Aqueous NaOH battery

Materials Science and Engineering B 177 (2012) 1788–1792

Contents lists available at SciVerse ScienceDirect

Materials Science and Engineering B journal homepage: www.elsevier.com/locate/mseb

Short communication

Looking beyond lithium-ion technology – Aqueous NaOH battery Manickam Minakshi ∗ School of Chemical and Mathematical Sciences, Murdoch University, Murdoch, WA 6150, Australia

a r t i c l e

i n f o

Article history: Received 21 May 2012 Received in revised form 27 August 2012 Accepted 16 September 2012 Available online 27 September 2012 Keywords: MnO2 Aqueous Sodium Rechargeable

a b s t r a c t The objective of this work is to investigate a water based sodium battery technology. The new concept proposed here for an aqueous rechargeable battery is replacing lithium hydroxide with a sodium hydroxide electrolyte in the patented technology developed at Murdoch University. Alternative energy storage system using abundantly available sodium as the aqueous electrolyte coupled with Zn anode and environmentally friendly MnO2 cathode are investigated and found feasible. Sodium intercalation and de-intercalation mechanism is identified in MnO2 |NaOH|Zn cell yielding 225 against 142 mAh/g for LiOH counterpart. The preliminary studies of the aqueous NaOH battery showed improved energy density and voltaic efficiency. © 2012 Elsevier B.V. All rights reserved.

1. Introduction The safety of the lithium-ion battery is an important factor in its utility. This factor, along with a battery’s storage and multiple charge–discharge capacity, exert a profound influence on acceptance of a power storage device in the market place. The organic electrolyte currently in use has several safety issues including flammability and electrochemical instability. The rechargeable Liion battery fails rapidly at temperatures in excess of 60 ◦ C and thermal runaway occurs above 80 ◦ C [1,2]. A potential battery fire in any consumer electronics or in an electric vehicle is a serious concern. The search for more robust and less flammable electrolytes is complicated by the many criteria required for an efficient power storage device. For example, solid polymer electrolytes possess the desired stability but show low ionic conductivity and give rise to the growth of dendrites on the solid electrode surfaces during multiple charge–discharge cycles [3,4]. For safety concerns, aqueous electrolytes are the natural choice in this field [5,6]. Water is cheaper than organic solvents, has fewer disposal and safety issues. The ionic conductivity of aqueous electrolytes is two orders of magnitude greater than that of organic electrolytes, allowing higher discharge rates and lower voltage drops due to electrolyte impedance. Batteries with aqueous electrolytes do not need to be sealed tight against the entry of water vapour [5]. The main disadvantage of using an aqueous electrolyte is that the battery voltage is limited to ∼2 V. To offset the effects of this voltage limit, the storage capacity and the charge–discharge

∗ Tel.: +61 8 9360 2719; fax: +61 8 9310 1711. E-mail addresses: [email protected], [email protected] 0921-5107/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.mseb.2012.09.003

characteristics of the aqueous system must be significantly superior to those of the non-aqueous system. Recently, authors have demonstrated that lithium intercalation can even occur reversibly in the battery that uses aqueous electrolytes and found the cells are cheap, rechargeable and safe [7]. Current benchmarks for the capacity of nickel-metal hydride (Ni-MH) and Li-ion batteries are 90 and 165 Wh/kg, respectively [6] and our aqueous rechargeable (Zn–MnO2 ) battery [7] using LiOH solution is comparable to the existing technology. Recently, factors like increase in demand for oil, CO2 emissions and global warming put an enormous pressure on the world energy market to find a competitive energy storage devices. Storing large amounts of electricity cheaply – something that will be essential for making renewable energy the primary source, rather than just the supplemental source is a current challenge. Such storage will make it practical to store energy from wind turbines and solar farms for later use. Although lithium-ion battery is a state-of-the-art for such devices but the trade-off between safety and cost has significantly hampered their utility. In contrast to lithium resources, sodium deposits are in plenty around the world and lower in cost. Next to lithium, the electrochemical equivalent and standard potential of sodium are the most attractive and beneficial for battery applications. To achieve it in an effective way, sodium-based energy storage systems have been identified as a key technology for the future as the lithium technology is more expensive [8,9]. Sodium is orders of magnitude more abundant than lithium and cheaper to use. The sodium-ion battery works at room temperature and uses sodium ions that is available widespread. Nevertheless, existing sodium technologies work at high temperature, where molten sodium and molten sulphur are the anode and cathode, respectively, and

