmosis (RO) process is one of the most promising technologies to produce high-quality recycling water at a reasonable cost. In this work we develop a new mathematical model to predict the performance of the RO process for the removal of several organic and non-organic compounds from wastewater simultaneously. The realistic operating conditions ensuring high rejection of multicompounds is explored via the simulation of the RO process first. The model is then used for optimising the operating conditions of the process while maximising the rejection and permeate recovery. Keywords: RO Process, Wastewater Treatment; Modelling; Simulation; Optimisation; Multi-compound rejection.
1. Introduction The main concern of any industrial wastewater is the existence of several organic and non-organic compounds which are harmful to human beings and marine life. Despite the availability of several wastewater treatment methods such as adsorption, biological, oxidation, and electrochemical process; the Reverse Osmosis (RO) process has demonstrated its competitive performance at a reasonable cost for the removal of these compounds. Several previous studies modelled the spiral wound RO process considering the removal of a single organic contaminant from wastewater (Sundaramoorthy et al., 2011; Al-Obaidi et al., 2017). However, only a few attempted the modelling of the spiral wound RO process for the removal of several organic and non-organic compounds from waste water. Al-Bastaki (2004) developed a lumped model to study the performance of a spiral wound RO process for removing Na2SO4 and methyl orange dye from wastewater. Note however, the development of a distributed model for the spiral wound RO process for the removal of multi-compounds simultaneously from wastewater has not been considered yet and is the main focus of this work. Also note, in the absence of experimental data in the literature for multi-compounds removal from wastewater, the model developed in this work has been validated against experimental data available for the removal of a single compound (chlorophenol) from wastewater. The model is then used to examine the rejection of organic and non-organic compounds under various operating parameters which in turn evaluates the process performance. This is followed
M. A. Al-Obaidi et al.
1868
by embedding the model in a multi-objective optimisation framework to simultaneously maximise the rejection and the total permeate recovery.
2. RO Process Model In this work RO filtration system of Sundaramoorthy et al. (2011) is used to investigate the simultaneous removal of multi-compounds from wastewater. Table 1 shows the new one-dimensional model equations for the process based on solution-diffusion model and the film theory. See Al-Obaidi et al. (2017) for a similar model but for single compound rejection. Due to the multi-compounds nature of the wastewater, the contribution of all of them to the osmotic pressure was considered in the new model. The model is validated using experimental data of chlorophenol removal from wastewater (Sundaramoorthy et al., 2011) which shows a very good agreement in terms of rejection and recovery of water (Fig. 1). The process model shown in Table 1 can be written in the compact form: f(x, u, v) = 0, where x is the set of all algebraic variables, u is the set of decision variables and v denotes the constant parameters of the process. The physical properties of seawater were used to calculate the diffusivity (ܦ ), viscosity (ρୠ ǡ ρ୮ ) and density (ɏୠ ǡ ɏ୮ ) due to low concentration of wastewater. The characteristics of the membranes used are given in Table 2. Five organic compounds and three inorganic species are assumed to be in the wastewater. The solute transport parameters (ୱ ) of the selected compounds were gathered from the literature and given in Table 3.
3. Process simulation: Effect of operating parameters Fig. 2 shows the variation of feed pressure, osmotic pressure, and water flux along the membrane length (x-axis). The feed pressure decreases along the x-axis due to pressure drop caused by the friction. This in turn reduces the water flux as a result of decreasing diving force. An increase of total osmotic pressure along the x-axis is noticed due to increasing accumulated concentration of the solutes at the membrane wall. The variation of inlet feed concentration of the pollutants found in wastewater is expected. Therefore, the RO process performance is investigated here using a range (350 to 500 ppm) of each pollutant concentration carried out at fixed feed flow rate, pressure, and temperature. Fig. 3 shows the effect of increasing feed concentration of each component on the rejection and total permeate recovery. In the selected range of 350 – 500 ppm of solute feed concentration, there was no considerable effect on the rejection. However, increasing the feed concentration of all the components causes a continuous reduction in permeate recovery due to increased osmotic pressure. 0.89 R² = 0.9493
0.83
Rec (The.)
Rej (The.)
0.86
0.8 0.77 0.74 0.72 0.74 0.76 0.78 0.8 0.82 0.84 Rej (Exp.)
29 25 21 17 13 9 5
R² = 0.9879
8
11
14
Fig. 1. The model validation results
17
20
23
Rec (Exp.)
26
29
Simultaneous Removal of Organic Compounds from Wastewater
1869
Table 1. Model equations of a spiral wound RO for multi-compounds wastewater Model Equations Specifications
no.
