- Email: [email protected]
pressure retarded osmosis process for hypersaline solutions and fracking wastewater treatment
Desalination 350 (2014) 79–85
Contents lists available at ScienceDirect
Desalination journal homepage: www.elsevier.com/locate/desal
Dual-stage forward osmosis/pressure retarded osmosis process for hypersaline solutions and fracking wastewater treatment Ali Altaee a,1, Nidal Hilal b,⁎ a b
Faculty of Engineering and Physical Sciences, University of West of Scotland, Paisley PA1 2BE, UK Centre for Water Advanced Technologies and Environmental Research (CWATER), College of Engineering, Swansea University, Swansea SA2 8PP, UK
H I G H L I G H T S • • • • •
Dual-stage FO/PRO is proposed for hypersaline water treatment and power generation. Two designs were suggested: PRO–FO and FO–PRO systems. PRO–FO system generates higher power than the FO–PRO system. Increasing draw solution flow rates increased the permeate flow rate and TDS. Treated hypersaline water is suitable for RO treatment or discharge to sea.
a r t i c l e
i n f o
Article history: Received 26 March 2014 Received in revised form 8 July 2014 Accepted 10 July 2014 Available online xxxx Keywords: Hypersaline solution Forward osmosis Pressure retarded osmosis Fracking wastewater
a b s t r a c t Hypersaline solution with high TDS is not suitable for direct treatment by the conventional membrane and thermal processes. The current study proposes a dual-stage FO/PRO process for hypersaline solution treatment and power generation. The treatment process reduces the concentration of saline wastewater and hence renders it suitable for disposal directly to sea or treatment by the conventional membrane and thermal processes. The draw and feed solutions in the FO process were the hypersaline solutions and wastewater effluent, respectively. Five concentrations were evaluated for the process treatment with different concentrations ranging from 53 g/L to 157 g/L. The performance of FO membrane was estimated using pre-developed computer software. The results showed that a higher power can be generated from the PRO-FO system than from the FO-PRO system without compromising the concentration of hypersaline solution after dilution. The study also showed that although increasing the flow rate of draw solution resulted in an increase in the permeate flow rate, it caused a reduction in the dilution of draw solution. On the other hand, the study showed a negligible improvement in the performance of FO membrane upon increasing the feed solution flow rate. Finally, the simulation results showed that the concentration of diluted draw solution was suitable for the conventional membrane and thermal treatments or discharge to seawater after the dual-stage FO membrane treatment. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Hypersaline solution is a wastewater generated from industrial activities such as oil and gas industries. It is characterized by the highconcentration of Total Dissolved Solids (TDS) which is normally more than seawater concentration of 35 g/L. Such wastewater is difficult to treat by the conventional water treatment technologies such as Reverse Osmosis (RO) and Bioreactors [1–5]. Shale gas industry is one of the activities which largely contribute to hypersaline wastewater generation. Typically, water and additives are injected at high pressure into the gas well so as to open fractures in the shale and develop a flow path for the * Corresponding author. E-mail address: [email protected] (A. Altaee). 1 Tel.: +44 7986517994.
http://dx.doi.org/10.1016/j.desal.2014.07.013 0011-9164/© 2014 Elsevier B.V. All rights reserved.
gas to escape. Despite the rapid growth in shale gas industries over the last few years, the capacity for treating and handling of the fracking wastewater has remained underdeveloped [1,2]. The characteristics and composition of fracking wastewater vary from place-to-place and time-to-time throughout the production cycle. Drilling water, for instance, contains rock cuttings which are carried back to the land surface while flowback water contains high concentrations of additives. The volume of the flowback has been reported to vary from 1500 m 3 to 4500 m3 per well per week, but decreases with time upon the completion of fracking operation [6]. In addition, there is a large volume of production water which is collected during the production life of the gas well. Practically, production water is retained in the gas well and exposed to the shale formation for long period of time [6]. Fracking wastewater, in general, contains large amounts of suspended solids, high salinity (TDS), fluid additives, and other naturally occurring metal ions [1,3,5]. TDS is of particular importance because of its
80
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
negative impact on the biological treatment, aquatic life, freshwater salinity and water composition. Flowback wastewater, therefore, has a particular importance because of its high TDS which makes it unaffected by the conventional processes for water and wastewater treatment. The typical flowback TDS varies from 5000 mg/L to 250,000 mg/L with an average salinity about 125,000 mg/L [2,8]. However, TDS over 300,000 mg/L was reported in some fracking water samples. The current options for hypersaline and fracking water management vary from a simple dilution treatment to a complicated evaporation and crystallization processes [7]. Some of the proposed technologies have a limited efficiency in hypersaline wastewater treatment [6,7]. For example, dilution is an inexpensive treatment option but it has a limited capacity for decreasing the salinity of wastewater [7]. The new regulations for wastewater discharge recommend that the TDS of effluent should not exceed 500 mg/L which renders the conventional dilution treatment insufficient. Evaporation and crystallization processes were also proposed for hypersaline wastewater treatment. The technology has the advantage of reducing the wastewater TDS and the treated water can be reused but it is very expensive (about 0.25 USD per gallon) and energy-intensive [8]. Reusing of hypersaline water is often performed to reduce the generation of wastewater as in the shale gas industry. For example, the flowback water from the shale gas industry can be reused so many times but it has to be treated when the salinity reaches 100,000 mg/L [1]. However, the disadvantage of flowback water reuse is the high level of contaminants which may plug the gas wells [7]. Natural evaporation is also performed in the United States for the treatment of flowback water, but the process is slow and only suitable for dry hot climate [7]. Regardless of the treatment technology, hypersaline wastewater
may require a pretreatment for the removal of sediments, suspended solids and hydrocarbon residues. In this paper, a dual-stage forward osmosis (FO)/pressure retarded osmosis (PRO) process for the treatment of hypersaline wastewater was investigated. The draw and feed solutions in the first FO stage are, respectively, the hypersaline wastewater and the wastewater effluent. Because of its high salinity, the TDS of hypersaline wastewater cannot sufficiently be reduced to the desirable level by the first stage of the FO/PRO treatment. Therefore, the TDS of the diluted hypersaline solution from the first FO/PRO stage would be relatively high and a second FO/PRO stage is required to reduce its salinity to the desirable level (Fig. 1). In the second stage, the diluted hypersaline solution enters an FO membrane for further dilution by wastewater effluent. After leaving the second stage of membrane treatment, the hypersaline solution is either treated by RO/thermal process or discharged to seawater. The dual-stage FO process is able not only to reduce the TDS of the hypersaline solution but also to generate a useful power from the osmotic pressure gradient across the FO membrane using the PRO process. The impact of the hypersaline solution TDS, draw and feed solution flow rates and the position of the turbine system is evaluated in the current work. Different salinities, between 53,000 mg/L and 157,000 mg/L, were evaluated for the dual stage FO/PRO treatment. These salinities are reported for the flowback water in the shale gas industry [8]. The performance of FO membrane was estimated using pre-developed FO software [9]. A hydraulic pressure of 10 bar was assumed to be on the draw solution side of the FO membrane when it is operated on the PRO mode for power generation. Finally, the Van't Hoff equation was applied to calculate the osmotic pressure of the draw solution [9,10].
Pressure exchanger
FO/PRO Pump
Flowback water pretreatment processes
Wastewater effluent
RO Turbine
Sea
FO/PRO
Wastewater effluent
Optional position to compare between the PRO-FO and the FO-PRO systems Fig. 1. Schematic diagram of the dual stage FO/PRO system for flowback water treatment.