M. Minakshi / Materials Science and Engineering B 177 (2012) 1788–1792

2. Experimental MnO2 (␥-MnO2 type) was purchased from the Foote mineral company. Zn foil (99.9%) from Advent research materials, analytical reagent grade lithium hydroxide monohydrate (LiOH·H2 O) and sodium hydroxide (NaOH) from Sigma Aldrich were used in this study. The MnO2 active material was first mixed with 15 wt.% of carbon black (A-99, Asbury, USA) and with 10 wt.% of poly (vinylidene di fluoride) (PVDF, Sigma Aldrich) as a binder and then pressed into a disc shape with a diameter of 12 mm. The 12 mm diameter, 0.5 mm thick electrode weighing 35 mg, containing 75% by weight of MnO2 , i.e., 26 mg has a theoretical capacity to 1.0 e– discharge to Mn3+ of 308 mAh/g. An electrochemical test cell was constructed with the disk as the cathode, Zn metal as the anode and filter paper (Whatman filters 12) as the separator. The electrolyte was a 7 M solution of sodium hydroxide or saturated amount of lithium hydroxide (5 M) containing 1 M of ZnSO4 and the Li/Na amount was 2-fold excess to meet the amount of ions in the solution for intercalation mechanism. We have tried earlier with a range of molar concentrations of NaOH electrolyte (2, 5, 7 and 10 M) and optimised 7 M as the best performance and the reported data in this paper is for 7 M NaOH. The cells were charged/discharged galvanostatically at 0.25 mA/cm2 on mass density (8 mA/g) using an 8 channel battery analyser from MTI Corp., USA, operated by a battery testing system (BTS). The cut-off discharge and charge voltages were 1.0 and 1.8 V versus Zn (anode), respectively. All electrochemical measurements were carried out at ambient temperature (25 ◦ C). The cell configuration used for potentiostatic (standard three cell electrode) is described elsewhere [15]. For cyclic voltammetric studies, MnO2 was the working and zinc foil was the counter electrode, which was separated from the main NaOH or LiOH electrolyte by means of a porous frit. A mercury–mercuric oxide (Hg/HgO) served as the reference electrode with a concentration of 4.2 M KOH as the electrode filling solution. The electrode potential of the Hg/HgO is +0.098 V versus the normal hydrogen electrode (NHE). Reported potentials are relative to Hg/HgO.

0.002 -70 mV A1 -56 mV

0.001 Current / A

they have never found widespread use. An alternative strategic approach used in this study is sodium battery for a promising low temperature energy storage device. Sodium-ion cell mechanism is similar in some ways to lithiumion cells (a familiar “rocking-chair” type cell). Ions are shuttled between the battery’s positive and negative electrodes during charging and discharging, with an electrolyte serving as the medium for moving those ions. However, by simply replacing the lithium ions with sodium ions is problematic – sodium ions are 70% bigger than lithium ions [10] and do not fit well in the electrodes matrix empty spaces. Only selected cathode materials are suitable. As the ionic volume of sodium ion is larger, only those of materials possessing two-dimensional layered structures or three-dimension with corner sharing matrix, and or crystal structure forming suitable large-size tunnels can reversibly accommodate sodium ions [11]. Hence, in this study manganese dioxide (MnO2 ) is chosen as a cathode material. A novel battery is fabricated on low cost, safe and sustainable manganese oxide as cathode and zinc as anode with NaOH electrolyte for rechargeable battery system. Although sodium ion insertion had been performed by few researchers earlier but most of them are limited to structural characterisation and those reported for battery applications are confined to non-aqueous sodium electrolytes [12–14]. In the proposed system, as sodium is not involved in the anodic process, it is not a “rocking-chair sodiumion cell” but it is a “MnO2 |NaOH|Zn” hereinafter called as “NaOH cell” with sodium insertion on the cathode and zinc dissolution on the anode.