୵ሺ୶ሻ ൌ ୵ ൜οୠሺ୶ሻ െ ౭ሺ౮ሻ େ౦ሺሻሺ౮ሻ
ʹ͵Ǥͳͷሻ൬
౩ሺሻ
σ୬୧ ሺሺୠ
൰൨ൠ ǡ are the solute under
consideration and total number of solutes ୢౘሺ౮ሻ ୢ୶ ୢౘሺ౮ሻ ୢ୶
ൌ െ ୵ሺ୶ሻ ୠሺ୶ሻ ൌ ୢ౦ሺ౮ሻ
ൌെ
୲
ୢ୶
ቀ౭ሺ౮ሻ େ౦ሺሻሺ౮ሻቁ ୲
ౘሺ౮ሻ ୢେౘሺሻሺ౮ሻ ୲
ୢౘሺ౮ሻ
୲
ୢ୶
ൌ െ ୠሺ୶ሻ
οୠሺ୶ሻ ൌ ୠሺ୶ሻ െ ୮
ୢ୶
େౘሺሻሺ౮ሻ ୢౘሺ౮ሻ
ౘሺ౮ሻ
ୢ୶
ୢ
ൌ
ୢ୶
ቂୠሺ୧ሻሺ୶ሻ
ୢେౘሺሻሺ౮ሻ ୢ୶
ቃെ
ቀ౭ሺ౮ሻ େౘሺሻሺ౮ሻቁ ౘሺ౮ሻ ୈమ ౘሺሻሺ౮ሻ ୲
൰Ǥଷଷଷ
Ǥ଼ଷ Ǥଵଶଽ ሺ୧ሻሺ୶ሻ ሺʹݐ ሻ ൌ ʹͶǤͻୠሺ୧ሻሺ୶ሻ Ǥଵଵ ሺ୶ሻ ୮ሺ୶ሻ ୫ሺ୧ሻሺ୶ሻ Ǥଷଽ Ǥଵଷହ ሺ୧ሻሺ୶ሻ ሺʹ ሻ ൌ ͳͶǤͶୠሺ୧ሻሺ୶ሻ Ǥଵଷ ሺ୶ሻ ୮ሺ୶ሻ ୫ሺ୧ሻሺ୶ሻ
୮ሺ୶ሻ ൌ
Feed solute concentration (kmol/m³)
1
2,3, 4,5
6,7
୲
ሺ୧ሻሺ୶ሻ ൌ ͳǤͳ ൬
୫ሺ୧ሻሺ୶ሻ ൌ
Permeate flux (m/s) at any point along the xaxis. The second term is total osmotic pressure (atm) Feed flow rate (m³/s), velocity (m/s), pressure (atm) and permeate flow rate (m³/s) and operating pressure (atm)
େౘሺሻሺ౮ሻ
ሺ୶ሻ ൌ
౭
ଶౘሺ౬ሻሺ౮ሻౘሺ౮ሻ ρౘሺ౬ሻሺ౮ሻ
ଶ౦ሺ౬ሻሺ౮ሻ ௧ ౭ሺ౮ሻ ρ౦ሺ౬ሻሺ౮ሻ
Mass transfer coefficient (m/s) of solute (Wankat, 1990) Dimethylphenol (Srinivasan et al., 2011) Chlorophenol (Sundaramoorthy et al., 2011) Dimensionless solute concentration and Reynolds number at feed and permeate channels. ɏ୵ ǣ The molal density of water (55.56 kmol/m3)
8 9 10
11, 12, 13
ె౭ሺబሻ
୮ሺ୧ሻሺሻ ൌ
౩ሺሻ େౘሺሻሺబሻ ୣౡሺሻሺబሻ ె౭ሺబሻ
౭ሺబሻ ା౩ሺሻ ୣౡሺሻሺబሻ
୮ሺ୧ሻሺሻ ൌ
౩ሺሻ େౘሺǡైሻ ౭ሺైሻ ା౩ሺሻ
ቀେ౭ሺሻሺ౮ሻ ିେ౦ሺሻሺ౬ሻቁ ቀେౘሺሻሺ౮ሻିେ౦ሺሻሺ౬ሻ ቁ
ె౭ሺైሻ ୣౡሺሻሺైሻ
ె౭ሺైሻ ୣౡሺሻሺైሻ
ൌ ൬
୮ሺ୧ሻሺୟ୴ሻ ൌ
౭ሺ౮ሻ
୩ሺሻሺ౮ሻ
େ౦ሺሻሺబሻ ାେ౦ሺሻሺైሻ
౦ሺైሻ ౘሺబሻ
ͳͲͲ
ሺ୧ሻ ൌ
൰
େౘሺሻሺైሻିେ౦ሺሻሺ౬ሻ େౘሺሻሺైሻ
14, 15
Wall solute concentration (kmol/m³) and solute flux (kmol/m² s)
16, 17
Recovery and rejection
18, 19
ଶ
ୱሺ୧ሻሺ୶ሻ ൌ ୱሺ୧ሻ ሺ୵ሺ୧ሻሺ୶ሻ െ ୮ሺ୧ሻሺୟ୴ሻ ሻ ൌ
Permeate solute concentration (kmol/m³) at x=0 and x=L and average one
ͳͲͲ
Table 2. Membrane specifications and geometry (Ion Exchange, India) Parameter Value Membrane material TFC Polyamide Feed and permeate spacers thickness ( ǡ ୮ ) 0.8E-3 m, 0.5E-3 m 0.934 m, 8.4 m Module length and width (ǡ ) Water Permeability Constant ሺ୵ ሻ and friction ୫ ୟ୲୫ୱ 9.5188E-7ቀ ቁ, 8529.45ቀ ర ቁ ୟ୲୫ୱ ୫ factor ሺሻ
