Dr. Gunter Gerbeth
Director Institute of Fluid Dynamics
Phone: +49 351 260 - 3480, 3484
Fax: +49 351 260 - 3440

Dr. Gerd Mutschke
Institute of Fluid Dynamics
Phone: +49 351 260 - 2480
Fax: +49 351 260 - 12480

Petra Vetter
Secretary Institute of Fluid Dynamics
Phone: +49 351 260 - 3480
Fax: 13480, 3440

Get more information

Project B3: Liquid Metal Batteries

PI: Tom Weier (HZDR)

Partners: HZDR, TUI, CU

1. Scientific case for the project

1.1 Background

With the growing role of solar and wind power in the German energy landscape, large scale storage becomes a key enabler for a functional power grid. In this setting, cost per unit stored energy is the main criterion for a successful technology. Liquid Metal Batteries (LMBs) are a very promising concept for economic storage [1]. This type of battery was investigated already in the 1960s [2] mainly in the context of thermally regenerative electrochemical systems (TRES). The renewed interest in LMBs is accompanied by massive funding from the side of the US government [3] as well as the industry. However, the operability of LMBs has so far been demonstrated for very small units only while the expected economic benefits rely on large scale implementations. LMBs operate at elevated temperature and consist of three horizontal liquid layers in stable density stratification (Fig. 8). Adjacent layers are immiscible in each other. Since the batteries are virtually self-assembling, they allow for an extremely simple and therefore cost-efficient design. The liquid-liquid interfaces show very fast charge transfer kinetics facilitating high current and power densities. Transferring the attractive concept to an industrial scale means to first understand and then address a number of issues mainly rooted in fluid mechanical instabilities. These instabilities can destroy the stable density stratification thereby causing battery failure.
Depending on the batterie's cross section, which is determined by the achievable current density and the projected charge/discharge time, different instabilities will occur. Tall cells are most susceptible to the Tayler instability (TI). This topic is currently studied at HZDR [6] in frame of the "Helmholtz-Initiative für mobile und stationäre Energiespeichersysteme" (EWI).
Image of a tall LMB stabilized against the Tayler instability by a return current
Fig. 8: Tall LMB stabilized against the Tayler instability by a return current

By contrast, material combinations recently favoured for LMBs support moderate current densities only, calling for shallow cells with large interfacial area. In shallow cells, the role and the characteristics of the TI are still to be determined. In addition, shallow cells are dominated by the interfacial regions. It is to be expected that interface instabilities related to those observed in aluminium reduction cells (ARCs) have to be taken into account as well since they can lead to a direct contact of anode and cathode material. Another fluid dynamic aspect is solutocapillary and thermocapillary Marangoni convection, which will likely be driven by temperature and concentration differences building up during charging and discharging.

1.2 Most important goals of the planned work

To function properly, LMBs rely crucially on the persistence of the density stratification. A thorough understanding of the interplay of Lorentz forces and stratification in LMBs is therefore essential in order to assess their scaling behavior. Consequently, one important goal of the project is to study the fluid dynamics of layered systems carrying an electric current normal to the interfaces. In part, these investigations can draw on the existing body of research on ARCs. ARCs are known to develop long wave interfacial instabilities due to the Lorentz force mediated coupling of gravity waves [4]. In addition, theory [5] predicts that three-layer systems as LMBs are susceptible to short wave instabilities as well. This theory, originally developed for vacuum arc remelting, should be adapted to the specifics of LMBs. In this context, it needs to be emphasized that tolerable wave amplitudes in LMBs are much lower than those in ARCs. While the cryolite layer depth in ARCs is in the order of a few centimeters, only a few millimetres of electrolyte are allowable in LMBs due to the low cell voltage.
Role and features of the TI in shallow basins will be studied using an in-house code [6]. Experimentally, velocity distributions in the cells and the position of the interfaces shall be determined using Ultrasound Doppler velocimetry (UDV) and an adapted version of the Contactless Inductive Flow Tomography (CIFT). Flow measurements will profit greatly from a close collaboration with the YIG.
Following the characterization of the different instabilities and their influence on battery operation, countermeasures have to be devised. Such countermeasures may consist in geometry modifications and/or a redistribution of the magnetic field as exemplified with the central bore and the axial return current in case of the tall cell LMB [6], see Fig. 8.

It goes without saying that electrochemical problems are at the heart of any kind of battery development. Active materials and electrolytes have to be identified and characterized, taking into account the solutions developed for TRES [2] and the new results from MIT [1]. This part of the project will be conducted at TUI and has two main objectives: 1) development of a low temperature system usable for basic research allowing, by virtue of its moderate temperature, for easy access by a variety of flow measurement techniques. 2) Screening of candidates for active materials and electrolytes in order to find combinations with low cost, low melting temperature and high specific power and energy.

2. Existing competencies and infrastructure

HZDR’s competencies relevant to the present proposal lie in experimental and theoretical knowledge of MHD instabilities and practical experiences with a wide variety of measurement techniques for liquid metals, including own developments like CIFT. Several large scale facilities for liquid metal experiments (LIMMCAST, see C3) have been built and are operated at HZDR or nearing construction (DRESDYN, see A2), among them a TI experiment demonstrating for the first time TI in a liquid metal. Furthermore, Marangoni convection in molten salts has been investigated by particle image velocimetry (PIV) [7], a technique used routinely for electrolyte flows at HZDR. Detailed experience on various interfacial instabilities at liquid-liquid interfaces exists at CU [8,9]. The group of Prof. Bund (TUI) has a strong background in fundamental and applied electrochemistry. The present project will substantially profit from the group's competencies in the field of electrochemical energy storage and conversion.

3. Resource planning and Budget Justification

One PhD student at HZDR will be responsible for the experimental and part of the numerical work on the fluid mechanics of shallow LMBs. At TUI, another PhD student shall investigate the electrochemical aspects of LMBs.

Links: The project is closely related to projects A2 and A3.


[1] D.J. Bradwell, H. Kim, A.H.C. Sirk, and D.R. Sadoway, 2012, Magnesium-antimony liquid metal battery for stationary energy storage. J. Am. Chem. Soc., Vol. 134, 1895-1897.
[2] E.J. Cairns, H. Shimotake, 1969, High-temperature batteries. Science, Vol. 164, 1347-1355.
[3] D. Kramer, 2009, DOE names winners of long-shot energy research grants, Physics Today, Vol. 62(12), 26-27.
[4] P. Davidson, 2001, An Introduction to Magnetohydrodynamics. Cambridge University Press.
[5] A.D. Sneyd, 1985, Stability of fluid layers carrying a normal electric current. J. Fluid Mech., Vol. 156, 223-236.
[6] F. Stefani, T. Weier, T. Gundrum, G. Gerbeth, 2011, How to circumvent the size limitation of liquid metal batteries due to the Tayler instability. Energ. Conv. Manag., Vol. 52, 2982-2986.
[7] A. Cramer, S. Landgraf, E. Beyer, G. Gerbeth, 2011, Marangoni convection in molten salts: Physical modelling toward lower Prandtl numbers. Exp. Fluids, Vol. 50, 479-490.
[8] A. Pedchenko, S. Molokov, J. Priede, A. Lukyanov, P.J. Thomas, 2009, Experimental model of the interfacial instability in aluminium reduction cells. Europhys. Lett. 88, 24001-5.
[9] J. Priede, 2011, Edge pinch instability of oblate liquid metal drops in a transverse AC magnetic field. J. Fluid Mech. 676, 218-236.