Elsevier

Journal of Energy Chemistry

Volume 66, March 2022, Pages 390-396
Journal of Energy Chemistry

Self-discharge mitigation in a liquid metal displacement battery

https://doi.org/10.1016/j.jechem.2021.08.015Get rights and content

Abstract

Recently, a disruptive idea was reported about the discovery of a new type of battery named Liquid Displacement Battery (LDB) comprising liquid metal electrodes and molten salt electrolyte. This cell featured a novel concept of a porous electronically conductive faradaic membrane instead of the traditional ion-selective ceramic membrane. LDBs are attractive for stationary storage applications but need mitigation against self-discharge. In the instant battery chemistry, Li|LiCl-PbCl2|Pb, reducing the diffusion coefficient of lead ions can be a way forward and a solution can be the addition of PbO to the electrolyte. The latter acts as a supplementary barrier and complements the function of the faradaic membrane. The remedial actions improved the cell’s coulombic efficiency from 92% to 97% without affecting the voltage efficiency. In addition, the limiting current density of a 500 mAh cell increased from 575 to 831 mA cm−2 and the limiting power from 2.53 to 3.66 W. Finally, the effect of PbO on the impedance and polarization of the cell was also studied.

Graphical abstract

  1. Download : Download high-res image (85KB)
  2. Download : Download full-size image

The addition of PbO acts as a mitigator to suppress self-discharge and enhance the performance of the liquid metal displacement battery.

Introduction

Climate change is one of the biggest challenges of the current era. Europe is committed to achieving the vision of a climate-neutral society by 2050 [1]. This transition motivates the search for innovative solutions that result in the mitigation of the adverse effects of energy consumption and conversion. This also drives how electrical energy is being produced, consumed, and stored. Implementation of electric vehicles is also a key step toward battling CO2 emissions. Batteries are the true enablers for this vision, but these need to be made safer, better performing, more sustainable, and affordable [2]. Batteries should exhibit high performance beyond their capabilities today. Furthermore, reliable batteries can provide greater resiliency to the existing grid while also lowering emissions [3]. Battery chemistries that have been deployed in stationary storage applications include lithium-ion, lead-acid, redox-flow, and high-temperature batteries such as sodium-sulfur and sodium-nickel chloride (ZEBRA) [4].

The transition to a more electricity-based society needs disruptive ideas that can enable the creation of sustainable batteries at the price point of the market. In line with these efforts Group Sadoway laboratory at MIT has reported the discovery of a new type of battery named liquid displacement battery (LDB) [3] which exploits the advantages of the sodium-nickel chloride electrochemistry without requiring the use of the β”-Al2O3 Na-ion conducting membrane. The fragility and brittleness of the β”-Al2O3 ceramic set severe design constraints and limit its performance. Also, its vulnerability to attack by transition-metal ions in a chloroaluminate melt is a concern. LDB has a new concept of a porous electronically conductive membrane instead of an ion-selective ceramic conductor. This membrane removes the barrier for researchers to choose only liquid Na for the negative electrode and solid transition-metal halides for the positive electrode [3].

The schematic of a liquid displacement battery based on the electrochemistry of Li|LiCl–PbCl2|Pb is shown in Fig. 1. The negative electrode is comprised of liquid Li-Pb, where Li is the active species and Pb serves as host metal. The eutectic composition of LiCl-KCl is used as a molten-salt electrolyte. Finally, liquid Pb is serving as the positive electrode and PbCl2 exists in the electrolyte in a dissolved state. The electrochemistry of Pb/Pb2+ displacement reactions in molten LiCl-KCl is explained [3] as2LiPbl+2Cl(LiCl-KCl(l))-2LiCl(LiCl-KCll)+2e-PbCl2(LiCl-KCl(l))+2e-Pbl+2ClLiCl-KCll-

This leads to the overall cell displacement reaction:2LiPbl+PbCl2Pbl+2LiCl

The use of membranes is a widely accepted concept in the field of batteries and fuel cells. Mostly, ion-exchange membranes are used as an electrolyte barrier between the half-cells. The membranes fabricated for this study enable ion exchange because of their porous nature and are also electronically conductive to allow for a “faradaic protection” reaction. When Pb2+ ions reach the lower surface of the membrane, they will be reduced faradaically into Pb metal. The lead droplets, formed at the surface of the membrane, sink back into the pool of Pb (l). This self-discharge reaction, which is caused by diffusion of lead ions from the bottom of the electrolyte towards the membrane, is of course detrimental. The membrane acts here simply as a lead barrier preventing irreversible lead transfer from the positive electrode to the negative electrode. LDBs built in the past exhibited a coulombic efficiency of 92% and an energy efficiency of 71% while operating at a current density of 150 mA cm−2 at a temperature of 410 °C. The loss in coulombic efficiency is not caused by irreversible side reactions but rather simply by the self-discharge reaction, explained above [3]. LDBs are promising but require further mitigation strategies to reduce self-discharge. The latter is the scope of this study.