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
membrane multiplied by its hydraulic pressure as in the following equation:
2. Methodology The schematic diagram of dual stage FO/PRO system for hypersaline water treatment is illustrated in Fig. 1. The hypersaline solution from flowback water is initially treated for the removal of any impurities which may damage the FO membrane such as colloidal particles and hydrocarbon substances. After the pretreatment, the hypersaline solution goes to the first stage of the FO/PRO for treatment using wastewater effluent as a feed solution. Inside the FO membrane, fresh water transports in the direction of the osmotic gradient or from the wastewater effluent to the hypersaline solution resulting in the dilution of hypersaline solution. However, the first stage of the FO/PRO is not able to sufficiently reduce the salinity of hypersaline solution to be treated by the conventional RO membrane or be discharged to the natural water streams. Therefore, a second FO/PRO stage is designed to further reduce the salinity of hypersaline solution from the first stage of the FO/PRO process. In the second FO/PRO stage, the diluted hypersaline solution is passed on one side of the FO membrane while the wastewater effluent is on the other side of the membrane. Freshwater transports across the FO membrane will further dilute the hypersaline solution to a level that can either be treated by the RO membranes, or discharged to the seawater. Part of the dual-stage FO system is designed to generate power from the osmotic pressure gradient across the membrane by the PRO process. Two scenarios were evaluated to optimize the position of turbine in the system. In scenario one, the position of the turbine is suggested to be after the first stage of the FO Treatment ; in another word PRO-FO system; while in scenario two it is the FO-PRO system in which the turbine is placed after the second stage of the FO membrane treatment (Fig. 1). In the PRO–FO design, the diluted hypersaline solution goes first to the turbine to generate power. Part of the pressurized hypersaline solution returns to the Pressure Exchanger (PX) to pressurize the hypersaline feed solution to the FO membrane (Fig. 1). After leaving the turbine, the diluted hypersaline solution goes to the second FO membrane system for further treatment. The FO-PRO design operates in a similar way to the PRO-FO design with the difference that the diluted hypersaline solution goes to the turbine after the second stage of the FO membrane treatment. Depending on the brine management system of the area, the diluted hypersaline solution from the second stage membrane treatment could either be discharged to the seawater or sent to an RO/thermal process for further treatment. The performance of the FO membrane was estimated using predeveloped software. An HTI forward osmosis CTA membrane with water permeability coefficient, Aw, of about 0.79 L/m2 h was evaluated in the current study [11]. The water and salt flux, Jw and Js respectively, are estimated from the following equations: J w ¼ Aw ðΔP−ΔπÞ
ð1Þ
J s ¼ B C f −in −C p :
ð2Þ
In Eq. (1), P is the feed pressure (bar) and π is the osmotic pressure on solution (bar). Where in Eq. (2), Cf-in is the concentration of feed solution (mg/L), B is the salt permeability coefficient (kg/m2 h), and Cp is the permeate concentration (mg/L). The B factor is estimated from the following equation: B¼
ð1−RÞ J w : R
81
ð3Þ
Membrane rejection, R, is a function of permeate to feed concentration; R = 1 − (Cp/Cf), which is for HTI CTA membrane about 98% and 99.5% for monovalent and divalent ions respectively. The power density (Pw − W/m2) was estimated from the water flux across the FO
P density ¼ J w P:
ð4Þ
It is important to differentiate between the power density which is the power generated per unit membrane area and the power generated in the turbine system. The latter is a function of the hydraulic pressure and the membrane flow rate: P w ¼ Q ds−out P
ð5Þ
where, Qds-out is the flow rate of the draw solution out of the FO membrane (Watt). It is assumed here that the TDS of wastewater effluent is b200 mg/L which is a typical salinity for the domestic wastewater effluent [12,13]. The TDS of hypersaline solution is assumed to be equivalent to that of the flowback water from Marcellus shale gas industry. It has been reported that the TDS and composition of flowback water vary from place to place and time to time (Table 1). As shown in Table 1, five samples with TDS values that vary from 53,272 mg/L to 157,930 mg/L were evaluated for the dual-stage FO/PRO treatment [8]. The concentrations of monovalent and divalent ions are different from one sample to another but in general the concentration of monovalent ions is higher than the divalent ones. 3. Results and discussions 3.1. PRO-FO versus FO-PRO Two scenarios were evaluated here to optimize the position of FO and PRO process in the dual-stage system. The first scenario is the PRO-FO system in which the turbine system positioned after the firststage of FO membrane while the second scenario is the FO-PRO system with the turbine system positioned after the second stage of the FO membrane (Fig. 1). In both cases, i.e. PRO-FO and FO-PRO, 10 bar hydraulic pressure was assumed to be applied on the draw solution side of the membrane in the PRO process. Samples 1 to 5 were used as the draw solution in the FO and the PRO process, salinity range between 50 g/L to 157 g/L (Table 1). The impact of FO and PRO process arrangement on the permeate flow rate, net driving pressure (NDP), and the concentration of diluted draw solution was evaluated at different draw solution to feed solution (Qds-in/Qf-in) flow rate ratios (Fig. 2). The simulation results showed that the permeate flow from the firststage of the FO membrane was higher in the FO-PRO system than in the PRO-FO system (Fig. 2a). Obviously this was due to the higher net driving pressure in the FO-PRO system as shown in Fig. 2b. In contrast, the permeate flow rate was lower in the FO-PRO system than in the PRO-FO system due to the higher NDP in the second stage of the PROFO than in the second stage of the FO-PRO mode (Fig. 2a and b). Fig. 2c shows the impact of the Qds-in/Qf-in flow rate ratio on the concentration of diluted draw solution leaving the FO membrane (Cds-out). In the first-stage of the FO membrane, the Cds-out was higher in the PROFO than in the FO-PRO. The lower permeate flow rate generated from the PRO-FO system explains this phenomenon (Fig. 2a). The draw solution was exposed to a lower dilution in the PRO-FO system than in the FO-PRO system. However, in the second-stage of the FO membrane there was insignificant difference in the Cds-out between the PRO-FO system and FO-PRO system (Fig. 2c). In other words, both designs were able to achieve the same dilution level at the end of the treatment process. In fact the concentration of draw solution after the second-stage of the FO membrane was slightly lower in the PRO-FO system. This observation was also noticed at 107,272 mg/L (sample 2) concentration draw solution (results are not shown here). Practically, the effect of concentration polarization increases with the increasing concentration of the draw solution [14]; i.e. higher in the first-stage than in the second-
82
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
Table 1 The concentration and composition of flowback water samples (ref. [8]). Sample TDS (mg/L)
Na (mg/L)
Mg (mg/L)
Ca (mg/L)
Cl (mg/L)
SO4 (mg/L)
HCO3 (mg/L)
Sample 1: 53,272 Sample 2: 107,761 Sample 3: 123,801 Sample 4: 145,239 Sample 5: 157,930
629 31,500 47,734 54 54,843
1770 725 899 4730 6062
15,680 7820 6800 15,200 3600
35,000 67,100 68,000 125,000 93,000
10 560 20 60 10
183 56 348 195 415
stage of the FO membrane. This probably explains the equal draw solution concentrations in both designs. The power density (Pdensity) and the gross power (Pw) were calculated in both systems; the PRO-FO and the FO-PRO (Fig. 3). The simulation results showed that the PRO-FO system is more efficient than the FOPRO system in terms of the generated power density and the gross power through the turbine system. It is also observed from Fig. 3 that the power density increased with the increasing draw solution flow
1200
a 1000
Qp (L/h)
800
rate. This was due to the higher permeate flow rate at higher draw solution flow rate (Fig. 2a). The simulation results show that the efficiency of the PRO-FO system is higher than that of the FO-PRO system. As shown in Fig. 3, the dual-stage membrane treatment generates higher power when it is operated on the PRO-FO mode than on the FO-PRO mode. For example in sample 1, the power density generated from the PROFO system was almost 2.65 times higher than that from the FO-PRO system. The higher power generated from the PRO-FO operation mode didn't affect the quality and concentration of the treated saline solution. In both cases, the PRO-FO and the FO-PRO, the draw solution from the second-stage of FO treatment has the same concentration. As a matter of fact, the treated draw solution was more diluted in the PRO-FO system than in the FO-PRO system. Therefore, for the rest of this study the PRO-FO system will only be considered for the hypersaline solution treatment because of its higher dilution efficiency and capability to generate higher energy than the FO-PRO system.