1789

(a) A2

(b)

0.000

-0.001

C1 C1 -390 mV -285 mV

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

Potential vs. Hg/HgO / V Fig. 1. Typical cyclic voltammogram of MnO2 electrode (first cycle) using aqueous (a) 7 M NaOH and (b) saturated amount of LiOH electrolytes under identical conditions (scan rate: 25 ␮v s−1 ; potential limit: 0.2 to −0.45 V and back).

The structural determination of the un-cycled and discharged MnO2 was done by Siemens D500 X-ray diffractometer 5635 using Cu K␣ radiation and a scan speed of 1◦ /min. Elemental concentration of the MnO2 cathode (85 wt.%) mixed with carbon black and (15 wt.%) binder (10 wt.%) in powder form were analysed through proton induced X-ray emission (PIXE) and proton induced Gammaray emission (PIGE) techniques for elements heavier than Al, and lighter than Al, respectively, with both techniques being nondestructive. This ion beam measurement was performed at ANSTO using the 2 MV tandem ion beam accelerator with 2.6 MeV protons. 3. Results and discussion Fig. 1 shows a typical cyclic voltammogram (CV) of MnO2 electrode. The behaviour of MnO2 in aqueous NaOH can be compared to that in aqueous LiOH electrolyte under identical conditions. The CV’s of MnO2 in the two electrolytes are quite different, reduction peak (C1 ) in NaOH observed at −390 mV but in the presence of LiOH it observed at −285 mV. On reversing the sweep, during oxidation, no such big difference in anodic potential is seen for those two electrolytes. Based upon what is postulated by us in the earlier publications [16,17] the cathodic peak for LiOH was assigned to lithium-intercalated MnO2 phase in addition to a variety of Mn3+ intermediates (MnOOH, Mn2 O3 or Mn3 O4 ) were formed through proton insertion [17] and the corresponding anodic peak was the reverse of this reaction to a formation of MnO2 phase. A small shoulder (A2 ) at 29 mV seen for LiOH electrolyte is assigned to ␦-MnOOH phase suggesting that the process is not completely reversible. All details pertaining to the products those formed during redox processes for LiOH electrolyte is available in the literature [16,17]. For NaOH electrolyte, reversible sodium insertion and extraction is proposed. During reduction, a peak shift of around 105 mV was observed at more negative potential implying the bigger size of sodium ions are difficult to intercalate [18] than extracting from the host MnO2 as the oxidation peaks have occurred at quite similar regions for both the electrolytes with a difference of only 14 mV. During the reverse anodic sweep oxidation peak A1 at −55 mV was observed for NaOH indicating that the process is reversible. Tiny shoulders at around 20 and 120 mV are also seen but they are not well-defined as observed for Li counterpart. This implies ␦-MnOOH and other Mn3+ non-rechargeable products that are not readily formed, promote the NaOH cell to have enhanced electrochemical behaviour. Fig. 2 shows the changes in the CV profile for NaOH

1790


0.0020

NaOH 25

1

0.0015

Current / A

0.0010 0.0005 0.0000

-0.0005 -0.0010 -0.0015

25 1

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

Potential vs. Hg/HgO / V Fig. 2. Effect of repeated voltammetric cycling on the current vs. potential profile of MnO2 in 7 M aqueous NaOH electrolyte. The numbers in figure denote cycle number.

NaOH Vs. LiOH cell

2.0 (a)

Fig. 4. X-ray diffraction pattern of manganese dioxide (MnO2 ) in NaOH electrolyte (a) before and (b) after discharged at shorter and (c) longer period of time (see details for text).

Cell Voltage / V vs. Zn

1.55 V

1.5

(b) 1.35 V

1.0 Excess of storage 36% in NaOH cell

0.5

0.0

0

40

80

120

160

200

240

-1 Cell Capacity / mAh g Fig. 3. The first discharge–charge profile of Zn–MnO2 cells using aqueous (a) saturated amount of LiOH and (b) 7 M NaOH electrolyte under identical conditions (0.25 mA/cm2 ).

