Inﬂuence of Mixed Electrolyte on the Performance of Iron-Ion/Hydrogen Redox Flow Battery

Improved charge/discharge performance of Iron-ion/Hydrogen redox ﬂow battery (RFB) electrolyte with a mixed FeSO 4 and FeCl 2 is reported. Addition of Cl − ions into a sulfate electrolyte changes the charge/discharge behavior of the sulfate electrolyte leading to a reduction in charging potential for a mixed FeSO 4 and FeCl 2 electrolyte system. This suggests that a sulfate/chloride electrolyte system can lead to improved charge/discharge of the Fe-ion/H 2 RFB. Reverse addition of FeSO 4 to FeCl 2 showed a decrease in the mixed electron transfer efﬁciency (experimental current relative to theoretical) equivalent to a decrease in electrolyte performance. We deduce that 0.8 M FeCl 2 corrosive electrolyte can be replaced by less corrosive mixture of 46 mol % Cl − in 0.8 M FeSO4 to achieve the same performance that can be obtained using an all chloride system. iron chlo- ride is that it exhibits high vapor pressure and tend to be corrosive. However, with appropriate electrolyte optimization, the battery’s per- formance can be greatly enhanced and it can be implemented into current storage systems. This study investigates the battery’s electrolyte performance im-provement through the mixing of iron chloride and iron sulfate elec- trolytes. The study uses simple cyclic voltammetric (CV) technique and the electron transfer efﬁciency as a measure to quantify the elec- trolyte performance. Electron transfer efﬁciency measured by the use of simple CV analysis compares the actual number of electrons trans- ferred in the reaction (calculated using Randles-Sevcik equation – see Eq. 1) over the theoretical number of electrons involved in an electrochemically balanced equation.

There is a need to reduce the reliance on finite energy resources such as fossil fuels used in energy production. Burning of fossil fuels produces green house gases such as CO 2 which may be directly linked to global warming and severe weather conditions. Furthermore, fossil fuels are costly. 1 Alternative energy, such as solar and wind, can be unstable for everyday use. Advanced storage technologies such as redox flow batteries (RFB) are being investigated as solutions because the electrolyte can be charged and stored for later usage during which the battery is discharged. This allows the battery to be very versatile in small-scale and large-scale operations because it allows for consistent energy distribution and management across the entire day. RFB presents a potentially cheap and nontoxic option of using more renewable energy resources. In addition, it offers megawatt storage capacity depending on the electrolyte used. 1 Popular redox flow batteries such as the vanadium flow battery, are viable solutions, but carry a high upfront and maintenance cost. 1 Vanadium has poor solubility in solvents and tends to precipitate out of solutions. Furthermore, lead, chromium, vanadium and bromine are amongst other electrolyte materials used but are considered costly, and more toxic when compared with iron, for example. The hydrogen ion/iron ion redox flow battery has captured the interest of the scientific community due to the availability and cheapness of the materials needed for fabrication. So far, the best two electrolytes are ion sulfate and iron chloride. [1][2][3][4][5][6] The iron-ion/hydrogen-ion RFB system also has a high energy conversion efficiency ranging up to 90% and nearly unlimited charge/discharge cycles without any optimization. The main problems impeding the implementation of this battery are related to electrolyte, kinetics and overpotential issues. 2 Amongst these problems, electrolyte issue is the greatest challenge. The regeneration of the electrolyte is documented as one of the biggest challenges facing the use of the pure iron sulfate electrolytes. Furthermore, with sulfate electrolytes, it is hard to improve the level of Fe concentration. Fe sulfate solution is also associated with high overpotential, lower power density, high viscosity that affects transport of electrolyte to electrodes and poor charge transfer. Associated with pure iron chloride is that it exhibits high vapor pressure and tend to be corrosive. [2][3][4][5][6] However, with appropriate electrolyte optimization, the battery's performance can be greatly enhanced and it can be implemented into current storage systems.
This study investigates the battery's electrolyte performance improvement through the mixing of iron chloride and iron sulfate electrolytes. The study uses simple cyclic voltammetric (CV) technique * Electrochemical Society Active Member. z E-mail: ekalu@eng.fsu.edu and the electron transfer efficiency as a measure to quantify the electrolyte performance. Electron transfer efficiency measured by the use of simple CV analysis compares the actual number of electrons transferred in the reaction (calculated using Randles-Sevcik equationsee Eq. 1) over the theoretical number of electrons involved in an electrochemically balanced equation.
An electrolyte comprised of a mixture of sulfate and chloride electrolytes is expected to contribute proportional individual excellent performance properties to the sum of both electrolytes used together.