600
3.2. The impact of draw solution flow rates
400 PRO-FO: Qp1 PRO-FO: Qp2
200
FO-PRO: Qp1 FO-PRO: Qp2
0 0
0.2
0.4
0.6
0.8
1
1.2
Qds-in/Qf-in 85
b
75
NDP (bar)
65 55 45 35
PRO-FO-NDP 1 PRO-FO-NDP 2 FO-PRO-NDP 1 FO-PRO-NDP 2
25 15 0
0.2
0.4
0.6
0.8
1
1.2
The impact of the draw solution flow rate on the performance of FO membrane was investigated using Qds-in/Qf-in ratios that vary from 0.2 to 1. Fig. 4 shows the impact of the draw solution flow rate on the permeate flow rate. Increasing the draw solution flow rate resulted in an increase in the permeate flow rate of the first and the second-stage of the FO treatment. This was due to the lower concentration polarization effect at the FO membrane surface and to the higher bulk feed concentration which resulted in a higher osmotic pressure across the FO membrane [15,16]. However, the permeate flow rate was higher in the firststage than in the second-stage of the FO membrane treatment because of the higher osmotic pressure across the membrane (Fig. 4c and d). Freshwater transport across the FO membrane diluted the concentration of draw solution in the first-stage of the FO process and hence its osmotic pressure decreased (Fig. 4c–d). As a result, the net driving pressure was lower in the second-stage than in the first-stage of the FO membrane. The simulation results show that the differential osmotic pressure and the permeate flow rate increased with the increasing
90000
c
80000
10
16
9
14
8
Power (kW)
Cds-out (mg/L)
12 7
70000 60000 50000
10
5
8
4
6
3
PRO-FO-Qds-out -1 PRO-FO-Qds-out-2 FO-PRO-Qds-out-1 FO-PRO-Qds-out-2
40000
6
Power: PRO-FO
2
Power density: PRO-FO
1
30000
4
Power: FO-PRO
Power Density (W/m2)
Qds-in/Qf-in
2
Power density: FO-PRO
0
0.2
0.4
0.6
0.8
1
1.2
Qds-in/Qf-in
0 0
0.2
0.4
0.6
0.8
1
0 1.2
Qds-in/Qf-in Fig. 2. The impact of Qds-in/Qf-in flow rate ratio on the permeate flow, NDP and concentration of draw solution in the PRO-FO and FO-PRO operation modes (samples 1 to 5).
Fig. 3. Power and power density at different Qds-in/Qf-in flow rate ratios (samples 1 to 5).
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
83
900
1400 TDS-53 g/l TDS-107 g/l TDS-123 g/l TDS-145 g/l TDS-157 g/l
1200
a
TDS-53 g/l TDS-107 g/l TDS-123 g/l TDS-145 g/l TDS-157 g/l
800 700
b
1000
Qp-2 (L/h)
Qp-1 (L/h)
600 800 600
500 400 300
400
200 200
100 0
0 0
0.2
0.4
0.6
0.8
1
1.2
0
0.2
0.4
Qds-in/Qf-in
0.6
0.8
1
1.2
Qds-in/Qf-in 70
120
c
TDS- 53 g/l TDS- 107 g/l TDS- 123 g/l TDS- 145 g/l TDS- 157 g/l
100
TDS- 53 g/l TDS- 107 g/l TDS- 123 g/l TDS- 145 g/l TDS- 157 g/l
60
d
50
∏ -2 (bar)
∏ -1 (bar)
80
60
40 30
40
20 20
10
0
0 0
0.2
0.4
0.6
0.8
1
1.2
0
0.2
0.4
Qds-in/Qf-in
0.6
0.8
1
1.2
Qds-in/Qf-in
Fig. 4. The impact of draw solution flow rate on the permeate flow rate and the differential osmotic pressure in the FO membrane (testing condition PRO-FO; applied hydraulic pressure 10 bar).
80% TDS-53 g/l TDS-107 g/l TDS-123 g/l TDS-145 g/l TDS-157 g/l
70% 60%
Re-1 (%)
concentration of draw solution from sample 1 to sample 5 (Fig. 4). This was true for all draw solution concentrations apart from sample 4 in which the net driving pressure was lower than that in sample 3. This can be attributed to the higher osmotic pressure of hypersaline solution in sample 2 than in sample 4; 98.4 bar and 96 bar respectively. The ionic composition of the draw solution in sample 3 possesses higher osmotic pressure than that in sample 4. For the same reason, the permeate flow rate in sample 2 was higher than that in sample 4 draw solution concentration (Fig. 4a and b). In addition to the permeate flow rate and osmotic pressure driving pressure, the impact of draw solution flow rate on stage one and two recovery rates was evaluated (Fig. 5). With exception to sample 4, which exhibited an osmotic pressure lower than sample 2, the FO recovery rate increased with the increasing concentration of the draw solution from sample 1 to sample 5 (Fig. 4c-d). The higher osmotic pressure across the FO membrane resulted in a higher permeate flow rate and hence the recovery rate increased (Fig. 4a and b). The recovery rates of the FO membrane also increased with the increasing Qds-in/Qf-in ratio because of the higher draw solution bulk concentration and osmotic pressure [17].
50% 40% 30% 20% 10% 0% 0
0.4
0.6
0.8
1
1.2
90% TDS-53 g/l TDS-107 g/l TDS-123 g/l TDS-145 g/l TDS-157 g/l
80%
b
60%
Re-2 (%)
The impact of changing the feed solution flow rates on the performance of the FO membrane is illustrated in Fig. 6. A significant increase in the osmotic pressure gradients across the FO membrane occurred upon increasing the Qds-in/Qf-in ratio while increasing the Qf-in/Qds-in ratio didn't cause a noticeable change in the osmotic pressure gradient across the membrane (Fig. 6a). Practically, increasing the draw solution flow rate reduces the effect of concentration polarization at the membrane surface and increases the concentration of bulk solution. This wasn't the case when higher feed solution flow rates are applied because of the negligible osmotic pressure (TDS b 200 mg/L) of the feed solution. Fig. 6b shows the impact of feed and draw solution flow rates on the
0.2
Qds-in/Qf-in
70%
3.3. Impact of changing the feed solution flow rates
a
50% 40% 30% 20% 10% 0% 0
0.2
0.4
0.6
0.8
1
1.2
Qds-in/Qf-in Fig. 5. The impact of the draw solution flow rate on the FO membrane recovery rate (testing condition PRO-FO).
84
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
Qf-in/Qds-in 5
6
7
8
9
90
Δ∏ -Qds-in/Qf-in (bar)
90
a
80
80
70
70
60
60
50
50
40
40
30
30 Qds/Qf: ∆∏1 Qds/Qf: ∆∏2 Qf/Qds: ∆∏1 Qf/Qds: ∆∏2
20 10 0.2
0.4
0.6
0.8
20 10 0 1.2
0 0
of FO treatment was 35,374 mg/L and increased to 51,972 mg/L at Qds-in /Qf-in ratio equals to 0.2 (initial draw solution TDS is 123,801 mg/L, sample 2). The concentration limits of these diluted draw solutions may allow further treatment by RO membranes or discharge to seawater. However, when the Q ds-in /Qf-in ratio increased to 0.8 the concentration of diluted draw solution from the second-stage of the FO membrane was 57,516 mg/L; which is a challenging concentration for further treatment [18–22]. As such, higher draw solution dilution is achievable at lower Qds-in / Qf-in ratios. On the other hand, increasing the flow rate of feed solution has no impact on the performance of FO and hence it would only increase the power consumption instead of improving the process performance.