M. A. Al-Obaidi et al.
1870
Table 3. Physical and transport parameters of the eight selected organic compounds Compound Bs (m/s) Membrane Reference Ion Exchange, India
Chlorophenol
8.4680E-8
Ion Exchange, India
Phenol Methyl orange dye Aniline Ammonium Cyanide Sulphate
6.5367E-7 3.2000E-9 4.1900E-6 1.1696E-7 2.1861E-6 3.9869E-8
Permionics, India FilmTec SW30 DESAL-3B SEPA-SSIC DESAL-3 SEPA-SSIC
12 10 8 6 4 2 0
Total osmotic pressure
0
0.2
Srinivasan et al. (2011) Sundaramoorthy et al. (2011) Srinivasan et al. (2010) Al-Bastaki (2004) Hidalgo et al. (2014) Bódalo et al. (2005) Bódalo et al. (2004) Bódalo et al. (2003)
4.60E-06 4.40E-06 4.20E-06 4.00E-06 3.80E-06 3.60E-06 3.40E-06
Feed pressure, Pb
0.4 0.6 0.8 Membrane Length, L (m)
Water Flux, Jw (m/s)
1.5876E-8
Pressure, Pb (atm)
Dimethylphenol
1
12.5
Dimethylphenol
12
Chlrophenol
11.5 11
325
375
425
475
Feed Concentration, Cb(0) (ppm)
525
%Rec
%Rej(i)
Fig. 2. The variation of operating parameters along the membrane length 100 90 80 70 60 50 40 30
Phenol Methyl orange dye Aniline
10.5
Ammonium
10
Cyanide Sulphate
Fig. 3. Effect of compound concentration on rejection and recovery rate (operating conditions: 2.583E-4 m³/s, 10 atm, 30 °C) Fig. 4 shows the impact of temperature on the removal of all compounds and permeate recovery. Increasing temperature causes more flexibility of membrane chains resulting in increasing convective transport by elevating the water flux. Also, it is noticed that diffusion transport increases due to increase in temperature, which is accompanied by a continuous reduction of average density and viscosity of the mixture. The effect of inlet feed pressure is shown in Fig. 5. Increasing the pressure from 10 to 20 atm causes an increase in the rejection due to increase in water flux, which dilutes the permeate. However, it seems that there is an optimum pressure which would maximise rejection of some of compounds. The impact of feed flow rate is shown in Fig. 6. The increase of feed flow rate from 2E-4 to 2.583E-4 m³/s causes a little increase in the rejection parameter but a remarkable decrease in permeate recovery. Increasing the feed flow rate causes a reduction in the osmotic pressure as a result of decreasing the membrane wall concentration. Increasing the feed flow rate from 2.583E-4 to 3E-4 m³/s causes a little
Simultaneous Removal of Organic Compounds from Wastewater
1871
decrease in rejection and steady decrease in the permeate recovery. The reduction of permeate flux is due to reduction of wastewater residence time inside the module. Dimethylphenol
100 90 80 70 60 50 40 30
15
Chlrophenol
%Rej(i)
14
30
32
34
36
38
40
Phenol
13
Methyl orange dye Aniline
12.5
Ammonium
12
Cyanide
13.5
28
%Rec
14.5
42
Sulphate
Operating Temperature,Tb (°C)
Fig. 4. Effect of operating temperature on rejection and recovery rate (operating conditions: 2.583E-4 m³/s, 10 atm, 350 ppm) 28
Dimethylphenol
25
Chlrophenol
22
Phenol
19 16
9
11
13
15
17
19
%Rec
%Rej(i)
100 90 80 70 60 50 40 30
Methyl orange dye Aniline
13
Ammonium
10
Cyanide
21
Sulphate
Feed Pressure, Pb(0) (atm)
Fig. 5. Operating pressure verses rejection and recovery rate (operating conditions: 2.583E-4 m³/s, 30 °C, 350 ppm) 16
Dimethylphenol
15
Chlrophenol
14
Phenol
13
Methyl orange dye Aniline
12 11 2.20E-04
2.50E-04
2.80E-04
Feed Flow rate, Fb(0) (m³/s)
10 3.10E-04
%Rec
%Rej(i)
100 90 80 70 60 50 40 30 1.90E-04
Ammonium Cyanide Sulphate
Fig. 6. Effect of operating feed flow rate on rejection and recovery rate (operating conditions: 10 atm, 30 °C, 350 ppm)