electrolyte when the MnO2 electrode is subjected to continuous cycling (up to 25 cycles). The cathodic and anodic peak currents decreased up to 10% in efficiency for the first ten cycles and then stabilised suggesting that the MnO2 material is versatile for reversible sodium ion intercalation/de-intercalation over a number of cycles in aqueous NaOH electrolyte. At the 25th cycle, MnO2 material investigated in NaOH was degraded to 24% which may not be readily attractive for battery applications. To enhance the reversibility in sodium battery, a separate study will be carried out on incorporating few additives into MnO2 structure. However, a detailed study on reversibility of MnO2 containing additives in LiOH battery is reported [16,17]. The effect of replacing LiOH with NaOH in Zn/MnO2 cell was determined by carrying out discharge cycles on two identical cells, but using two different electrolytes, one with 7 M NaOH (named NaOH cell) and the other saturated amount of LiOH (named LiOH cell). For easy comparison, the same current density of 0.25 mA/cm2 was used. The results for the first discharged–charged cycle for both the cells are shown in Fig. 3. The discharge characteristics for the cell with LiOH are quite different from that of the NaOH cell (Fig. 3). The discharge capacities of the MnO2 cathode for the NaOH cell was

225 mAh/g (300 Wh/kg) compared to 142 mAh/g (210 Wh/kg) for the LiOH cell using a 1 V cut-off voltage. The cell voltage for NaOH was 0.2 V below as compared to LiOH cell, however, the decrease in cell potential during discharge was slower comprising 36% excess of energy storage. The observed lower voltage for NaOH cell is due to the cathodic effect, the energy gained from inserting Na into a host MnO2 structure was lower than that for inserting Li from LiOH cell. The reason explained by Ong et al. [19] is the fact that Na tend to form weaker bonds with O than Li and would depend strongly on the local environment in the MnO2 crystal. However, the weaker Na O bonding [20] had a lower diffusion barrier for Na+ to intercalate in MnO2 than the Li+ in the LiOH cell and hence the higher capacity. The NaOH cell was fully reversible with a capacity of 225 mAh/g and delivering a discharge voltage of 1.35 V resulting in a higher energy density of 300 Wh/kg. The active material utilisability (consumed amount of Mn4+ in MnO2 ) was calculated to be 73% of the 1 e– capacity of the true MnO2 content and this value is well agreed with the elemental concentration analysis shown in Table 1. The cell capacity may be limited by the presence of planar Zn anode, nevertheless, this problem can be alleviated using porous Zn anode [21]. During the discharge process, there was a gradual decrease in potential until 1.2 V corresponding to Mn4+/3+ redox characteristics with a sodium intake and then a sharp drop in voltage was achieved at a discharge cut-off voltage of 1.0 V. During the charge process there was a gradual increase in voltage to the charge cut-off voltage of 1.8 V corresponding to Mn3+/4+ redox reactions. The performance of MnO2 cathode in aqueous NaOH electrolyte has a better performance in terms of higher energy density and material utilisation while comparing to the MnO2 cathode in non-aqueous (NaClO4 ) solvents [22]. The smooth voltage profile indicates intercalation/de-intercalation occurred within a single phase and this has been supported by X-ray diffraction (XRD) of the discharged product. The products formed during the cell discharge were examined by XRD. The XRD pattern of the material (MnO2 ) before any electrochemical treatment (with carbon included for conductivity) is shown in Fig. 4a. The observed diffraction peaks are the characteristic for ␥-MnO2 material. The high intensity peak corresponding to the carbon black (denoted as “C”) mixed with ␥MnO2 for conductivity is also evidenced in the parent material. A comparison of the XRD spectra of the starting material and that of


1791

Table 1 Elemental analysis on the MnO2 electrode before and after discharged at various levels in NaOH electrolyte. Concentrations of elements were determined by PIXE technique. Sample

Na (␮g/cm2 )

Mn (␮g/cm2 )

Utilisability of MnO2 cathode (%)

Zn (␮g/cm2 )

Standard MnO2 sample (before any electrochemical treatment) Short discharged (equivalent capacity 100 mAh/g) Long discharged (equivalent capacity 225 mAh/g)

0

85.996

100

0

2.895

54.343

33

0.168

6.287

26.926

73

0.213

2.0

NaOH cell

Cell Voltage / V

1.5 1

1.0

25

10

0.5

0.0

0

50

100

150

200

250

-1

Cell Capacity / mAh g

Fig. 5. Cyclability of Zn–MnO2 cells using aqueous 7 M NaOH electrolyte.

