A facilitated bulk liquid membrane transport approach is studied for Mg(II) extraction from seawater and reversed osmosis brine simulated media using 18-Crown-6 and dibenzo (DB)-18-Crown-6. The work is based on investigating the experimental parameters affecting the transport efficiency, such as pH of feed and receiving phase, type of membrane solvent, temperature, type and concentration of the carrier, and stripping solution conditions. The transported amount of magnesium ions from feed phase (Mg(II) = 0.059 M, NaCl = 0.01 M, pH = 3.3) across a chloroform membrane (18C6 = 0.001 M) into the receiving phase (SCN 0.1 M pH 3) was found to be %97 (±0.7) after 2.5 hr. The selectivity of the method was evaluated by performing competitive transport experiments on the mixtures containing Ca2+, Na+, K+, and Mg2+ ions.

INTRODUCTION

Magnesium is the third most plentiful element dissolved in seawater, with an average of 0.13%. Most magnesium and magnesium compounds are produced from seawater, well and lake brines and bitterns (Al-Mutaz & Wagialla 1990), as well as from minerals such as dolomite, magnesite, brucite, carnallite. Elemental magnesium alloys are increasingly used as a lightweight metal in the manufacture of engine parts, jet engines, missiles, and rockets. Likewise, dry magnesium salts and solutions are widely used in industrial, agricultural, medical, and environmental contexts (Kramer 2000).

The most common method for the initial separation of magnesium from seawater and brines is solar evaporation which induces a chemical precipitation. Solar evaporation is usually applied for processing highly concentrated brines such as the Dead Sea and Great Salt Lake waters (Epstein 1976; Tripp 2009). This method is based on the fact that MgCl2 is more soluble than NaCl. The chemical precipitation method is typically applied to low concentrated brines, such as seawater (Bhatti et al. 1984).

Traditional methods of magnesium production, i.e., processing of the hydromineral sources, despite their profitability, do not satisfy increasing ecological standards (Mero 1965). Consequently, new alternative ecological clean technologies, based on ion exchange of desalinated seawater with Mg2+ ions (Birnhack & Lahav 2007), nanofiltration/crystallization process (Drioli et al. 2004; Lahav et al. 2013) from desalination brines, precipitation of Mg(OH)2(s) on the surface area of self-synthesized magnetite (Fe3O4) micro-particles (Lehmann et al. 2014), and use of positively charged polyamide composite nanofiltration hollow-fiber membranes for lithium and magnesium separation (Li et al. 2015) have been developed. Furthermore, as the worldwide need for metals increases and the grades of ore decreases, the incentive to find more effective and efficient separation and purification methods grows.

Extraction techniques for metal separation (Mohite & Khopkar 1988) are of increasing interest due to the importance of environmental protection problems. In recent years, the use of various types of liquid membranes has attracted increasing attention due to their ability for selective metal ions separation in aqueous solutions (Danesi et al. 1981; Kamiński & Kwapiński 2000). Transport of metal ions across a membrane has an important role in industrial processes and has useful practical applications in separation science (Jafari et al. 2009). This type of membrane has been widely utilized for carrier facilitated separations and acquired significant importance for its use in separation, purification, or analytical application in diverse areas like water and waste water treatment (Shamsipur et al. 2010; Dalali et al. 2012). In liquid membranes, the extraction, stripping, and regeneration operations are combined in one single step which has the advantages of low cost and low energy consumption (León & Guzmám 2004; Revathi Reddy et al. 2010).

It seems that the main challenge to do an efficient liquid membrane transport process with unlimited feed like seawater (containing the macro components of Mg2+, Ca2+, K+, and Na+) is finding alternative carriers, and membrane conditions (Rounaghi et al. 1997), such as solvent type, stripping reagent, etc. Effective transport requires the carrier, which is distributed primarily to the membrane phase, to provide a selective complex with the desired metal ion (Kikuchi & Sakamoto 2000). Macrocyclic ligands, i.e., crown ethers as carrier (Izatt et al. 1986), have been developed to be used in several transport procedures for their selectivity to cation ions. In some studies (Heo 2012), selectivity of crown ethers to alkaline and alkaline earth ions is mentioned.

