Microsoft Word - 43poletti.docx


 CHEMICAL ENGINEERING TRANSACTIONS  
 

VOL. 47, 2016 

A publication of 

 
The Italian Association 

of Chemical Engineering 
Online at www.aidic.it/cet 

Guest Editors: Angelo Chianese, Luca Di Palma, Elisabetta Petrucci, Marco Stoller
Copyright © 2016, AIDIC Servizi S.r.l., 
ISBN 978-88-95608-38-9; ISSN 2283-9216 

Effect of MgCl2 on Energy Generation by Reverse 
Electrodialysis 

Ahmet H. Avcia, Pinkey Sarkara, Diego Messanab, Enrica Fontananovab, Gianluca 
Di Profiob, Efrem Curcioa,b  
a 
Department of Environmental and Chemical Engineering, University of Calabria DIATIC-UNICAL, Via P. Bucci CUBO 45A, 

87036 Rende (CS) Italy  
b 
Institute on Membrane Technology, National Research Council of Italy ITM-CNR, Via P. Bucci CUBO 17C, 87036 Rende 

(CS) Italy 
ahmethalilavci@hotmail.com 

One of the membrane-based technique for harvesting salinity gradient energy is Reverse Electrodialysis 
(RED). Most of the studies in literature are based on utilizing river and seawater by mimicking them with NaCl 
solutions. However, real solutions contain many different ions; this study deals with the impact of Mg2+ ion on 
power density (Pd). Investigations have been carried out using six solutions of NaCl and MgCl2 prepared in 
varying compositions from 0 % to 100 %. Increasing Mg2+ concentration resulted in a remarkable decrease of 
Pd and open circuit voltage (OCV), and in a significant increase in stack resistance Rstack. Electrochemical 
impedance test (EIS) and ion chromatography analysis helped to clarify reasons for loss of power. In 
particular, it was found out that dominant negative contribution is due to increasing Cation Exchange 
Membrane (CEM) resistance in the presence of Mg2+.  

1. Introduction 
According to US Energy Information Administration, world net electricity generation is expected to increase 
from 20 trillion to 40 trillion kWh in the coming 30 years. Coal, natural gas, nuclear and renewable energies 
are sources which presently satisfy the increasing demand of energy. Among all these sources, renewable 
energy is the fastest-growing source of electric power with an annual 2.8 % increase (U.S. Energy Information 
Administration, 2013). An emerging renewable energy source is Salinity Gradient Power (SGP), originally 
proposed for sea and river water more than 60 years ago (Pattle, 1954). The total technical potential of SGP is 
estimated to be around 647 gigawatts, which is 23 % of global electricity consumption. There are two common 
technologies which harvest SGP by utilizing membrane based processes: Reverse Electrodialysis (RED) – 
object of the present investigation - and Pressure Retarded Osmosis (PRO). RED is distinguished from PRO 
in the case of utilizing high concentration solutions like brine and seawater, instead of utilizing seawater and 
river water (Post et al. 2007). Possible application areas of SGP techniques are estuaries where freshwater 
rivers run into seawater, high salinity wastewater (brine from desalination or salt mining) and saltwater lakes 
(International Renewable Energy Agency, 2014). In a typical RED system, cation exchange membranes 
(CEM) and anion exchange membranes (AEM) are stacked alternately; driven by a concentration gradient 
between a Low Concentration Compartment (LCC) and a High Concentration Compartment (HCC), the 
diffusive flux of ions generates an electrochemical membrane potential recorded as a voltage across 
electrodes. In last 10 years, researches were focused mostly to increase efficiency and Pd of RED, as 
maximum result 2 W/m2 power generated by using river and seawater (Vermaas et al., 2013). Nevertheless, 
since most of the results were obtained mimicking solutions only by using water and NaCl, further 
investigation is necessary to have more realistic outcome (Tufa et al., 2014; 2015). The objective of our study 
was to enlighten the reduction effect of Mg2+ on Pd and transport mechanism of ions.  