10
Δ∏ -Qf-in/Qds-in (bar)
4
1
Qds-in/Qf-in
4. Conclusions
Qf-in/Qds-in 4
5
6
7
8
9
10 1000
b
Qp-Qds-in/Qf-in (L/h)
900
900
800
800
700
700
600
600
500
500
400
400
300
300 Qds/Qf: Qp1 Qds/Qf: Qp2 Qf/Qds: Qp1 Qf/Qds: Qp2
200 100 0 0
0.2
0.4
0.6
Qp-Qf-in/Qds-in (L/h)
1000
200 100
0.8
1
0 1.2
8
9
10
Qds-in/Qf-in Qf-in /Q ds-in 4
5
6
7
90000
c
80000
80000
70000
70000
60000
60000
50000
50000
40000
40000
30000
30000
20000
Qds/Qf: Cds-out-1 Qds/Qf: Cds-out-2
20000
10000
Qf/Qds: Cds-out-1 Qf/Qds: Cds-out-2
10000
0
Cds-out-Qf-in/Qds-in (mg/L)
Cds-out-Qds-in/Qf-in (mg/L)
90000
0 0
0.2
0.4
0.6
0.8
1
1.2
Qds-in/Qf-in Fig. 6. Impact of feed and draw solution flow rates on the performance of FO membrane; a) impact on the osmotic pressure gradient b) impact on the permeate flow rate c) impact on the concentration of draw solution out (Cds-out).
permeate flow rate. No tangible change in the permeate flow rate was found upon increasing the flow rate of the feed solution while higher draw solution flow rates resulted in an increase in the permeate flow rate. Increasing the flow rate of draw solution not only increased the permeate flow rate and osmotic pressure, but also increased the concentration of draw solution from the treatment process (Fig. 6c). Actually, this parameter should be carefully calibrated because it affects the dilution of the hypersaline water which is the main objective of the membrane treatment. At low TDS concentration, the effect of increasing the flow rate of feed solution on the performance of FO membrane is negligible (Fig. 6). It is desirable, therefore, to have Qds-in/Qf-in ratios between low and average to achieve a sufficient dilution of the draw solution (saline water). For example, at Qds-in /Qf-in ratio equals to 2, the concentration of diluted draw solution from the second-stage
A dual-stage membrane process was investigated for the hypersaline solution treatment in which the hypersaline water and wastewater effluent were the feed and draw solution respectively. Two different systems were evaluated; i.e. PRO-FO and FO-RPO systems. The simulation results showed that the efficiency of PRO-FO design is higher than that of the FO-PRO design especially in terms of the power generation. The power generated from the PRO-FO system was 2.65 times higher than that from the FO-PRO system while the concentration of the diluted draw solution was the same in both designs. A higher draw solution dilution was achieved in the PRO-FO design at a Qds-in/Qf-in ratio equals to 0.2. Using higher Qds-in/Qf-in ratios resulted in a higher NDP across the membrane and hence increased the permeate flow rate. But it resulted in a higher draw solution concentration from the second-stage of FO membrane, which complicates the disposal and management of the diluted draw solution. The simulation results also showed that the increase of feed solution flow rate on the performance of FO membrane was negligible. Therefore, it is desirable not to use high feed flow rates as the power consumption will be increased. The complicated composition and the high concentration of hypersaline solution investigated in this study make it unsuitable for direct conventional thermal and membrane treatment [7,8]. The advantage of dual-stage FO membrane treatment is to bring the concentration of hypersaline solution low enough to be treated by the conventional processes or discharge properly to seawater. Experimental work would be required to investigate the process potential and feasibility. References [1] Penn State Extension, Marcellus Shale Wastewater Issues in Pennsylvania—Current and Emerging Treatment and Disposal Technologies, Penn State College of Agricultural Sciences, Cooperative Extension, 2011. [2] Gaudlip Tony, Water Use and Water Reuse/Recycling in Marcellus Shale Gas Exploration and Production, Penn State College of Agricultural Sciences, Cooperative Extension, Marcellus Shale Educational Webinar Series, 2010. [3] Paul T. Sun, Charles L. Meyer, Cor Kuijvenhoven, Sudini Padmasiri, Vladimir Fedotov, Treatment of water from fracturing operation for unconventional gas production, in: Ronald D. Neufeld (Ed.), Contemporary Technologies for Shale-Gas Water and Environmental Management, Water Environment Federation, Alexandria, Va, 2012, pp. 61–81. [4] Michael E. Tamblin, Pilot Project to Recycle and Treat Marcellus Shale Water, Clear Waters, winter 2010. 40–46. [5] O. Lefebvre, F. Habouzit, V. Bru, J.P. Delgenes, J.J. Godon, R. Moletta, Treatment of hypersaline industrial wastewater by microbial consortium in a sequencing batch reactor, Environ. Technol. 25 (2004) 543–553. [6] Adrienne Beckman, Archis Ambulkar, Art Umble, Diego Rosso, Joe Husband, Joseph Cleary, Julian Sandino, Mikel Goldblatt, Ron Horres, Ronald Neufeld, Russell Mau, Sam Jeyanayagam, Considerations for Accepting Fracking Wastewater at Water Resource Recovery Facilities, Water Environment Federation, 2014. (20/03/2014). [7] D. Yoxtheimer, Water use and Water Reuse/Recycling in Marcellus Shale Gas Exploration and Production, Penn State College of Agricultural Sciences, Cooperative Extension, Marcellus Shale Educational Webinar Series, 2010. [8] Tom Lewis, Fracking water disposal/recycling processes for unconventional shale gas wastewater, http://www.gwpc.org/sites/default/files/event-sessions/19r_ Lewis_Tom.pdf (25/03/2014). [9] Ali Altaee, Abdelnasser Mabrouk, Karim Bourouni, A novel forward osmosis membrane pretreatment of seawater for thermal desalination processes, Desalination V326 (2013) 19–29.
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85 [10] Ali Altaee, Forward Osmosis, Potential use in desalination and water reuse, J. Membr. Sep. Technol. 1 (2012) 79–93. [11] Gordon T. Gray, J.R. McCutcheon, M. Elimelech, Internal concentration polarization in forward osmosis: role of membrane orientation, Desalination 197 (2006) 1–8. [12] Maruf Mortula, Sina Shabani, Removal of TDS and BOD from synthetic industrial wastewater via adsorption, 2012 International Conference on Environmental, Biomedical and Biotechnology, vol. 41, IACSIT Press, Singapore, 2012. [13] Feed water type and analysis, tech manual excerpt, http://www.dow.com/ PublishedLiterature/dh_003b/0901b8038003b4a0.pdf?filepath=liquidseps/pdfs/ noreg/609-02010.pdf&fromPage=GetDoc (20/01/2013). [14] J.R. McCutcheon, M. Elimelech, Influence of concentrative and dilutive internal concentration polarization on flux behavior in forward osmosis, J. Membr. Sci. 284 (2006) 237–247. [15] Ata M. Hassan, Review of development of the new NF-seawater desalination process from pilot plant to commercial production plant stages, The 6th Saudi Engineering Conference, KFUPM, Dhahran, December 2002, 2002. [16] Ata Hassan, Fully integrated NF-thermal seawater desalination process and equipment, US Patents No 2006/0157410 A1, July 20, 2006.
85
[17] Ali Altaee Guillermo Zaragoza Adel Sharif, Pressure retarded osmosis for power generation and seawater desalination: performance analysis, Desalination 336 (2014) 50–57. [18] Khaled Walha, Raja Ben Amar, Francis Quemeneur, Pascal Jaouen, Treatment by nanofiltration and reverse osmosis of high salinity drilling water for seafood washing and processing, Desalination 219 (2008) 231–239. [19] Weiyi Li, William B. Krantz, Emile R. Cornelissen, Jan W. Post, Arne R.D. Verliefde, Chuyang Y. Tang, A novel hybrid process of reverse electrodialysis and reverse osmosis for low energy seawater desalination and brine management, Appl. Energy 104 (2013) 592–602. [20] Qiang Li, Mengdi Wang, Jun Feng, Wei Zhang, Yuanyuan Wang, Gu. Yanyan, Cunjiang Song, Shufang Wang, Treatment of high-salinity chemical wastewater by indigenous bacteria — bioaugmented contact oxidation, Bioresour. Technol. 144 (2013) 380–386. [21] Nikolay Voutchkov, Overview of seawater concentrate disposal alternatives, Desalination 273 (2011) 205–219. [22] D.W. Westcot, Reuse and disposal of higher salinity subsurface drainage water, Agric. Water Manag. 14 (1988) 483–511.