In this study, an attempt has been made to use facilitated bulk liquid membrane (BLM) transport using crown ethers 18-Crown-6 and DB-18-Crown-6 to find optimum conditions, such as pH, temperature, type and concentration of the carrier, solvent and ionic strength for extraction of magnesium ion from the simulated seawater media. Results show that the best extraction of magnesium was carried out at an acidic feed and received phase using 18-Crown-6 dissolved in chloroform as transport membrane in the presence of 0.01 M of sodium chloride. The selectivity of method was evaluated by performing competitive transport experiments on simulated seawater mixtures containing Ca2+, Na+, K+, and Mg2+ ions.

EXPERIMENTAL

Reagents

Dichloromethane, chloroform, and carbon tetrachloride as organic solvent membrane were purchased from Merck (Darmstadt, Germany). These solvents were washed three times with distilled water to remove the stabilizers and impurities. 18-Crown-6 ether (18C6) and dibenzo-18-Crown-6 ether (DB18C6) were purchased from Sigma-Aldrich (Steinheim, Germany). Analytical grade metal salts (mainly chloride such as MgCl2.6H2O) were purchased from Merck or Fluka (Buchs, Switzerland) and used for preparation of Mg2+ Ca2+, K+, and Na+ stock solution. Metal ions working solutions were prepared by dilution of their stock solutions in deionized water. Solutions of ethylene diamine tetra acetic acid (EDTA for metal ion standardizing), hydrochloric acid (HCl), and sodium thiocyanate were prepared by dissolving respective analytical reagent grade chemicals in deionized water.

Transport procedure

For separating the two aqueous phases, one transport cell (Figure 1), including a cylindrical glass with inside diameter of 3.7 cm and holding a glass tube inside it with a diameter of 2 cm (Jabbari et al. 2001) was utilized. By placing it in a thermostated water bath, the temperature of the transport cell was kept constant at 30 °C. A feed phase (total volume of 5 mL) was included of Mg+2 (0.059 M) and NaCl (0.01 M), and adjusted by HCl solution at pH = 3.3. The receiving phase (total volume of 5 mL) was included of SCN solution (0.1 M) and set at pH of 3. These aqueous phases were separated by 20 mL chloroformic solution of 18C6 (0.001 M). Facilitated BLM transport work was performed by stirring the organic phase with a speed of 100 rpm using a Teflon-coated magnetic bar. After a given time, concentration of Mg2+ in the aqueous phases was separated and determined by EDTA titration to end point with Eriochrome Black T (pH = 10).
Figure 1

BLM and schematic representation of the ion exchange mechanism.

Figure 1

BLM and schematic representation of the ion exchange mechanism.

RESULTS AND DISCUSSION

Selection of membrane solvent

The main objective of this study is identification of the best conditions for magnesium separation from seawater and reversed osmosis (RO) brine. Organic solvents have an important role in BLM transport (Ajwani et al. 2012). According to the high concentration of magnesium ions in these samples and in order to prevent evaporation of the solvent, the organic phase with a higher density was selected. The organic phase should include features such as low volatility and viscosity, suitable dielectric constant, polarity and also the ability for a good exchange phase with high distribution coefficient (Zidi et al. 2011). Therefore, several heavy solvents, such as dichloromethane, chloroform, and carbon tetrachloride were tested. Results of this study are shown in Table 1. It appears that suitable dielectric constant and polarization ability of chloroform leads to better behavior in magnesium transport than other solvents. Extraction percent (%E) and stripping (%S) are calculated from the following formulas (Fakhari et al. 2006): 
formula
where, C0f, Cf are initial and final concentration of Mg2+ in the feed phase and CR is concentration of Mg2+ in the receiving phase after the transport process.
Table 1

Effect of membrane solvent type on transport procedure

%E Solvent 
16.95 Dichloromethane 
20.34 Chloroform 
8.47 Carbon tetrachloride 
%E Solvent 
16.95 Dichloromethane 
20.34 Chloroform 
8.47 Carbon tetrachloride 

Experiment conditions: [Mg+2] = 0.059 M; temperature = 25 °C; feed pH = 4; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2 hr.

