When it comes to green energy, the intermittent nature of renewable sources like wind, solar, and tidal power presents a difficult problem for the electrical grid management. Peak energy production often doesn’t correlate well with peak energy demand, necessitating a means of storing excess energy when consumption is low. As renewable energy sources become more prevalent, and the need to curb fossil fuel emissions continues to increase, finding a new grid energy storage solution has never been more important. It is the final piece of technology required to bring about wide scale adoption of renewable energy sources like solar panels and wind turbines.
Molten salt batteries, especially liquid metal batteries, are increasingly gaining interest from the energy community as a grid energy storage solution for renewable energy sources. Combining high energy and power densities, long life times, and low cost materials, they have the potential to meet the unique demands of grid scale energy storage. A molten salt battery is a class of battery that uses a molten salts electrolyte. The components of molten salt batteries are solid at room temperature, allowing them to be stored inactive for long periods time. During activation, the cathode, anode and electrolyte layers separate due to their relative densities and immiscibility. The molten salt layer in the middle serves as an electrolyte with a high ionic conductivity, and is the medium through which the ionic species travel as the battery charges and discharges.
Molten Salt Batteries carry several inherent advantages over their solid state contemporaries. Since some (or all in the case of liquid metal batteries) of the components are liquid, the batteries possess a higher current density, longer cycle life, and simplified manufacturing scheme in large scale applications. Since no membranes or separator systems are involved, cycle life is higher and energy efficiency can be retained over a longer period of time. The grid scale energy storage company Ambri has previously shown that a lead-antimony and lithium liquid metal battery should retain 85 percent of its initial efficiency over a decade of daily charge/discharge cycles. Since the battery is essentially a container containing 3 liquid phases, construction is as simple as pouring the heavier metal into the bottom, the electrolyte in the middle, and the lighter electrode on top.The major drawback of this design is the high operating temperature required to keep the components in the liquid state. However in a grid scale application these elevated temperatures can easily be maintained using the heat generated during the charge and discharge cycles.
The first molten salt batteries actually weren't intended to operate for very long periods of time at all, but were instead used as single activation primary batteries for bombs and rockets. Invented by German WWII era scientist Georg Otto Erb, the first practical cells were called thermal batteries and while they were never used during the war, the United States Ordnance Development Division would eventually acquire the technology and use it to power rockets, bombs, and even nuclear weapons. These early batteries could last indefinitely (over 50 years) in the solid state while supplying a huge burst of power. Today thermal batteries are still used as the primary source of power for missiles like the AIM-9 Sidewinder, BGM-109 Tomahawk and the MIM-104 Patriot.
In 1966 Ford Motor Company invented the Sodium-Sulfur (NaS) liquid metal battery for electric vehicle application. The high power density and high energy capacity looked promising but the high operating temperature of 290-390 °C caused Ford to drop research and development. In 1983, Tokyo Electric Power Company (TEPCO) and Nippon Gaishi Kaisha (NGK) realized the potential for NaS battery system as a solution for grid storage and began research and development of the technology. In 1993 the first large-scale prototype of such a system was field tested at TEPCO’s Tsunashima substation. The system consisted of three 2 MW, 6.6 KV battery banks. This laid the groundwork for NGK/TEPCO consortium’s current line of grid storage NaS batteries, which produce 90 MW of storage capacity every year.
Meanwhile in Pretoria, South Africa, 1985, the Zeolite Battery Research Africa Project (ZEBRA) led by Dr. Johan Coetzer at the Council for Scientific and Industrial Research, invented the first sodium nickel chloride battery. It had a specific energy of 90 Wh/kg, a notably stable beta alumina solid electrolyte, and enhanced corrosion resistance over NaS. This design, while novel, has yet to see large scale commercial grid storage applications and remains a hot topic in battery research and development. They have however been deployed by FIAMM Sonick and used in the Modec Electric Van.
The thermal batteries used to power rockets and missiles are primary batteries and intended to deliver a high power over a short period of time, on the order of a few seconds to a little over an hour. There are generally two types of design. The first involves the use of a fuze strip, consisting of zirconium metal powder and barium chromate in a ceramic paper along the edge of heat pellets to ignite the burning process. The fuze strip is ignited by a squib which applies an electric current. The second involves a hole at the center of the battery stack that fills with a mixture of incandescent particles and hot gases upon electrically triggered ignition. This process is faster, on the order of tens of milliseconds versus the hundreds of milliseconds with the fuze strip design. Today’s thermal batteries utilize cathodes of made up of iron disulfide or cobalt disulfide with lithium silicon or lithium aluminum alloys. Older chemistries however made use of magnesium or calcium anodes and calcium chromate, tungsten oxide or vanadium cathodes. All of these designs used a molten salt electrolyte layer, normally consisting of lithium chloride and potassium chloride. Eutectic electrolytes have also used lithium bromide and potassium bromide have also been employed to increase cycle life.
