Energy storage update

This category covers a number of chemistries, all based on lithium. Lithium batteries have become the most common for on-grid domestic energy storage systems due to their high efficiency, long life, low maintenance and deep-cycle capabilities.

All lithium chemistries have high ‘specific energy’ (the energy stored per unit mass, in watt-hours per kilogram), usually above 100 Wh/kg, making them compact and light. They can be deeply cycled without great reductions in lifespan (this varies depending on the lithium chemistry, cell design etc) and charged and discharged rapidly, with high discharge rates having little effect on capacity. They also have high charge and discharge efficiency, stable output voltage during discharge and are effectively fully sealed, so no off-gassing occurs unless they are heavily overcharged.

They do have some disadvantages, such as higher initial cost compared to lead-acid (although the gap is closing), and the need for accurately controlled charging and over-discharge protection. There is also a higher fire risk for some chemistries. Lithium batteries should be used with a battery management system (BMS) to prevent damage (usually built-in in packaged systems).

Lithium battery recycling facilities are still limited in Australia, so lithium batteries can be more difficult to recycle than lead-acid.
Lithium iron phosphate, commonly known as LiFePO4 or LFP, is perhaps the most popular lithium chemistry, often used for large-scale energy storage system and also in pre-packaged storage systems, such as the Fronius Solar Battery and the BYD B-Box LV. They can have extremely long cycle lifespans (up to 10,000 cycles), especially if not deeply discharged each cycle, and LiFePO4 cells are resistant to fire, even when extremely abused, such as being damaged or short circuited.

Lithium titanate are robust, long-lived lithium batteries; they’ve been around for decades, but are not common in consumer-grade devices. They are normally used in industrial systems where reliability and long lifespan are paramount, especially with high power transfer requirements, as they can charge and discharge at very high rates compared to cells such as LiFePO4. Examples include fast-charge electric buses, grid backup and military applications. A few consumer-grade electric vehicles also use LTO batteries, such as the now-unavailable Honda Fit EV.

Lithium manganese oxide (LiMn2O4, Li2MnO3 or LMO) batteries are mostly used in consumer products, but they have also made their way into energy storage systems such as the Magellan HESS and the Varta Pulse 3. They feature a low internal resistance and so allow fast charging and discharging. The chemistry is often combined with lithium nickel manganese cobalt oxide (NMC, used in the Tesla Powerwall 2, Sonnenschein @ Home Lithium and LG Chem Resu, among others) to improve lifespan and energy density; there are many variations of these chemistries. The resulting high charge and discharge capabilities and good energy densities, plus their good fire resistance, make them popular for power tools and electric vehicles. The materials used in their manufacture are mostly common and cheap; reduced cobalt content reduces the manufacturing cost.

This chemistry was, up until a few years ago, the mainstay of the home energy storage industry. Lead-acid batteries have been in use for over a century and are a tried and proven technology. Although still common in systems that are assembled from individual cells, lead-acid is being replaced by lithium chemistries in on-grid domestic energy storage systems.

Lead-acid batteries are readily available, have reasonable resistance to overcharging, are lower cost than many other technologies and are almost 100% recyclable.

Disadvantages include that they can’t be regularly deep cycled without reducing lifespan and their low energy density makes them heavy for a given battery capacity. In addition, their primary reactive materials are toxic and corrosive, and they can produce explosive hydrogen gas when charging. They shouldn’t be stored partially discharged and must be fully charged regularly to prevent sulphation (where permanent lead sulphate crystals form on the plates). They also suffer from the Peukert effect, where the effective capacity reduces as discharge rate increases.

Flooded-cell lead-acid batteries are generally reliable, but require the highest maintenance of all the lead-acid battery types, including electrolyte level checking and top-up and periodic ‘equalise charging’. They can also produce explosive hydrogen during normal charging cycles, so must be installed in ventilated enclosures.

Sealed lead-acid batteries are effectively sealed, with the electrolyte either contained in a fibreglass separator (absorbent glass mat, or AGM, batteries) or formed into a thick gel with the addition of a silica-based gelling agent (gel cell batteries). They are safer to handle and transport than flooded cell batteries and have little-to-no hydrogen off-gassing during regular charge cycles, no need for equalising charges and no need to check or top up electrolyte. Lifespan depends greatly on the battery design and how they are treated—solar-grade gel cells can easily last as long as equivalent flooded lead-acid cells.

A variation of lead-acid technology is the UltraBattery, which effectively combines a sealed lead-acid battery and carbon electrode ultracapacitor into each cell, enabling the batteries to provide higher currents and withstand deeper discharges than regular lead-acid batteries. They can have a long cycle life, can be stored in a partial state of charge without damage and can have greater charging efficiency.

Flow batteries are quite different from conventional batteries. Instead of solid plates separated by a liquid electrolyte, they use two liquids as the electrodes, separated by a membrane. They have the advantage that capacity can be increased simply by adding larger electrolyte tanks. The only flow battery available for domestic storage applications, the Z-Cell Redflow zinc-bromine battery, comes in 10 kW modules which can be added together to make batteries as large as required.

They have a long cycle life, can be shut down for extended periods at any state of charge, can be cycled to 100% depth of discharge for every cycle, are not a fire risk and can operate in temperatures up to 50 °C.