The type and concentration of the carrier

The type and concentration of the carrier for metal ion transporting is another important factor that should be selected properly. In facilitated BLM transport, the carrier forms a complex with the guest cation and helps it to transfer through the hydrophilic membrane (Ajwani et al. 2012). For this purpose, the behavior of two crown ethers (18-Crown-6 and DB-18-Crown-6) was studied and the results are shown in Table 2. It seems the greater flexibility of 18-Crown-6 works better than the hydrophobicity of DB-18-Crown-6 (possessing two benzenes) in the chloroform membrane for magnesium transport. It should be noted that DB-18-Crown-6, due to possessing two benzene groups, is more hydrophobic than 18-Crown-6. For the same reason, it is easier for 18-Crown-6 ether to adjust its structure for better reacting with Mg2+, and this leads to its greater flexibility.

Table 2

Effect of carrier type, tested concentration is 0.01 M in chloroform

%E Structure Crown type 
11.86  DB18C6 
18.36  18C6 
%E Structure Crown type 
11.86  DB18C6 
18.36  18C6 

Experiment conditions: [Mg+2] = 0.059 M; membrane solvent = chloroform; temperature = 25 °C; feed pH = 4; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2 hr.

In the next stage of this study, the concentration of 18-Crown-6 as carrier in the range of 0.0001 to 0.1 M was evaluated. Results show that 0.001 M with an extraction of 20.33% has the best performance. It seems that at higher concentration of carriers, due to increasing irreversible interaction between metal cations and 18-Crown-6 in membrane, the extraction efficiency was decreased. In the facilitated BLM transport process, the complexation reaction between the carrier and the cation should not be so strong, since its delivery is not done in receiving phases. A high concentration of carrier ligand leads to irreversible reaction with the metal ions and the transfer process will be stopped.

Effect of type and concentration of the counter ion

Transport of Mg2+ across the membrane via the predicting mechanism of Figure 1, needs to apply a contour ion. For this purpose, different types of 0.01 M salt solution, such as KCl, KI, KClO3, NaNO3, NaBr, and NaCl, were studied against 0.059 M of magnesium ions to find a better contour ion. According to the results of Table 3, in the presence of Br, magnesium transport is higher. However, due to economic reasons and the abundant presence of Cl in seawater and RO brine, which are our target samples, Cl was selected as the contour ion. In order to determine the optimum concentration of contour ion, NaCl at a concentration range of 0.0001–2 M was tested in the membrane containing 18-Crown-6. As a result (Figure 2), the maximum extraction percent of magnesium ion was found at the concentration of 0.01 M of NaCl.
Table 3

Effect of type of the counter ion

%E Contour ion (0.01 M) 
16.59 KCl 
16.59 KI 
20.34 KClO3 
15.25 NaNo3 
22.03 NaBr 
20.34 NaCl 
%E Contour ion (0.01 M) 
16.59 KCl 
16.59 KI 
20.34 KClO3 
15.25 NaNo3 
22.03 NaBr 
20.34 NaCl 

Experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18 crown 6 in chloroform; temperature = 25 °C, feed pH = 4; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2 hr.

Figure 2

Effect of Cl concentration in feed phase, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; temperature = 25 °C, feed pH = 4; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2 hr.

Figure 2

Effect of Cl concentration in feed phase, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; temperature = 25 °C, feed pH = 4; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2 hr.

Effect of pH of feed solution on extraction efficiency

At this stage, the pH of the feed phase was adjusted in the range of 1–10, using concentrated NaOH and HCl solutions. Results of magnesium ions transport in different pH values are plotted in Figure 3. As is clear, at pH = 3.3, the extraction percent of magnesium (22.8%) is maximum compared with the other points. This is due to the ability of H+ to pull crown ether to the boundary zone of the membrane, and the increase of Cl for the contour ion role. Therefore, this pH was selected as the optimum point for future work.
Figure 3

Effect of feed pH on Mg2+ transport, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; temperature = 25 °C; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2 hr.

Figure 3

Effect of feed pH on Mg2+ transport, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; temperature = 25 °C; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2 hr.