Materials and Methods
An equimolar 0.8 M amounts of FeSO 4 and FeCl 2 and their acids (H 2 SO 4 and HCl, respectively) were mixed and degassed with N 2 for one hour. Pt, graphite and glassy carbon working electrodes were used along with Pt wire counter electrode and Ag/AgCl reference electrodes in both divided and undivided cell set-up. In the divided cell set-up, Nafion 212 membrane was used as the separator. In both set-ups, linear scanning voltammetry (LSV) and cyclic voltammetery (CV) analysis were carried out using a Gamry Instruments Interface 1000 potentiostat/galvanostat/ZRA. Scan rates varied from 0.025 to 0.01 V/s. For simple H-cell charge/discharge arrangements, both Pt catalyzed carbon cloth and Pt foil working electrodes were evaluated for preliminary and worst case performances without system optimization.

Results and Discussion
The balanced electrochemical reactions of iron-ion/hydrogen ion RFB during discharge are: Cathode: Anode: Overall reaction: As seen in Eqs. 2 and 3, a single electron exchange is involved in the battery's discharge or charge. A simple CV technique utilized in the analysis of Fe 3+ containing solution results in the characteristic reversible redox duck-like curves shown in Fig. 1a. Graphite electrode was used as working electrodes for both curves whose anodic peak currents are 0.016A and 0.028A for the sulfate and chloride solutions, respectively. Both anodic peaks occurred at approximately 0.68V which is representative of the Fe 2+ to Fe 3+ in both electrolytes. As expected, iron chloride exhibited higher peak currents during the redox process, presumably due to the higher charge carrying capacity of the chloride ions (0.00052 cm 2 /Vs) compared to the sulfate ions (0.00027 cm 2 /Vs). Tucker et al., 4 proposed that the two best electrolytes for iron-ion/hydrogen ion RFB are iron chloride and iron sulfate. 4 However, chloride is corrosive and will increase cost of materials used in the battery fabrication. To capture the excellent properties of both chloride and sulfate electrolytes, we hypothesized that a mixture of both electrolytes in various volume ratios could yield a better performing electrolyte mixture than a single electrolyte. Figure 1b shows the CV curve at 25 mV/s of a mixture of 50% (V/V) 0.8M iron sulfate with 0.8M iron chloride. The result clearly indicates that chloride addition to sulfate electrolyte enhances the peak currents of the redox reaction, while exhibiting reversibility over the same voltage range of 0.68V and 0.3 V for the oxidation and reduction peaks, respectively on graphite WE. A similar observation was made when GC and platinum electrodes were used as the working electrodes.
To better understand the influence of chloride addition into sulfate electrolyte, the actual electrons transferred (assuming Eq. 2 reaction occurs) is calculated using Randles-Sevcik equation and this is compared to the theoretical number of electrons (Eq. 1). for electron transfer efficiency or electrolyte performance. The electrolyte 'performance' as used here, compares the theoretical charge transfer or current flow versus the measured current flow on a particular electrode. For instance in a 100% Fe (II) sulfate electrolyte, the theoretical current flow for one electron transfer using Randles-Secvik equation is compared to experimentally measured current to determine the electron transfer efficiency as approximately equal to 68%. The performance here then means that only 68% of charge passed could account for the Fe oxidation reaction. However, after certain incremental chloride addition, the electron transfer efficiency is observed to increase to 91% signifying that about 91% of charge passed could account for the one electron Fe oxidation reaction of interest.
The electron transfer efficiency results obtained with three different working electrodes for FeSO 4 is shown in Fig. 1c. Graphite exhibited stable electron transfer efficiency regardless of the scan rate. Our speculation is that the overpotential for the reversible reaction: Fe 3+ + e Fe 2+ is lower in graphite as compared to the other two electrodes. In addition, hydrogen evolution competes with the Fe 3+ reduction, however hydrogen overvoltage is much higher in graphite than in Pt and thus graphite is more favorable for the iron electrocatalysis. A review of the literature shows that carbon is a preferred electrode material for the redox reaction of iron even though Pt is one of the best electrode materials available. Furthermore, graphite is much cheaper than GC and Pt. As such, it was the most optimum electrode material for further use in Fe 2+ /Fe 3+ redox reaction studies.