Contents lists available at ScienceDirect
Desalination journal homepage: www.elsevier.com/locate/desal
Dual-stage forward osmosis/pressure retarded osmosis process for hypersaline solutions and fracking wastewater treatment Ali Altaee a,1, Nidal Hilal b,⁎ a b
Faculty of Engineering and Physical Sciences, University of West of Scotland, Paisley PA1 2BE, UK Centre for Water Advanced Technologies and Environmental Research (CWATER), College of Engineering, Swansea University, Swansea SA2 8PP, UK
H I G H L I G H T S • • • • •
Dual-stage FO/PRO is proposed for hypersaline water treatment and power generation. Two designs were suggested: PRO–FO and FO–PRO systems. PRO–FO system generates higher power than the FO–PRO system. Increasing draw solution flow rates increased the permeate flow rate and TDS. Treated hypersaline water is suitable for RO treatment or discharge to sea.
a r t i c l e
i n f o
Article history: Received 26 March 2014 Received in revised form 8 July 2014 Accepted 10 July 2014 Available online xxxx Keywords: Hypersaline solution Forward osmosis Pressure retarded osmosis Fracking wastewater
a b s t r a c t Hypersaline solution with high TDS is not suitable for direct treatment by the conventional membrane and thermal processes. The current study proposes a dual-stage FO/PRO process for hypersaline solution treatment and power generation. The treatment process reduces the concentration of saline wastewater and hence renders it suitable for disposal directly to sea or treatment by the conventional membrane and thermal processes. The draw and feed solutions in the FO process were the hypersaline solutions and wastewater effluent, respectively. Five concentrations were evaluated for the process treatment with different concentrations ranging from 53 g/L to 157 g/L. The performance of FO membrane was estimated using pre-developed computer software. The results showed that a higher power can be generated from the PRO-FO system than from the FO-PRO system without compromising the concentration of hypersaline solution after dilution. The study also showed that although increasing the flow rate of draw solution resulted in an increase in the permeate flow rate, it caused a reduction in the dilution of draw solution. On the other hand, the study showed a negligible improvement in the performance of FO membrane upon increasing the feed solution flow rate. Finally, the simulation results showed that the concentration of diluted draw solution was suitable for the conventional membrane and thermal treatments or discharge to seawater after the dual-stage FO membrane treatment. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Hypersaline solution is a wastewater generated from industrial activities such as oil and gas industries. It is characterized by the highconcentration of Total Dissolved Solids (TDS) which is normally more than seawater concentration of 35 g/L. Such wastewater is difficult to treat by the conventional water treatment technologies such as Reverse Osmosis (RO) and Bioreactors [1–5]. Shale gas industry is one of the activities which largely contribute to hypersaline wastewater generation. Typically, water and additives are injected at high pressure into the gas well so as to open fractures in the shale and develop a flow path for the * Corresponding author. E-mail address: [email protected] (A. Altaee). 1 Tel.: +44 7986517994.
http://dx.doi.org/10.1016/j.desal.2014.07.013 0011-9164/© 2014 Elsevier B.V. All rights reserved.
gas to escape. Despite the rapid growth in shale gas industries over the last few years, the capacity for treating and handling of the fracking wastewater has remained underdeveloped [1,2]. The characteristics and composition of fracking wastewater vary from place-to-place and time-to-time throughout the production cycle. Drilling water, for instance, contains rock cuttings which are carried back to the land surface while flowback water contains high concentrations of additives. The volume of the flowback has been reported to vary from 1500 m 3 to 4500 m3 per well per week, but decreases with time upon the completion of fracking operation [6]. In addition, there is a large volume of production water which is collected during the production life of the gas well. Practically, production water is retained in the gas well and exposed to the shale formation for long period of time [6]. Fracking wastewater, in general, contains large amounts of suspended solids, high salinity (TDS), fluid additives, and other naturally occurring metal ions [1,3,5]. TDS is of particular importance because of its
80
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
negative impact on the biological treatment, aquatic life, freshwater salinity and water composition. Flowback wastewater, therefore, has a particular importance because of its high TDS which makes it unaffected by the conventional processes for water and wastewater treatment. The typical flowback TDS varies from 5000 mg/L to 250,000 mg/L with an average salinity about 125,000 mg/L [2,8]. However, TDS over 300,000 mg/L was reported in some fracking water samples. The current options for hypersaline and fracking water management vary from a simple dilution treatment to a complicated evaporation and crystallization processes [7]. Some of the proposed technologies have a limited efficiency in hypersaline wastewater treatment [6,7]. For example, dilution is an inexpensive treatment option but it has a limited capacity for decreasing the salinity of wastewater [7]. The new regulations for wastewater discharge recommend that the TDS of effluent should not exceed 500 mg/L which renders the conventional dilution treatment insufficient. Evaporation and crystallization processes were also proposed for hypersaline wastewater treatment. The technology has the advantage of reducing the wastewater TDS and the treated water can be reused but it is very expensive (about 0.25 USD per gallon) and energy-intensive [8]. Reusing of hypersaline water is often performed to reduce the generation of wastewater as in the shale gas industry. For example, the flowback water from the shale gas industry can be reused so many times but it has to be treated when the salinity reaches 100,000 mg/L [1]. However, the disadvantage of flowback water reuse is the high level of contaminants which may plug the gas wells [7]. Natural evaporation is also performed in the United States for the treatment of flowback water, but the process is slow and only suitable for dry hot climate [7]. Regardless of the treatment technology, hypersaline wastewater
may require a pretreatment for the removal of sediments, suspended solids and hydrocarbon residues. In this paper, a dual-stage forward osmosis (FO)/pressure retarded osmosis (PRO) process for the treatment of hypersaline wastewater was investigated. The draw and feed solutions in the first FO stage are, respectively, the hypersaline wastewater and the wastewater effluent. Because of its high salinity, the TDS of hypersaline wastewater cannot sufficiently be reduced to the desirable level by the first stage of the FO/PRO treatment. Therefore, the TDS of the diluted hypersaline solution from the first FO/PRO stage would be relatively high and a second FO/PRO stage is required to reduce its salinity to the desirable level (Fig. 1). In the second stage, the diluted hypersaline solution enters an FO membrane for further dilution by wastewater effluent. After leaving the second stage of membrane treatment, the hypersaline solution is either treated by RO/thermal process or discharged to seawater. The dual-stage FO process is able not only to reduce the TDS of the hypersaline solution but also to generate a useful power from the osmotic pressure gradient across the FO membrane using the PRO process. The impact of the hypersaline solution TDS, draw and feed solution flow rates and the position of the turbine system is evaluated in the current work. Different salinities, between 53,000 mg/L and 157,000 mg/L, were evaluated for the dual stage FO/PRO treatment. These salinities are reported for the flowback water in the shale gas industry [8]. The performance of FO membrane was estimated using pre-developed FO software [9]. A hydraulic pressure of 10 bar was assumed to be on the draw solution side of the FO membrane when it is operated on the PRO mode for power generation. Finally, the Van't Hoff equation was applied to calculate the osmotic pressure of the draw solution [9,10].
Pressure exchanger
FO/PRO Pump
Flowback water pretreatment processes
Wastewater effluent
RO Turbine
Sea
FO/PRO
Wastewater effluent
Optional position to compare between the PRO-FO and the FO-PRO systems Fig. 1. Schematic diagram of the dual stage FO/PRO system for flowback water treatment.