The effects of type and concentration of stripping reagent in receiving phase

Transport of Mg2+ in the mentioned membrane is based on reaction between 18-Crown-6 as a chelating agent and this cation. The complex after passing through the aqueous/organic interface is transferred to the receiving phase, but in the absence of stripping reagent, transport is negligible. For this purpose, the effect of several stripping reagents (0.1 M of EDTA, SCN, sodium citrate, nitric acid, oxalic acid, and citric acid) on the stripping percent (%S) was studied. Figure 4 presents the results of this experiment, which accordingly show SCN with the best behavior as stripping reagent. To find the optimum concentration of SCN, its concentration was investigated in a range of 0.0001–2 M and a concentration of 0.1 M was chosen as the optimum amount.
Figure 4

Effects of type of stripping reagent in receiving phase with concentration of 0.1 M for every reagent, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; feed pH = 3.3; temperature = 25 °C; pH of the receiving solution = 3; extraction time = 2 hr.

Figure 4

Effects of type of stripping reagent in receiving phase with concentration of 0.1 M for every reagent, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; feed pH = 3.3; temperature = 25 °C; pH of the receiving solution = 3; extraction time = 2 hr.

Effect of pH of receiving phase on stripping process

Here, due to the importance of the role of pH of the receiving phase on the stripping process, this parameter was adjusted at different amounts by using concentrated NaOH and HCl solutions in the presence of 0.1 M SCN as the optimum stripping reagent. Results of this study are shown in Figure 5, and as is clear, the best stripping percent was obtained at pH = 3. Therefore, this pH was selected as the optimum pH of the receiving phase. It seems that the main role of acidic media is establishing a tendency for the metal complex to be present between the aqueous/organic boundary zone.
Figure 5

Effect of receiving pH on Mg2+ stripping, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; feed pH = 3.3; stripping reagent = 0.1 M of SCN; temperature = 25 °C; extraction time = 2 hr.

Figure 5

Effect of receiving pH on Mg2+ stripping, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; feed pH = 3.3; stripping reagent = 0.1 M of SCN; temperature = 25 °C; extraction time = 2 hr.

Effect of time and cell temperature on transport

Since BLM transport is time-consuming, finding the optimal time for the work is very important. Figure 6 shows the changes in extraction and stripping percent of magnesium ions via time in hours. As can be seen, maximum extraction and stripping efficiency can be achieved at 2.5 hours and at higher times there are no great changes. Therefore, this time was selected for further work.
Figure 6

Effect of time on BLM transport of Mg2+, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; feed pH = 3.3; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; temperature = 25 °C.

Figure 6

Effect of time on BLM transport of Mg2+, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; feed pH = 3.3; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; temperature = 25 °C.

Temperature of the transport cell is the next factor whose effect was investigated at the three points of 25, 30, and 35 °C. To do this work, the transport cell was inserted into a water bath and the results of this study are plotted in Figure 7 for the feed phase as extraction percent and receiving phase as stripping percent.
Figure 7

Effect of temperature on BLM transport of Mg2+, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; feed pH = 3.3; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2.5 hr.

Figure 7

Effect of temperature on BLM transport of Mg2+, experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18-Crown-6 in chloroform; [Cl] = 0.01 M; feed pH = 3.3; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2.5 hr.

Selectivity of BLM system

As the aim of this study was separation and extraction of magnesium ion from seawater and RO brine, the selectivity of the BLM system was evaluated by Mg2+ transport studies in the presence of Ca2+ (450 mg/L), Na+ (13,680 mg/L), and K+ (800 mg/L) ions in the feed solution close to the reported concentration of these ions in seawater. Table 4 shows the percentage transport of Mg2+ (0.059 M) and these cations into the receiving phase. As is observed, in the absence of any masking agents, the interfering effects of these foreign metals ions are low.

Table 4

Selectivity study of BLM transport of Mg2+ in several interference mixture simulated seawater

No. Mixture % Transported into receiving phase 
Mg2+ 98 
 Ca2+ nd 
Mg2+ 98 
 Na+ 7.8 
Mg2+ 98 
 K+ 2.1 
Mg2+ 97 
 Ca2+ nd 
 Na+ 8.3 
 K+ 3.7 
No. Mixture % Transported into receiving phase 
Mg2+ 98 
 Ca2+ nd 
Mg2+ 98 
 Na+ 7.8 
Mg2+ 98 
 K+ 2.1 
Mg2+ 97 
 Ca2+ nd 
 Na+ 8.3 
 K+ 3.7 

Experiment conditions: [Mg+2] = 0.059 M; membrane = 0.001 M of 18 crown 6 in chloroform; [Cl] = 0.01 M; feed pH = 3.3; stripping reagent = 0.1 M of SCN; pH of the receiving solution = 3; extraction time = 2.5 hr. nd: not detected.