While the reported battery chemistry is relatively recent, the electrochemical reactions explained above are well established and have been used for lead deposition from molten chlorides. The characteristics of the mixtures of molten salts and metals are reported elsewhere [5]. The diffusion coefficient of lead ions in molten chlorides has been reported to be 2 × 10−5 cm2 s−1; the cathodic deposition of lead was found to be diffusion controlled [6], [7], [8], [9] for the dilute solution of PbCl2 in molten KCl-LiCl at 400 °C. A formulation based on applying Faraday’s law and species transport equations led to a relationship highlighting the direct proportionality of the self-discharge current density to the diffusion coefficient. The mathematical formulation is given in the supplementary information.

Different remedies can be used to reduce self-discharge such as increasing the thickness of the electrolyte (salt) layer, reducing the solubility of PbCl2 in the electrolyte, lower the operating temperature, freezing the electrolyte during cell idling, binding of PbCl2 to some other chemical moiety inside the Pb, or by using an electrolyte featuring a miscibility gap. The idea of this study is to convert PbCl2 into another complex compound leading to a reduction of the diffusion of the Pb-ions.

Section snippets

Chemicals

Lead (99.99%, anhydrous, Sigma Aldrich) was used in the electrodes. The electrolyte was prepared using the eutectic composition for a mixture of lithium chloride (LiCl, 99.9% anhydrous supplied by Alfa Aesar) and potassium chloride (KCl, 99.9% anhydrous supplied by Alfa Aesar). Membranes were prepared using titanium nitride (TiN) powder (Alfa Aesar, 99.7%, <10 μm) and magnesia (MgO) nano-powder (Inframat Advanced Materials). Lead-oxide was also obtained from Alfa Aesar.

Preparation of membrane

The porous TiN membrane

Mitigation of self-discharge

The activation of the liquid metal electrodes and the molten salt electrolyte was performed through charge–discharge cycling (CDC) until the cell reached the operating capacity of 500 mAh. The cycling was done step-wise by operating the cell at the capacities of 25, 250, and 500 mAh as shown in Fig. 3 (i.e., CDC at 50, 100, and 200 mA for charging times of 0.5, 2.5, and 2.5 h, respectively). During the activation phase, the coulombic efficiency at 25 mAh capacity increased from 25.5% to 90.1%

Conclusions

A formulation involving Faraday’s law and the Nernst-Planck equation leads to a relationship that indicates that the self-discharge current density is directly proportional to the diffusion coefficient. It directs this study to the solution of adding lead oxide (PbO) to the electrolyte so as to reduce the diffusion coefficient as well as the Pb ion concentration by initiating two additional side reactions. The limiting current density and power were improved from 575 to 831 mA cm−2 and from

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was financially supported by the research unit of Group Sadoway Laboratory, Department of Materials Science and Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA, 02139-4307, United States. Kashif Mushtaq is grateful to the Portuguese Foundation for Science and Technology (FCT) for his Ph.D. scholarship (PD/BD/128041/2016). This work was also financially supported by the Base Funding (UIDB/00511/2020) of the Laboratory for Process Engineering,

References (12)

  • W. Zhang et al.

    J. Energy Chem.

    (2020)
  • V.M.B. Nunes et al.

    Appl. Energy

    (2016)
  • M. Urquidi-Macdonald et al.

    Electrochim. Acta

    (1990)
  • K. Mushtaq et al.

    J. Energy Storage

    (2021)
  • H. Yin et al.

    Nat. Energy

    (2018)
  • J. Figgener et al.

    J. Energy Storage

    (2020)
There are more references available in the full text version of this article.

Cited by (7)

  • Anode reaction mechanisms of Na|NaCl-CaCl<inf>2</inf>|Zn liquid metal battery

    2022, Journal of Energy Chemistry
    Citation Excerpt :

    Conceptually, the thermodynamic driving force for cell discharge can be interpreted as emanating from a strong interaction between A and B, in which A’s activity can be extremely low (aA(in B) can be as low as 10−10) [11]. Various types of LMBs have been developed to date, including Na||Sn [16], K||Hg [17], Na||Hg [18–20], Na||Bi [21], Mg|| (Mg-Sb) [12], Li|| (Li-Sb-Pb) [22], Li|| (Li-Sb-Sn) [23], (Li-Pb) ||Pb [24,25], (Ca-Mg) || (Ca-Mg-Bi) [26], Sn||Bi [27], Li|| (Li-Sb-Bi) [28], Li|| (Bi-Pb) [29], (Zn-Sn) ||Bi [30], Li||Te, Na||Te, K||Te, and Li|| (Te-Sn) [31] LMBs. The full liquid structure enables the LMB to work as high as 1300 mA cm−2 current density with a coulombic efficiency of 100% [32,33].

View all citing articles on Scopus
View full text