                               
 
 

 

 
   

                                                  
DOI: 10.3303/CET1647061

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Please cite this article as: Avci A., Sarkar P., Messana D., Fontananova E., Di Profio G. , Curcio E., 2016, Effect of mgcl2 on energy generation 
by reverse electrodialysis, Chemical Engineering Transactions, 47, 361-366  DOI: 10.3303/CET1647061

361



2. Materials and methods 
2.1 Experimental setup 

A SGP-RE stack (Figure. 1) with an active membrane area of 0.01 m2 (10 cm x 10 cm) was operated in cross-
flow mode and equipped with 25 cell pairs for all achieved experiments provided by REDstack B.V (The 
Netherlands). Stack basically has 2 inlet and 2 outlet channels for saline solutions, 2 electrolyte compartments 
continously recirculated, a cathode and an anode. AEM-80045 and CEM-80050 were main constituents of one 
cell pair which were provided by Fujifilm Manufacturing Europe B.V (The Netherlands).  

 

Figure 1: Salinity gradient power reverse electrodialysis unit used in experiments 

Important characteristics of Ion Exchange Membranes (IEMs) are tabulated in Table. 1. Each cell pair was 
supported with 270 mm polyethylene gaskets and PET spacers (Deukum GMBH, Germany). Besides 
repeating cell units, at the both end of the stack there are electrode compartments which consist anode and 
cathode made of inert Ti–Ru/Ir mesh had a dimension of 10 cm x 10 cm (MAGNETO Special Anodes B.V., 
The Netherlands).  

Table 1: Properties of ion exchange membrane 

Membrane code  Thickness (mm) Ion exchange capacity 
(mmol/g membrane) 

Density of fixed charge 
(mol/L) 

Membrane area 
resistance (Ωcm2)

Fuji-AEM-80045 129 ± 2 1.4 ± 0.1 3.8 ± 0.2 1.551 ± 0.001 
Fuji-CEM-80050 114 ± 2 1.1 ± 0.1 2.4 ± 0.2 2.974 ± 0.001 

2.2 Solutions 

LCC and HCC solutions were prepared by dissolving required amount of NaCl and MgCl2·6H2O (Sigma-
Aldrich, Italy) in deionized water (PURELAB, Elga LabWater®, 0.055 µScm-1). All solutions were fed to the 
stack at a flow rate of 20 L/h by Masterflex L/S digital peristaltic pumps Mod. no. 7528-10 6–600 rpm (Cole-
Palmer, US). The electrolyte solution was 0.3 M potassium hexacyanoferrate(II), 0.3 M potassium 
hexacyanoferrate(III) and 2.5 M sodium chloride (Sigma-Aldrich S.r.l., Italy). For the recirculation of electrolyte 
solution, Masterflex L/S digital peristaltic pump (Cole-Palmer, US) was operated at 30 L/h. 

2.3 Electrochemical measurements 

Representative electric circuit for RED stack and attached measurement devices are shown in Figure. 2. A 
high dissipation five-decade resistance box in the range of 0.1–1000 Ω (CROPICO, Bracken Hill, US) was 
used to load the SGP-RE system. DC voltage drop across the RED stack was measured by a 3½ digital 
multimeter with accuracy of ±0.5 % in the range of 200 mV to 200 V (Valleman, DVM760), and the current 
flowing across the load resistors was measured by Agilent 34422A 6½ digit multimeter. 
All the experiments were performed by varying external resistance load whereas electric potential drop (V) 
and current (I) were measured simultaneously by ammeter and voltmeter, respectively. After fitting a straight 
line for the collected data V(V) vs. I(A), open circuit voltage OCV (V) which is maximum obtainable voltage 
value, and Rstack (Ω) which is total resistance of stack, were calculated from Eq(1): = −    (1) 

362



Intercept of fitted first order linear line gives OCV value where = 0 and slope of the line gives Rstack similarly. 
After having OCV and Rstack, Pd (W/m

2) one of the most important performance criterion of RED, can be 
calculated equation given by Eq(2). =    (2) 
It is also possible to obtain maximum Pd value by plotting calculated Pd vs. i (A/m

2) which follows a parabolic 
trendline in the form of Eq(3). = + +    (3) 
which allows to calculate maximum Pd by Eq(4). 

 , = −    (4) 

 

Figure 2: Electric circuit diagram of the experimental apparatus. The ammeter (A) is connected in series and 
voltmeter (V) connecter in parallel with the resistance box (Rload) 

2.4 Impedance tests 

All impedance tests were carried out by using home designed impedance test which allows 3.14 cm2 active 
membrane area. Two identical solutions were fed and recirculated by using gear pumps to the cell immersed 
in a thermostatic bath and continuously stirred. Inside of the cell, four electrodes (working, counter, reference 
and sense electrodes) were arranged to apply AC current and measure potentials as shown in Figure. 3.  