CONCLUSIONS

As mentioned before, with the increasing consumption of metals such as magnesium, extraction has become important, especially from the easy to access source, seawater. Here, a facilitated BLM transport method is presented for extraction of Mg(II) from simulated seawater using 18-Crown-6 as carrier. The operational conditions leading to Mg(II) transport were established as: 0.01 M of sodium chloride, pH of feed = 3.3, membrane consisting of chloroform solution of 18C6 (0.001 M), 0.1 M SCN as stripping reagent and pH = 3 in the receiving phase and overall transport time of 2.5 hours. Despite the good results of the selectivity study of the method in the mixture solutions, it seems we need to find a more selective carrier or extraction condition for magnesium, even if its extraction efficiency is lower, because the used feed phase material (seawater) is completely unlimited.

REFERENCES

REFERENCES
Ajwani
P.
Lokwani
L.
Sharma
U.
2012
Bulk liquid membrane transport of alkali metal cations using non cyclic ionophores
.
Proceedings of the National Academy of Sciences, India Section A: Physical Sciences
82
,
91
95
.
Al-Mutaz
I. S.
Wagialla
K. M.
1990
Production of magnesium from desalination brines
.
Resources, Conservation and Recycling
3
,
231
239
.
Bhatti
A. S.
Dollimore
D.
Dyer
A.
1984
Magnesia from seawater: a review
.
Clay Minerals
19
,
865
875
.
Danesi
P. R.
Horwitz
E. P.
Vandegrift
G. F.
Chiarizia
R.
1981
Mass transfer rate through liquid membranes: interfacial chemical reactions and diffusion as simultaneous permeability controlling factors
.
Separation and Purification Technology
16
,
201
211
.
Drioli
E.
Curcio
E.
Criscuoli
A.
Profio
G. D.
2004
Integrated system of CaCO3, NaCl and MgSO4 7H2O from nanofiltration retentate
.
Journal of Membrane Science
239
,
27
38
.
Fakhari
A.
Khorrami
A. R.
Shamsipur
M.
2006
Selective uphill Zn2+ transport via a bulk liquid membrane using an azacrown ether carrier
.
Separation and Purification Technology
50
,
77
81
.
Izatt
R. M.
Lindh
G. C.
Bruening
R. L.
Bradshaw
J. S.
Lamb
J. D.
Christensen
J. J.
1986
Design of cation selectivity into liquid membrane systems using macrocyclic carriers
.
Pure and Applied Chemistry
58
,
1453
1460
.
Kamiński
W.
Kwapiński
W.
2000
Applicability of liquid membranes in environmental protection
.
Polish Journal of Environmental Studies
9
,
37
43
.
Kramer
D. A.
2000
Magnesium, its Alloys and Compounds, U.S.G.S Open-File Report 01–341
.
Lahav
O.
Telzhensky
M.
Zewuhn
A.
Gendel
Y.
Gerth
J.
Calmano
W.
Birnhack
L.
2013
Struvite recovery from municipal-wastewater sludge centrifuge supernatant using seawater NF concentrate as a cheap Mg(II) source
.
Separation and Purification Technology
108
,
103
110
.
Mero
J. L.
1965
The Mineral Resources of the Sea
.
Elsevier Publishing Co.
,
New York
,
USA
.
Revathi Reddy
T.
Ramkumar
J.
Chandramouleeswaran
S.
Reddy
A. V. R.
2010
Selective transport of copper across a bulk liquid membrane using 8-hydroxy quinoline as carrier
.
Journal of Membrane Science
351
,
11
15
.
Tripp
T. G.
2009
Production of magnesium from Great Salt Lake, Utah, USA
.
Natural Resources and the Environment
15
,
54
61
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).