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
membrane multiplied by its hydraulic pressure as in the following equation:
2. Methodology The schematic diagram of dual stage FO/PRO system for hypersaline water treatment is illustrated in Fig. 1. The hypersaline solution from flowback water is initially treated for the removal of any impurities which may damage the FO membrane such as colloidal particles and hydrocarbon substances. After the pretreatment, the hypersaline solution goes to the first stage of the FO/PRO for treatment using wastewater effluent as a feed solution. Inside the FO membrane, fresh water transports in the direction of the osmotic gradient or from the wastewater effluent to the hypersaline solution resulting in the dilution of hypersaline solution. However, the first stage of the FO/PRO is not able to sufficiently reduce the salinity of hypersaline solution to be treated by the conventional RO membrane or be discharged to the natural water streams. Therefore, a second FO/PRO stage is designed to further reduce the salinity of hypersaline solution from the first stage of the FO/PRO process. In the second FO/PRO stage, the diluted hypersaline solution is passed on one side of the FO membrane while the wastewater effluent is on the other side of the membrane. Freshwater transports across the FO membrane will further dilute the hypersaline solution to a level that can either be treated by the RO membranes, or discharged to the seawater. Part of the dual-stage FO system is designed to generate power from the osmotic pressure gradient across the membrane by the PRO process. Two scenarios were evaluated to optimize the position of turbine in the system. In scenario one, the position of the turbine is suggested to be after the first stage of the FO Treatment ; in another word PRO-FO system; while in scenario two it is the FO-PRO system in which the turbine is placed after the second stage of the FO membrane treatment (Fig. 1). In the PRO–FO design, the diluted hypersaline solution goes first to the turbine to generate power. Part of the pressurized hypersaline solution returns to the Pressure Exchanger (PX) to pressurize the hypersaline feed solution to the FO membrane (Fig. 1). After leaving the turbine, the diluted hypersaline solution goes to the second FO membrane system for further treatment. The FO-PRO design operates in a similar way to the PRO-FO design with the difference that the diluted hypersaline solution goes to the turbine after the second stage of the FO membrane treatment. Depending on the brine management system of the area, the diluted hypersaline solution from the second stage membrane treatment could either be discharged to the seawater or sent to an RO/thermal process for further treatment. The performance of the FO membrane was estimated using predeveloped software. An HTI forward osmosis CTA membrane with water permeability coefficient, Aw, of about 0.79 L/m2 h was evaluated in the current study [11]. The water and salt flux, Jw and Js respectively, are estimated from the following equations: J w ¼ Aw ðΔP−ΔπÞ
ð1Þ
J s ¼ B C f −in −C p :
ð2Þ
In Eq. (1), P is the feed pressure (bar) and π is the osmotic pressure on solution (bar). Where in Eq. (2), Cf-in is the concentration of feed solution (mg/L), B is the salt permeability coefficient (kg/m2 h), and Cp is the permeate concentration (mg/L). The B factor is estimated from the following equation: B¼
ð1−RÞ J w : R
81
ð3Þ
Membrane rejection, R, is a function of permeate to feed concentration; R = 1 − (Cp/Cf), which is for HTI CTA membrane about 98% and 99.5% for monovalent and divalent ions respectively. The power density (Pw − W/m2) was estimated from the water flux across the FO
P density ¼ J w P:
ð4Þ
It is important to differentiate between the power density which is the power generated per unit membrane area and the power generated in the turbine system. The latter is a function of the hydraulic pressure and the membrane flow rate: P w ¼ Q ds−out P
ð5Þ
where, Qds-out is the flow rate of the draw solution out of the FO membrane (Watt). It is assumed here that the TDS of wastewater effluent is b200 mg/L which is a typical salinity for the domestic wastewater effluent [12,13]. The TDS of hypersaline solution is assumed to be equivalent to that of the flowback water from Marcellus shale gas industry. It has been reported that the TDS and composition of flowback water vary from place to place and time to time (Table 1). As shown in Table 1, five samples with TDS values that vary from 53,272 mg/L to 157,930 mg/L were evaluated for the dual-stage FO/PRO treatment [8]. The concentrations of monovalent and divalent ions are different from one sample to another but in general the concentration of monovalent ions is higher than the divalent ones. 3. Results and discussions 3.1. PRO-FO versus FO-PRO Two scenarios were evaluated here to optimize the position of FO and PRO process in the dual-stage system. The first scenario is the PRO-FO system in which the turbine system positioned after the firststage of FO membrane while the second scenario is the FO-PRO system with the turbine system positioned after the second stage of the FO membrane (Fig. 1). In both cases, i.e. PRO-FO and FO-PRO, 10 bar hydraulic pressure was assumed to be applied on the draw solution side of the membrane in the PRO process. Samples 1 to 5 were used as the draw solution in the FO and the PRO process, salinity range between 50 g/L to 157 g/L (Table 1). The impact of FO and PRO process arrangement on the permeate flow rate, net driving pressure (NDP), and the concentration of diluted draw solution was evaluated at different draw solution to feed solution (Qds-in/Qf-in) flow rate ratios (Fig. 2). The simulation results showed that the permeate flow from the firststage of the FO membrane was higher in the FO-PRO system than in the PRO-FO system (Fig. 2a). Obviously this was due to the higher net driving pressure in the FO-PRO system as shown in Fig. 2b. In contrast, the permeate flow rate was lower in the FO-PRO system than in the PRO-FO system due to the higher NDP in the second stage of the PROFO than in the second stage of the FO-PRO mode (Fig. 2a and b). Fig. 2c shows the impact of the Qds-in/Qf-in flow rate ratio on the concentration of diluted draw solution leaving the FO membrane (Cds-out). In the first-stage of the FO membrane, the Cds-out was higher in the PROFO than in the FO-PRO. The lower permeate flow rate generated from the PRO-FO system explains this phenomenon (Fig. 2a). The draw solution was exposed to a lower dilution in the PRO-FO system than in the FO-PRO system. However, in the second-stage of the FO membrane there was insignificant difference in the Cds-out between the PRO-FO system and FO-PRO system (Fig. 2c). In other words, both designs were able to achieve the same dilution level at the end of the treatment process. In fact the concentration of draw solution after the second-stage of the FO membrane was slightly lower in the PRO-FO system. This observation was also noticed at 107,272 mg/L (sample 2) concentration draw solution (results are not shown here). Practically, the effect of concentration polarization increases with the increasing concentration of the draw solution [14]; i.e. higher in the first-stage than in the second-
82
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
Table 1 The concentration and composition of flowback water samples (ref. [8]). Sample TDS (mg/L)
Na (mg/L)
Mg (mg/L)
Ca (mg/L)
Cl (mg/L)
SO4 (mg/L)
HCO3 (mg/L)
Sample 1: 53,272 Sample 2: 107,761 Sample 3: 123,801 Sample 4: 145,239 Sample 5: 157,930
629 31,500 47,734 54 54,843
1770 725 899 4730 6062
15,680 7820 6800 15,200 3600
35,000 67,100 68,000 125,000 93,000
10 560 20 60 10
183 56 348 195 415
stage of the FO membrane. This probably explains the equal draw solution concentrations in both designs. The power density (Pdensity) and the gross power (Pw) were calculated in both systems; the PRO-FO and the FO-PRO (Fig. 3). The simulation results showed that the PRO-FO system is more efficient than the FOPRO system in terms of the generated power density and the gross power through the turbine system. It is also observed from Fig. 3 that the power density increased with the increasing draw solution flow
1200
a 1000
Qp (L/h)
800
rate. This was due to the higher permeate flow rate at higher draw solution flow rate (Fig. 2a). The simulation results show that the efficiency of the PRO-FO system is higher than that of the FO-PRO system. As shown in Fig. 3, the dual-stage membrane treatment generates higher power when it is operated on the PRO-FO mode than on the FO-PRO mode. For example in sample 1, the power density generated from the PROFO system was almost 2.65 times higher than that from the FO-PRO system. The higher power generated from the PRO-FO operation mode didn't affect the quality and concentration of the treated saline solution. In both cases, the PRO-FO and the FO-PRO, the draw solution from the second-stage of FO treatment has the same concentration. As a matter of fact, the treated draw solution was more diluted in the PRO-FO system than in the FO-PRO system. Therefore, for the rest of this study the PRO-FO system will only be considered for the hypersaline solution treatment because of its higher dilution efficiency and capability to generate higher energy than the FO-PRO system.