ME MB R A N E
N aC l 

solution
N aC l 

solution

R eference 
elec trode

R eferenc e 
elec trode

IMP E D A N C E  
A N A LY ZE R

C
O

U
N

T
E

R
 E

L
E

C
T

R
O

D
E W

O
R

K
IN

G
 E

L
E

C
T

R
O

D
E

 

Figure 3: Schematic illustration of electrochemical impedance test 

363



To evaluate specific resistances that contribute to total compartment resistance, a potentiostat/galvanostat 
combined with a frequency response analyzer, Metrohm Autolab PGSTAT302N was used. By applying AC 
current over a range, voltage drop through the membrane was measured. AC current in the range of 1000-
0.01 Hz was generated through the cell and response was recorded in the software of Metrohm Autolab 
PGSTAT302N. After collecting all necessary data, an equivalent circuit model (described in section 3.2) was 
fitted by the help of software Nova 1.9.16 by Metrohm Autolab B.V. (Fontananova et al., 2014). 

2.5 Ion Chromatography 

To quantify transported ions across ion exchange membranes, Metrohm 861 Advanced Compact Ion 
Chromatograph was used. Characterization of anions and cations done by using Metrosep A Supp 5 - 250/4.0 
and Metrosep C 4 – 250/4.0 separation column, respectively. 3.2 mM Na2CO3 + 1mM NaHCO3 was used as 
eluent of anion column, and 2mM HNO3 0.25mM oxalic acid was used as eluent of cation column.  

3.  Results 
3.1 Performance of stack  

Figure. 4 includes results from performance analysis of RED stack for six solutions which contain different 
amount of NaCl and MgCl2 (Table 2). Best performance was achieved by using pure NaCl solution. 
Introduction of 10 % of MgCl2 resulted in 20 % and 60 % decrease of OCV and Pd, respectively; a decreases 
of OCV and Pd up to 58 % and 94 % in the case of pure MgCl2 solution was observed.  
  

 

Figure 4. Gross power density vs current density  

Table 2: Solutions composition, open circuit voltage (OCV), stack resistance (Rstack) gross power density (Pd).   

Soln # HCC Composition  LCC Composition OCV (V) Rstack (Ω) Pd(W/m
2) 

1 4.00 M NaCl 0.50 M NaCl 1.70 2.78 1.06 
2 3.60 M NaCl+ 0.40 M MgCl2 0.45 M NaCl+ 0.05 M MgCl2 1.36 4.44 0.43 
3 3.20 M NaCl+ 0.80 M MgCl2 0.40 M NaCl+ 0.10 M MgCl2 1.30 4.67 0.36 
4 2.40 M NaCl+ 1.60 M MgCl2 0.30 M NaCl+ 0.20 M MgCl2 1.29 5.11 0.32 
5 1.60 M NaCl+ 2.40 M MgCl2 0.20 M NaCl+ 0.30 M MgCl2 1.15 6.38 0.21 
6  4.00 M MgCl2 0.50 M MgCl2 0.72 8.92 0.06 

 
 

Rstack results refer to all components of resistance including membranes. In order to discriminate the specific 
contribution of a specific ion exchange membrane typology, EIS tests were carried out on CEM and AEM.   

364



3.2 EIS tests = + + ∆    (5) 
The internal area resistance Ri (Ωm

2) of cell consists of three parts (Eq(5)): Rm+s arises from membrane and 
solution resistances, Rdl is related to the electrical double layer, and Rdbl is due to concentration polarization in 
the boundary layer adjacent to the membrane (Fontananova et al., 2014). The equivalent circuit model for 
interpretation of EIS data is shown in Figure. 5. In equivalent circuit model, 3 different resistance was taken 
into account in series mode. Resistance of membrane and solution represented as simple resistance, 
resistance of electrical double layer represented as a resitance and a capacitance in parallel and resistance of 
diffusion boundary layer represented as a resistance and constant phase element in parallel. 

 

Figure 5: Resistance mechanisms on CEM and AEM (up), equivalent circuit model (down) 

Table 3 reports EIS data refer to LCC solutions composition. Ohmic resistances dominate on non-ohmic ones, 
and Rm for CEM membranes were the most sensitive to MgCl2 content, while Rm for AEM remained stable.  