A systematic addition of 0.8 M FeCl 2 into 0.8 M FeSO 4 was carried out and the results are shown in Fig. 2. Both Figures 2a and 2b show increasing electron transfer efficiency (81% to 96%) as the chloride addition goes from 1 -50% (V/V). Was this increase in electron transfer efficiency of the sulfate electrolyte really due to chloride addition? To answer this question, iron chloride was used as the starting electrolyte followed by systematic addition of 1-50% (V/V) 0.8M iron sulfate. As shown in Fig. 2b, the curve with the negative slope represents the addition of sulfate ion into the chloride electrolyte while the curve with a positive slope represents the gradual addition of chloride ion into the sulfate containing electrolyte. The electron transfer efficiency for the chloride electrolyte decreased from 81% to 52% as sulfate ion is added. The addition of chloride to sulfate had the opposite effect as shown by the curve with a positive slope where the electron efficiency increased from about 66% to 92%. This clearly confirms that chloride addition to sulfate electrolyte plays a significant role in enhancing the electrolyte's charge/discharge performance as shown in Fig. 1b CV where increased peak current was obtained after mixing chloride and sulfate electrolytes. Results in Fig. 2b further suggest that a mixture of chloride and sulfate can perform even better than chloride alone. For instance, based on Fig. 2b results, the addition of 46 mol % chloride ion into 56 mol % sulfate electrolyte results in equivalent electrolyte performance (electron transfer efficiency) as 100 mol % chloride ion of same concentration. In respect of corrosion, we speculate that a 100% chloride electrolyte is more corrosive than sulfate electrolyte with only 46 mol % chloride ion content. Our speculation is aligned with corrosion analysis performed elsewhere in which chloride ions are shown to be corrosive to various materials with an increase in chloride ion concentration. [7][8][9][10][11][12] Furthermore, Fig. 2b shows that with only 70 mol % chloride in a sulfate electrolyte, the electrolyte charge/discharge performance is improved to a value much higher than an all chloride (100% chloride) containing electrolyte.
The effect of chloride ions added to ammonium iron sulfate electrolyte is also investigated as shown in Figure 3. The incremental addition of iron chloride to the ammonium iron sulfate yielded increase in electron transfer efficiency similar to that observed for the mixed FeCl 2 and FeSO 4 electrolyte system shown in Fig. 2.
One of the significant challenges facing iron-ion/hydrogen RFB is the regeneration of Fe 3+ from Fe 2+ discharge product since it is identified as high energy consuming step. In this study, using Pt-catalyzed carbon cloth electrode, it requires about 3 V applied potential to charge 0.8 M iron sulfate as shown in Fig. 4a. However, after mixing iron sulfate 50% (V/V) and iron chloride 50% (V/V), the charging potential required was reduced to about 1.4V which is more than 50% reduction in the energy required for the sulfate only electrolyte shown in Fig. 4a. The reduced charge voltage supports the premise of higher electron transfer efficiency based on the relationship between electrolyte conductivity and species concentration. The charge voltage for 100% sulfate electrolyte under the experimental conditions was achieved at 3.0V. However, with the addition of chloride ion into the sulfate electrolyte, the charge voltage was reduced to 1.4V. This indicates improved charge transport. It can be shown that for the two electrolytes (1 and 2, respectively) at the same charge current, the relationship between the charge voltages (V ), transfer efficiencies (η) and electrolyte resistances (r ) is V 2 = V 1 ( η 2 η 1 )( r 2 r 1 ). Since the increase in electron transfer efficiency is attributed to chloride addition, it can also be responsible in reducing the charge voltage due to increase in reversibility of the redox reaction aiding Fe 3+ to Fe 2+ . This result is very promising and supports the earlier deduction reached through the evaluation of electron transfer efficiency in CV experiments. Thus, both results obtained using simple CV techniques and H-cell charge/discharge setup point to probable validity of a mixed chloride and sulfate electrolyte as better performing electrolyte for iron-ion/hydrogen RFB. Also, the authors are not aware of any existing report in the literature detailing the effect and use of mixed (chloride and sulfate) electrolyte in the iron-ion/hydrogen RFB. Figure 4b is a simple H-cell set-up discharge of 0.8 M iron sulfate using a platinum foil working electrode. Hydrogen was bubbled into the sulfuric acid anolyte solution and the system exhibited a steady decrease after an hour of discharge. Figure 4c represents the discharge of a mixed electrolyte of 0.8 M Fe 2 (SO 4 ) 3 and 0.8M FeCl 3 in a simple H-cell batch set-up. No effort was made to optimize the set-up. We observe the voltage decreased from 0.85 V to 0.33 V within the 1 st hour of the 3 hour discharge. A full cell setup studies is underway to demonstrate the observed CV data analysis. This will be reported in our future communication.

Conclusions
The conclusion of the CV results is in agreement with a full cell study in literature comparing the sulfate and chloride electrolyte performances for iron-ion/hydrogen-ion RFB. The new results from the present study show that the addition of Cl − increases performance of sulfate electrolyte. This may be due to the higher mobility of 0.00052 cm 2 /Vs Cl − ions. The work suggests that a mixed sulfate/chloride electrolyte system can lead to an improved charging/discharging of the Fe-ion/H 2 RFB. Based on the results, a 100% pure 0.8 M FeCl 2 corrosive electrolyte system can be replaced by less corrosive mixture of 46 mol % Cl − in 0.8 M FeSO4 to achieve the same performance of an all chloride electrolyte system. A similar increase in the performance of the electrolyte was observed for the addition of chloride to ammonium iron sulfate. Furthermore, optimization could be achieved in mixing other redox couples.