600
3.2. The impact of draw solution flow rates
400 PRO-FO: Qp1 PRO-FO: Qp2
200
FO-PRO: Qp1 FO-PRO: Qp2
0 0
0.2
0.4
0.6
0.8
1
1.2
Qds-in/Qf-in 85
b
75
NDP (bar)
65 55 45 35
PRO-FO-NDP 1 PRO-FO-NDP 2 FO-PRO-NDP 1 FO-PRO-NDP 2
25 15 0
0.2
0.4
0.6
0.8
1
1.2
The impact of the draw solution flow rate on the performance of FO membrane was investigated using Qds-in/Qf-in ratios that vary from 0.2 to 1. Fig. 4 shows the impact of the draw solution flow rate on the permeate flow rate. Increasing the draw solution flow rate resulted in an increase in the permeate flow rate of the first and the second-stage of the FO treatment. This was due to the lower concentration polarization effect at the FO membrane surface and to the higher bulk feed concentration which resulted in a higher osmotic pressure across the FO membrane [15,16]. However, the permeate flow rate was higher in the firststage than in the second-stage of the FO membrane treatment because of the higher osmotic pressure across the membrane (Fig. 4c and d). Freshwater transport across the FO membrane diluted the concentration of draw solution in the first-stage of the FO process and hence its osmotic pressure decreased (Fig. 4c–d). As a result, the net driving pressure was lower in the second-stage than in the first-stage of the FO membrane. The simulation results show that the differential osmotic pressure and the permeate flow rate increased with the increasing
90000
c
80000
10
16
9
14
8
Power (kW)
Cds-out (mg/L)
12 7
70000 60000 50000
10
5
8
4
6
3
PRO-FO-Qds-out -1 PRO-FO-Qds-out-2 FO-PRO-Qds-out-1 FO-PRO-Qds-out-2
40000
6
Power: PRO-FO
2
Power density: PRO-FO
1
30000
4
Power: FO-PRO
Power Density (W/m2)
Qds-in/Qf-in
2
Power density: FO-PRO
0
0.2
0.4
0.6
0.8
1
1.2
Qds-in/Qf-in
0 0
0.2
0.4
0.6
0.8
1
0 1.2
Qds-in/Qf-in Fig. 2. The impact of Qds-in/Qf-in flow rate ratio on the permeate flow, NDP and concentration of draw solution in the PRO-FO and FO-PRO operation modes (samples 1 to 5).
Fig. 3. Power and power density at different Qds-in/Qf-in flow rate ratios (samples 1 to 5).
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
83
900
1400 TDS-53 g/l TDS-107 g/l TDS-123 g/l TDS-145 g/l TDS-157 g/l
1200
a
TDS-53 g/l TDS-107 g/l TDS-123 g/l TDS-145 g/l TDS-157 g/l
800 700
b
1000
Qp-2 (L/h)
Qp-1 (L/h)
600 800 600
500 400 300
400
200 200
100 0
0 0
0.2
0.4
0.6
0.8
1
1.2
0
0.2
0.4
Qds-in/Qf-in
0.6
0.8
1
1.2
Qds-in/Qf-in 70
120
c
TDS- 53 g/l TDS- 107 g/l TDS- 123 g/l TDS- 145 g/l TDS- 157 g/l
100
TDS- 53 g/l TDS- 107 g/l TDS- 123 g/l TDS- 145 g/l TDS- 157 g/l
60
d
50
∏ -2 (bar)
∏ -1 (bar)
80
60
40 30
40
20 20
10
0
0 0
0.2
0.4
0.6
0.8
1
1.2
0
0.2
0.4
Qds-in/Qf-in
0.6
0.8
1
1.2
Qds-in/Qf-in
Fig. 4. The impact of draw solution flow rate on the permeate flow rate and the differential osmotic pressure in the FO membrane (testing condition PRO-FO; applied hydraulic pressure 10 bar).
80% TDS-53 g/l TDS-107 g/l TDS-123 g/l TDS-145 g/l TDS-157 g/l
70% 60%
Re-1 (%)
concentration of draw solution from sample 1 to sample 5 (Fig. 4). This was true for all draw solution concentrations apart from sample 4 in which the net driving pressure was lower than that in sample 3. This can be attributed to the higher osmotic pressure of hypersaline solution in sample 2 than in sample 4; 98.4 bar and 96 bar respectively. The ionic composition of the draw solution in sample 3 possesses higher osmotic pressure than that in sample 4. For the same reason, the permeate flow rate in sample 2 was higher than that in sample 4 draw solution concentration (Fig. 4a and b). In addition to the permeate flow rate and osmotic pressure driving pressure, the impact of draw solution flow rate on stage one and two recovery rates was evaluated (Fig. 5). With exception to sample 4, which exhibited an osmotic pressure lower than sample 2, the FO recovery rate increased with the increasing concentration of the draw solution from sample 1 to sample 5 (Fig. 4c-d). The higher osmotic pressure across the FO membrane resulted in a higher permeate flow rate and hence the recovery rate increased (Fig. 4a and b). The recovery rates of the FO membrane also increased with the increasing Qds-in/Qf-in ratio because of the higher draw solution bulk concentration and osmotic pressure [17].
50% 40% 30% 20% 10% 0% 0
0.4
0.6
0.8
1
1.2
90% TDS-53 g/l TDS-107 g/l TDS-123 g/l TDS-145 g/l TDS-157 g/l
80%
b
60%
Re-2 (%)
The impact of changing the feed solution flow rates on the performance of the FO membrane is illustrated in Fig. 6. A significant increase in the osmotic pressure gradients across the FO membrane occurred upon increasing the Qds-in/Qf-in ratio while increasing the Qf-in/Qds-in ratio didn't cause a noticeable change in the osmotic pressure gradient across the membrane (Fig. 6a). Practically, increasing the draw solution flow rate reduces the effect of concentration polarization at the membrane surface and increases the concentration of bulk solution. This wasn't the case when higher feed solution flow rates are applied because of the negligible osmotic pressure (TDS b 200 mg/L) of the feed solution. Fig. 6b shows the impact of feed and draw solution flow rates on the
0.2
Qds-in/Qf-in
70%
3.3. Impact of changing the feed solution flow rates
a
50% 40% 30% 20% 10% 0% 0
0.2
0.4
0.6
0.8
1
1.2
Qds-in/Qf-in Fig. 5. The impact of the draw solution flow rate on the FO membrane recovery rate (testing condition PRO-FO).
84
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85
Qf-in/Qds-in 5
6
7
8
9
90
Δ∏ -Qds-in/Qf-in (bar)
90
a
80
80
70
70
60
60
50
50
40
40
30
30 Qds/Qf: ∆∏1 Qds/Qf: ∆∏2 Qf/Qds: ∆∏1 Qf/Qds: ∆∏2
20 10 0.2
0.4
0.6
0.8
20 10 0 1.2
0 0
of FO treatment was 35,374 mg/L and increased to 51,972 mg/L at Qds-in /Qf-in ratio equals to 0.2 (initial draw solution TDS is 123,801 mg/L, sample 2). The concentration limits of these diluted draw solutions may allow further treatment by RO membranes or discharge to seawater. However, when the Q ds-in /Qf-in ratio increased to 0.8 the concentration of diluted draw solution from the second-stage of the FO membrane was 57,516 mg/L; which is a challenging concentration for further treatment [18–22]. As such, higher draw solution dilution is achievable at lower Qds-in / Qf-in ratios. On the other hand, increasing the flow rate of feed solution has no impact on the performance of FO and hence it would only increase the power consumption instead of improving the process performance.