M. A. Al-Obaidi et al.
1872
4. Process optimisation The multi-objective function is presented below: Max ୠሺሻ ǡ ୠሺሻ ǡ ୠ ሺ୧ሻ , Subject to: Equality constraints - Process Model f(x, u, v) = 0 Inequality constraints: Inlet pressure ͷ ୠሺሻ ʹͲ Inlet feed flow rate Operating temperature
୫య
ͳ െ Ͷ ୠሺሻ ͳ െ ͵ ୱ ʹͲι ୠ ͶͲι
୫య ୱ
The optimal values of feed flow rate, pressure, and temperature are 7.4515E-4 m³/s, 20 atm and 40 °C respectively. The maximum permeate recovery is found to be 13.54%, while the maximum rejection of the compounds are:
5. Conclusions A new one-dimensional model is developed for a spiral-wound RO process which is used to evaluate the impact of operating conditions on the rejection of several organic and nonorganic compounds from wastewater and permeate recovery. The simulation results confirmed the importance of feed pressure and temperature to drive high performance of RO process. Finally, the multi-objective optimisation problem finds the maximum values of the rejection of all the compounds and permeate recovery.
References N. Al-Bastaki, 2004, Removal of methyl orange dye and Na2SO4 salt from synthetic waste water using reverse osmosis, Chemical Engineering and Processing, 43, 1561-1567. M.A. Al-Obaidi, J-P. Li, C. Kara-Zaïtri, I.M. Mujtaba, 2017, Optimisation of reverse osmosis based wastewater treatment system for the removal of chlorophenol using genetic algorithms, Chemical Engineering Journal, 316, 91-100. A. Bódalo, J.L. Gómez, E. Gómez, G., León, M., Tejera, 2003, Reduction of sulphate content in aqueous solutions by reverse osmosis using celloulose acetate membranes, Desalination, 162, 55-60. A. Bódalo, J.L. Gómez, E. Gómez, G. León, M. Tejera, 2004, Sulfonated polyethersulfone membrane in the desalination of aqueous solutions, Desalination, 168, 277-282. A. Bódalo, J.L. Gómez, E. Gómez, G. León, M. Tejera, 2005, Ammonium removal from aqueous solutions by reverse osmosis using cellulose acetate membranes, Desalination, 184, 149-155. A.M. Hidalgo, G. León, M. Gómez, M.D. Murcia, M.D. Bernal, S. Ortega, 2014, Polyamide nenofiltration membranes to remove aniline in aqueous solutions, Environmental Technology, 35(9), 1175-1181. G. Srinivasan, S. Sundaramoorthy, D.V.R. Murthy, 2010, Spiral wound reverse osmosis membranes for the recovery of phenol compounds – Experimental and parameter estimation studies, American Journal of Engineering and Applied Science, 3(1), 31-36. G. Srinivasan, S. Sundaramoorthy, D.V.R. Murthy, 2011, Validation of an analytical model for spiral wound reverse osmosis membrane module using experimental data on the removal of dimethylphenol, Desalination, 281, 199-208. S. Sundaramoorthy, G. Srinivasan, D.V.R. Murthy, 2011, An analytical model for spiral wound reverse osmosis membrane modules: Part II — Experimental validation, Desalination, 277(1–3), 257-264. P.C. Wankat, 1990, Rate-Controlled Separation. Ist. Edn., Springer, pp: 873.