Sodium-Sulfur (NaS) batteries are fabricated from inexpensive and abundant materials. The typical design involves a solid electrolyte membrane between the anode and cathode encased in a steel cylinder protected with a chromium and molybdenum interior. Molten sodium at the heart of the cell serves as the anode that donates electrons to the external circuit. The sodium core is encased in a beta-alumina solid electrolyte (BASE) cylinder which facilitates the movement of Na+ ions to the exterior sulfur electrode which serves as the cathode while preventing the two electrodes from shorting. NGK currently runs a line of successful grid storage NaS batteries, and is considered the world’s largest grid-scale battery supplier serving North America, Asia, and Europe. Each 1 MW x 6 MWh standard battery system contains 20 modules capable of supplying 50 kW AC in an operating temperature range of 300-350 °C.
Sodium Nickel Chloride (Na-NiCl2) Batteries also use a molten sodium core, but instead uses the nickel as the positive electrode in the discharged state and nickel chloride in the charged state. Both forms of nickel electrode are insoluble in their liquid states and a sodium conducting beta alumina ceramic is used as the separator. In place of the pure elemental sodium found in NaS batteries, a tetrachloroaluminate (NaAlCl4) core is preferred. Na-NiCl2 batteries are sometimes called sodium metal halide batteries and in addition to boasting long operating life spans, the ability to be assembled in the discharged state, and a safer chemistry than NaS. Normal operating temperature range of Na-NiCl2 batteries are in the 270-350 °C range, but one company, Sumitomo was able to develop a similar chemistry using a salt that melts at 61 °C and operates at 90 °C; they had initially slated commercial trials for late 2015 so only time will tell how they function in the market.
A liquid metal battery is a new type of molten salt battery designed for grid storage applications. First proposed by Donald Sadoway, a Massachusetts Institute of Technology (MIT) materials professor, in 2009, the liquid metal battery consisted of a current collecting container filled a molten antimony cathode at the bottom, a salt electrolyte for the middle layer, and liquid magnesium metal anode at the top. Magnesium was initially chosen for its low cost and low solubility with molten salt electrolyte, but the higher operating temperature of 700 °C prompted him to switch the chemistry to a Lithium based anode in 2011. The higher operating temperature was undesirable because it resulted in higher rates of corrosion, lowering total storage efficiency, and increasing costs over the battery’s lifetime. The current design thus uses a liquid lithium anode, a molten mixture of lithium salts as an electrolyte and a led antimony cathode capable of operating at a reduced temperature of 450 °C thanks to the lower melting points of the new electrodes.
Sadoway’s liquid metal battery is particularly attractive for grid storage applications because of the low cost of its materials and high energy efficiency. Antimony (Sb), currently priced at about $1.23 USD per mol, yields a high cell voltage when used with alkaline-earth negative electrodes. When coupled with a lithium (Li) electrode, the liquid metal chemistry can achieve an average cell voltage of 0.92 V when measured under a 200 mA /cm2 galvanostatic discharge. Li melts at a significantly lower temperature of 180 °C, exhibits low solubility with lithium halide salts, which subsequently reduces the probability of self discharge. This gives it an edge in energy efficiency over alternative sodium (Na) based molten battery chemistries.
While information on Ambri’s current Li and Sb-Pb application is scarce, Sadoway has publicly confirmed that it is similar to his initial magnesium—antimony (Mg||Sb) chemistry. In the original 2012 design, a negative Mg electrode and positive Sb electrode are separated by a molten salt electrolyte of the formula MgCl2-KCl-NaCl. Density differences form the three distinct layers of anode, electrolyte and cathode. When discharged, Mg undergoes oxidation reaction to yield Mg2+ which dissolves into the electrolyte and 2 free electrons which are released to an external circuit. The Mg2+ cations are simultaneously reduced to Mg and deposited into the Sb cathode where they combine together to form a Mg-Sb liquid metal alloy. During charging, the reverse occurs, and electric current drives Mg the Mg-Sb alloy and returns it to the top negative electrode as liquid Mg. This expanding and contracting of the liquid electrodes is unique to liquid metal batteries, and allows the electrodes to effectively be regenerated with each charge and discharge cycle effectively increasing lifetime of the battery.
Companies like Donald Sadoway’s Ambri, NGK, and Sumitomo are continuing to push the boundaries of molten salt chemistries as investors and the energy industry at large begin to recognize the importance of better batteries for grid scale energy storage. Ambri plans to ship six 10 ton prototypes to a pilot grid in Alaska, wind and solar plants in Hawaii, and a substation in Manhattan. NGK and Mitsubishi Electric Corp are building a 50,000 kilowatt battery system for Kyushu Electric Power Company in a bid to support a Japanese national initiative to switch to solar power. In a recent 2015 publication by MIT titled The Future of Solar Energy it was revealed that humanity today consumes 15 terawatts of power from a variety of energy sources. The report also revealed that solar technology has already reached the point that humans need to harness the energy of the sun and meet this energy demand. In Germany, Italy, and Spain, solar power has already achieved grid parity, with Germany in the lead generating 45% of its power from the sun. It has become clear that energy storage is the last piece of the puzzle for the world to fully reap the benefits of renewable energy sources. Money spent on investment into grid scale energy storage will carry more weight than money spent in better renewable energy technologies.