10
Δ∏ -Qf-in/Qds-in (bar)
4
1
Qds-in/Qf-in
4. Conclusions
Qf-in/Qds-in 4
5
6
7
8
9
10 1000
b
Qp-Qds-in/Qf-in (L/h)
900
900
800
800
700
700
600
600
500
500
400
400
300
300 Qds/Qf: Qp1 Qds/Qf: Qp2 Qf/Qds: Qp1 Qf/Qds: Qp2
200 100 0 0
0.2
0.4
0.6
Qp-Qf-in/Qds-in (L/h)
1000
200 100
0.8
1
0 1.2
8
9
10
Qds-in/Qf-in Qf-in /Q ds-in 4
5
6
7
90000
c
80000
80000
70000
70000
60000
60000
50000
50000
40000
40000
30000
30000
20000
Qds/Qf: Cds-out-1 Qds/Qf: Cds-out-2
20000
10000
Qf/Qds: Cds-out-1 Qf/Qds: Cds-out-2
10000
0
Cds-out-Qf-in/Qds-in (mg/L)
Cds-out-Qds-in/Qf-in (mg/L)
90000
0 0
0.2
0.4
0.6
0.8
1
1.2
Qds-in/Qf-in Fig. 6. Impact of feed and draw solution flow rates on the performance of FO membrane; a) impact on the osmotic pressure gradient b) impact on the permeate flow rate c) impact on the concentration of draw solution out (Cds-out).
permeate flow rate. No tangible change in the permeate flow rate was found upon increasing the flow rate of the feed solution while higher draw solution flow rates resulted in an increase in the permeate flow rate. Increasing the flow rate of draw solution not only increased the permeate flow rate and osmotic pressure, but also increased the concentration of draw solution from the treatment process (Fig. 6c). Actually, this parameter should be carefully calibrated because it affects the dilution of the hypersaline water which is the main objective of the membrane treatment. At low TDS concentration, the effect of increasing the flow rate of feed solution on the performance of FO membrane is negligible (Fig. 6). It is desirable, therefore, to have Qds-in/Qf-in ratios between low and average to achieve a sufficient dilution of the draw solution (saline water). For example, at Qds-in /Qf-in ratio equals to 2, the concentration of diluted draw solution from the second-stage
A dual-stage membrane process was investigated for the hypersaline solution treatment in which the hypersaline water and wastewater effluent were the feed and draw solution respectively. Two different systems were evaluated; i.e. PRO-FO and FO-RPO systems. The simulation results showed that the efficiency of PRO-FO design is higher than that of the FO-PRO design especially in terms of the power generation. The power generated from the PRO-FO system was 2.65 times higher than that from the FO-PRO system while the concentration of the diluted draw solution was the same in both designs. A higher draw solution dilution was achieved in the PRO-FO design at a Qds-in/Qf-in ratio equals to 0.2. Using higher Qds-in/Qf-in ratios resulted in a higher NDP across the membrane and hence increased the permeate flow rate. But it resulted in a higher draw solution concentration from the second-stage of FO membrane, which complicates the disposal and management of the diluted draw solution. The simulation results also showed that the increase of feed solution flow rate on the performance of FO membrane was negligible. Therefore, it is desirable not to use high feed flow rates as the power consumption will be increased. The complicated composition and the high concentration of hypersaline solution investigated in this study make it unsuitable for direct conventional thermal and membrane treatment [7,8]. The advantage of dual-stage FO membrane treatment is to bring the concentration of hypersaline solution low enough to be treated by the conventional processes or discharge properly to seawater. Experimental work would be required to investigate the process potential and feasibility. References [1] Penn State Extension, Marcellus Shale Wastewater Issues in Pennsylvania—Current and Emerging Treatment and Disposal Technologies, Penn State College of Agricultural Sciences, Cooperative Extension, 2011. [2] Gaudlip Tony, Water Use and Water Reuse/Recycling in Marcellus Shale Gas Exploration and Production, Penn State College of Agricultural Sciences, Cooperative Extension, Marcellus Shale Educational Webinar Series, 2010. [3] Paul T. Sun, Charles L. Meyer, Cor Kuijvenhoven, Sudini Padmasiri, Vladimir Fedotov, Treatment of water from fracturing operation for unconventional gas production, in: Ronald D. Neufeld (Ed.), Contemporary Technologies for Shale-Gas Water and Environmental Management, Water Environment Federation, Alexandria, Va, 2012, pp. 61–81. [4] Michael E. Tamblin, Pilot Project to Recycle and Treat Marcellus Shale Water, Clear Waters, winter 2010. 40–46. [5] O. Lefebvre, F. Habouzit, V. Bru, J.P. Delgenes, J.J. Godon, R. Moletta, Treatment of hypersaline industrial wastewater by microbial consortium in a sequencing batch reactor, Environ. Technol. 25 (2004) 543–553. [6] Adrienne Beckman, Archis Ambulkar, Art Umble, Diego Rosso, Joe Husband, Joseph Cleary, Julian Sandino, Mikel Goldblatt, Ron Horres, Ronald Neufeld, Russell Mau, Sam Jeyanayagam, Considerations for Accepting Fracking Wastewater at Water Resource Recovery Facilities, Water Environment Federation, 2014. (20/03/2014). [7] D. Yoxtheimer, Water use and Water Reuse/Recycling in Marcellus Shale Gas Exploration and Production, Penn State College of Agricultural Sciences, Cooperative Extension, Marcellus Shale Educational Webinar Series, 2010. [8] Tom Lewis, Fracking water disposal/recycling processes for unconventional shale gas wastewater, http://www.gwpc.org/sites/default/files/event-sessions/19r_ Lewis_Tom.pdf (25/03/2014). [9] Ali Altaee, Abdelnasser Mabrouk, Karim Bourouni, A novel forward osmosis membrane pretreatment of seawater for thermal desalination processes, Desalination V326 (2013) 19–29.
A. Altaee, N. Hilal / Desalination 350 (2014) 79–85 [10] Ali Altaee, Forward Osmosis, Potential use in desalination and water reuse, J. Membr. Sep. Technol. 1 (2012) 79–93. [11] Gordon T. Gray, J.R. McCutcheon, M. Elimelech, Internal concentration polarization in forward osmosis: role of membrane orientation, Desalination 197 (2006) 1–8. [12] Maruf Mortula, Sina Shabani, Removal of TDS and BOD from synthetic industrial wastewater via adsorption, 2012 International Conference on Environmental, Biomedical and Biotechnology, vol. 41, IACSIT Press, Singapore, 2012. [13] Feed water type and analysis, tech manual excerpt, http://www.dow.com/ PublishedLiterature/dh_003b/0901b8038003b4a0.pdf?filepath=liquidseps/pdfs/ noreg/609-02010.pdf&fromPage=GetDoc (20/01/2013). [14] J.R. McCutcheon, M. Elimelech, Influence of concentrative and dilutive internal concentration polarization on flux behavior in forward osmosis, J. Membr. Sci. 284 (2006) 237–247. [15] Ata M. Hassan, Review of development of the new NF-seawater desalination process from pilot plant to commercial production plant stages, The 6th Saudi Engineering Conference, KFUPM, Dhahran, December 2002, 2002. [16] Ata Hassan, Fully integrated NF-thermal seawater desalination process and equipment, US Patents No 2006/0157410 A1, July 20, 2006.
85
[17] Ali Altaee Guillermo Zaragoza Adel Sharif, Pressure retarded osmosis for power generation and seawater desalination: performance analysis, Desalination 336 (2014) 50–57. [18] Khaled Walha, Raja Ben Amar, Francis Quemeneur, Pascal Jaouen, Treatment by nanofiltration and reverse osmosis of high salinity drilling water for seafood washing and processing, Desalination 219 (2008) 231–239. [19] Weiyi Li, William B. Krantz, Emile R. Cornelissen, Jan W. Post, Arne R.D. Verliefde, Chuyang Y. Tang, A novel hybrid process of reverse electrodialysis and reverse osmosis for low energy seawater desalination and brine management, Appl. Energy 104 (2013) 592–602. [20] Qiang Li, Mengdi Wang, Jun Feng, Wei Zhang, Yuanyuan Wang, Gu. Yanyan, Cunjiang Song, Shufang Wang, Treatment of high-salinity chemical wastewater by indigenous bacteria — bioaugmented contact oxidation, Bioresour. Technol. 144 (2013) 380–386. [21] Nikolay Voutchkov, Overview of seawater concentrate disposal alternatives, Desalination 273 (2011) 205–219. [22] D.W. Westcot, Reuse and disposal of higher salinity subsurface drainage water, Agric. Water Manag. 14 (1988) 483–511.