the product formed on its discharge at different discharged level is shown in Fig. 4. The XRD patterns of the discharged material at a shorter (equivalent capacity 100 mAh/g) and longer period of time (equivalent capacity 225 mAh/g) are shown in Fig. 4b and c, respectively. The discharged material showed an evolution of new peaks and the original peaks corresponding to un-cycled material shifted to lower 2Â values. These new peaks are assigned to NaMnO2 in accordance with the values quoted in the ICDD card numbers 0250844; 038-0965. The un-reacted MnO2 peaks are also observed in Fig. 4b. The peak intensities corresponding to the new phase (NaMnO2 ) are slightly higher for those powders discharged to a larger extent of time. This suggests that the discharge mechanism of MnO2 in NaOH involve Na+ insertion forming NaMnO2 similar to that of the lithium insertion behaviour occurred in LiOH aqueous electrolytes [15–17,21]. In a separate experiment, elemental concentration on the surface evaluated by proton induced X-ray and gamma-ray emission (PIXE and PIGE) analyses for the discharged MnO2 samples confirmed that Na+ intercalation is involved. The longer period of discharged time had a greater extent of sodium ions intercalated (as shown in Table 1) in the MnO2 sample. The amount of Zn incorporated from anodic dissolution, shown in Table 1, is too negligible in the MnO2 electrode during discharge processes that could inhibit the intercalation mechanism. The cyclability data for MnO2 as cathode in NaOH cell is shown in Fig. 5. The zinc–manganese dioxide (Zn–MnO2 ) battery using NaOH electrolyte had an excellent discharge capacities of 184 and 171 mAh/g at 10 and 25th cycles, respectively. This corresponded to an acceptable efficiency, after multiple cycles, MnO2 cathode lost 24% of its initial capacity. As the calculated discharge capacity is the product of the imposed discharge current and discharge time, shorter discharge time in Fig. 5 after successive cycles mean a lower discharge capacity. This implies that the total internal resistance of the NaOH cell may be high that shortens the time to reach the identical cut-off voltage. To overcome this problem, discharge current

rate can be lowered to deliver higher capacity. On the other hand, if the reversibility is hampered by means of structural instability or lower conductance (Na+ ) then a dopant material can be added into the host MnO2 or lowering the pH of NaOH electrolyte may enhance the reversibility. Due to the fat Na+ ion or lower conductance, host MnO2 structure is not versatile for multiple cycling while smaller Li+ ion or higher conductance of the LiOH system showed much potential for battery applications [23]. To enhance the cyclability of MnO2 |NaOH|Zn battery an extensive research need to be undertaken both in electrochemical and materialistic point of view. The larger storage capacity is promising for NaOH cell although the cycleability needs to be improved but the intention of the work is achieved through reversible intercalation mechanism in aqueous solutions. To summarise, low cost sodium ion battery is feasible for energy storage systems. Storing larger amounts of electricity cheaply, something that will be essential for making renewable energy the primary source of energy. The water based sodium battery (NaOH cell) is non flammable and has higher ionic conductivity necessary for rate capability and it could be a competitive to Ni-MH batteries. 4. Conclusions The proposed MnO2 |NaOH|Zn develops an important new family of energy storage devices based on an affordable, globally available element, sodium. The innovative science in this study involves reversible aqueous sodium electrochemistry at low temperature (against the available relatively high temperature, at which Na is molten). The electrochemical behaviour of zinc–MnO2 alkaline cell showed that the aqueous NaOH electrolyte functions quite differently from our conventional cell, which uses LiOH electrolyte. When a cell containing aqueous NaOH is discharged, sodium is intercalated into the host MnO2 structure and the mechanism appears to be quite identical to that of LiOH cell but the size of Na+ ion or its conductivity may render limitations in cyclability. The reversibility of the NaOH cell need to be enhanced for practical applications through incorporating few oxide additives to the host MnO2 matrix. However, the absence of non-rechargeable products like Mn3+ intermediates and ␦-MnO2 lead to a larger storage capacity which is 36% higher than that of lithium battery (LiOH) counterpart. Intercalation of sodium in aqueous solution is confirmed through XRD and PIXE analyses of the discharged MnO2 electrode. The sodium energy storage technology will offer immediate advantages over existing primary battery technologies in terms of high energy density, cost, safety and environmental considerations. The aqueous sodium battery has a potential interest for large scale energy storage systems such as smart grid applications. Acknowledgements The author (M.M.) wishes to acknowledge the Australian Research Council (ARC). This research was supported under Australian Research Council (ARC) Discovery Project funding scheme (DP1092543) and Centre for Research into Energy for Sustainable Transport (CREST) (Centre of Excellence, Project 1.1.5). The views