Table 3:  Rm, Redl, Rdbl, for varying MgCl2 concentration, CEM (left) and AEM (right) 

CEM AEM 

% MgCl2  Rm (Ωcm
2) Redl (Ωcm

2) Rdbl (Ωcm
2)     % MgCl2  Rm (Ωcm

2) Redl (Ωcm
2) Rdbl (Ωcm

2) 
0 2.410 0.048 0.518 0 1.350 0.014 0.131 

10 6.975 0.088 0.656 10 1.493 0.008 0.069 
20 9.399 0.101 0.958 20 1.369 0.009 0.061 
40 15.515 0.065 1.118 40 1.269 0.008 0.059 
60 18.943 0.069 1.302 60 1.459 - - 
80 20.994 - - 80 1.370 - - 

100 23.215 0.076 0.978 100 1.407 - - 

3.3 Mass transport 

Previous studies carried out by using monovalent ions show that ions’ mass transfer direction is the same 
direction of the driving force by concentration difference; on the other hand, it is possible to find some 
evidence of uphill transport, i.e. mass transfer takes place in the reverse direction (Vermaas et al., 2014). In 
this study, in the case of cation, ion transport was observed in the same direction for monovalent Na+, 
whereas tranport of Mg2+ is in the negative direction for solutions up to 20 % MgCl2 content (Figure. 6). 
  

365



 

Figure 6: Percentage of immigrated ions from HCC to LCC with varying MgCl2 concentration; sodium and 
magnesium 

4. Conclusion 
In this study, AEM-80045 and CEM- 80050 membranes supplied by Fujifilm were tested for energy generation 
by RED unit for different NaCl/MgCl2 solutions. Maximum gross power density obtained was 1.06 W/m

2 when 
using pure NaCl solutions (LCC:HCC=0.5M:4M), while a drastic decrease on power density was observed at 
increasing MgCl2 content, mainly caused by the negative impact of Mg

2+ cation on CEMs resistance. 
Presence of Mg2+ in feed solution of RED to mimic real solutions reveals that present RED membranes suffer 
from presence of multivalent ions. Therefore, IEM materials still need a significant improvement to compete in 
the field of renewable energy. 

Acknowledgements 

The financial support of the Education, Audiovisual and Culture Executive Agency (EU-EACEA) within the 
EUDIME “Erasmus Mundus Doctorate in Membrane Engineering” program (FPA 2011-0014, SGA 2014-0970, 
http://eudime.unical.it) is kindly acknowledged. 

Reference 

Fontananova, E., Zhang, W., Nicotera, I., Simari, C., Baak, W.v., Di Profio, G., Curcio, E., Drioli, E., 2014, 
Probing membrane and interface properties in concentrated electrolyte solutions, Journal of Membrane 
Science, 459, 177-189.  

International Renewable Energy Agency (2014). Salinity Gradient Energy. Bonn, Germany. 
Pattle, R.E., 1954, Production of electric power by mixing fresh and salt water in the hydroelectric pile, Nature, 

174, 660, DOI: 10.1038/174660a0 
Post, J.W., Hamelers, H.V.M., Buisman, C.J.N., 2009, Influence of multivalent ions on power production from 

mixing salt and fresh water with a reverse electrodialysis system, Journal of Membrane Science, 330, 65-
72 

Tufa, R.A., Curcio, E., Baak, W.v.,Veerman, J., Grasman, S., Fontananova, E., Di Profio, G., 2014, Potential 
of brackish water and brine for energy generation by salinity gradient power-reverse electrodialysis (SGP-
RE), Royal Society of Chemistry, 4, 42617-42623 

Tufa, R.A., Curcio, E., Brauns, E., van Baak, W., ., Fontananova, E., Di Profio, G., 2015, Membrane 
Distillation and Reverse Electrodialysis for Near-Zero Liquid Discharge and low energy seawater 
desalination, Journal of Membrane Science, 496, 325-333 

Vermaas, D.A., Veerman, J., Saakes, M., Nijmeijer, K., 2014, Influence of multivalent ions on renewable 
energy generation in reverse electrodialysis, Energy and Environmental Science, 7,1434-1445. 

U.S. Energy Information Administration (2013). International Energy Outlook 2013.  Washington, US. 

0 20 40 60 80 100
% MgCl2

-8

-4

0

4

8

12

%
 c

at
io

n 
tr

an
sp

or
t i

n 
LC

C

Na+

Mg2+

366