1792


expressed herein are those of the author (M.M.) and are not necessarily those of the ARC and CREST. References [1] Q. Wang, P. Ping, X. Zhao, G. Chu, J. Sun, C. Chen, Journal of Power Sources 208 (2012) 210–224. [2] F. Torabi, V. Esfahanian, Journal of the Electrochemical Society 158 (2011) A850–A858. [3] C.-K. Park, A. Kakirde, W. Ebner, V. Manivannan, C. Chai, D.-J. Ihm, Y.-J. Shim, Journal of Power Sources 97–98 (2001) 775–778. [4] R.A. Leising, M.J. Palazzo, E.S. Takeuchi, K.J. Takeuchi, Journal of the Electrochemical Society 148 (2001) A838–A844. [5] W. Li, J.R. Dahn, D.S. Wainwright, Science 264 (1994) 1115–1118. [6] Battery Types and Characteristics for HEV, ThermoAnalytics, Inc., 2007, http://www.thermoanalytics.com (accessed 27.08.12). [7] M. Minakshi, P. Singh, Journal of Solid State Electrochemistry 16 (2012) 1487–1492. [8] I.D. Gocheva, M. Nishijima, T. Doi, S. Okada, J. Yamaki, T. Nishida, Journal of Power Sources 187 (2009) 247–252. [9] D. Kim, S.-H. Kang, M. Slater, S. Rood, J.T. Vaughey, N. Karan, M. Balasubramanian, C.S. Johnson, Advanced Energy Materials 1 (2011) 333–336. [10] R.D. Shannon, Acta Crystallographica Section A 32 (1976) 751–767.

[11] M.M. Doeff, M.Y. Peng, Y. Ma, L.C. De Jonghe, Journal of the Electrochemical Society 141 (1994) L145–L147. [12] A. Caballero, L. Hernan, J. Morales, L. Sanchez, J. Santos Pena, M.A.G. Aranda, Journal of Materials Chemistry 12 (2002) 1142–1147. [13] L. Athouel, F. Moser, R. Dugas, O. Crosnier, D. Belanger, T. Brousse, Journal of Physical Chemistry C 112 (2008) 7270–7277. [14] R. Chen, T. Chirayil, P. Zavalij, M.S. Whittingham, Solid State Ionics 86–88 (1996) 1–7. [15] M. Minakshi, N. Sharma, D. Ralph, D. Appadoo, K. Nallathamby, Electrochemical and Solid-State Letters 14 (2011) A86–A89. [16] M. Minakshi, D.R.G. Mitchell, P. Singh, Electrochimica Acta 52 (2007) 3294–3298. [17] M. Minakshi, D.R.G. Mitchell, Electrochimica Acta 53 (2008) 6323–6327. [18] J. Gopalakrishnan, Bulletin of Materials Science 7 (1985) 201–214. [19] S.P. Ong, V.L. Chevrier, G. Hautier, A. Jain, C. Moore, S. Kim, X. Ma, G. Ceder, Energy & Environmental Science 4 (2011) 3680–3688. [20] M.W. Chase, NIST-JANAF Thermochemical Tables, vol. 12, American Chemical Society, New York, 1998. [21] M. Minakshi, D. Appadoo, D.E. Martin, Electrochemical and Solid-State Letters 13 (2010) A77–A80. [22] F. Sauvage, L. Laffont, J.-M. Tarascon, E. Baudrin, Inorganic Chemistry 46 (2007) 3289–3294. [23] M. Minakshi, Aqueous Rechargeable Battery, PCT (WO/2011/044644) (2011).