Their main drawback is the relatively low charge/discharge efficiency and the need for a regular maintenance cycle, during which time the battery is offline. This problem can be mitigated by scheduling the maintenance cycle at the time of least need, such as early in the morning.

While most batteries use either solid electrodes and liquid electrolyte, or liquid electrodes with a solid separator (as used in flow batteries), there is one battery type that has, until now, not been developed commercially—a completely liquid metal battery.
The liquid metal battery from Ambri uses two metals of different densities, one heavy, one light, separated by a molten salt with a density between that of the two metals. The different layers stay separate due to immiscibility (the inability to mix); there are no membranes or separators required.

Ambri originally used magnesium as the negative electrode and antimony as the positive electrode. As the cell discharges, electrons flow from the magnesium layer, and magnesium ions migrate through the salt layer to form a magnesium-antimony alloy. When recharging, the magnesium is stripped from the alloy and reforms the top magnesium layer.

Testing on this self-heating cell chemistry found almost no capacity reduction after 2000 charge/discharge cycles, with an expected lifespan over 15 years and tens of thousands of cycles, and the cells have proved to be easily scalable to 1 kWh and larger per cell. However, Ambri is now commercialising a different (proprietary) chemistry with a lower operating temperature, higher voltage and lower cost, which should result in batteries with a better cost per unit of storage.

Just when the Ambri battery will hit the market is unknown, but it should find uses in grid-level storage and even possibly domestic storage systems—if the price is right.

Other flow batteries

The Redflow battery is not the only flow battery around. Vanadium redox flow batteries have been in development in Australia and overseas for many years, with a number of grid-scale installations completed. Vsun Energy has installations in Australia and quite a few overseas, including in the USA, Germany, Japan, China, Indonesia, India, the UK, Italy and The Netherlands.

Another flow battery chemistry is zinc redox, and ViZn Energy Systems has recently secured a new round of funding after suspending operations earlier this year. In December 2017 they secured a contract to install a 400 kW, three-hour zinc flow battery at CMI’s headquarters in Seraing, Belgium, to be integrated with 2 MW of solar PV.

An interesting recent development comes from Stanford University, where they have come up with the design for a liquid metal flow battery. The negative electrode is made of an alloy of sodium and potassium which is liquid at room temperature. The separator, or membrane, is made of a ceramic consisting of potassium and aluminium oxide. For the positive side of the battery, the researchers experimented with four different liquids, but the water-based liquids quickly degraded the membrane. They are now looking for a non-water-based option to improve the battery’s performance—probably a good idea, given the explosive reactions that occur when sodium and potassium come into contact with water! www.stanford.io/2MaS4gr

One of the biggest issues with lithium batteries is the use of cobalt in many lithium chemistries. Cobalt is expensive, with prices tripling in the last couple of years, and more than half of the world’s cobalt coming from the politically unstable Democratic Republic of the Congo, where miners usually face poor working conditions and are often exploited.

Recognising these and other issues, a number of manufacturers are working to reduce or eliminate cobalt from their batteries altogether. These include Panasonic, which aims to get the cobalt out of their nickel-cobalt-aluminium batteries (used in Tesla’s electric vehicles); they have already reduced the amount of cobalt required per vehicle by almost 60% in the last six years (www.bit.ly/2LZhzBm). According to Tesla CEO Elon Musk, the ultimate goal is for Tesla vehicles to use no cobalt at all in their battery packs.

New York startup Conamix is aiming to develop their own cobalt-free electrode materials that can be used in existing battery manufacturing. Conamix has already attracted investment from the ‘deep diligence’ investment company +Volta Energy Technologies, so they must be onto something with their technologies.

BYD in China is also reducing the amount of cobalt in its batteries, to a nickel-cobalt-manganese ratio of 8:1:1 (current batteries generally use a 5:2:3 or 6:2:2 ratio), and this ratio is expected to account for around 57% of all electric vehicle batteries by 2030, according to Bloomberg (bloom.bg/2ndg5Et).

In the last battery technology update in Renew 137 we looked at a number of newer technologies. But have they been successful, and where are they now? Will they change the storage industry, or were they just pipe dreams?

The first is the Saltwater batteries from Aquion Energy. Unfortunately, Aquion went into receivership not long after issue 137 came out, but unlike most companies that fail, they were rebooted with new investment funds and are in the process of redesigning and improving their batteries—although they seem to have been doing this for the last two years now. You can buy Aquion batteries from a few suppliers that have stock of the old batteries, but the new restructured company is not supporting warranties for those batteries. Whether we will see the new and improved Aquion batteries and whether they will address the issues of low power output and low charge cycle efficiency remain to be seen.

While Aquion’s fortunes have been a tad variable, one Australian company has done a lot better. Redflow Limited makes zinc-bromine flow batteries in commercial and domestic designs, all based on versions of the same 10 kWh flow battery module. While they had a slow start, installations have been progressing, with numerous domestic and a small number of commercial installations so far, including for use in a digital TV network in Fiji, a factory in South Africa and a number of domestic and rural installations around Australia. Note that Redflow uses Thailand as its manufacturing base, rather than Australia.

Flywheel batteries

Flywheels were once touted as the solution to rapid response energy storage for on-grid applications. While they haven’t been as successful as manufacturers would have liked, there are some flywheel energy storage systems available, such as the Amber Kinetics M32 (available in Australia through Greenfields Energy in Adelaide), with a rated output of 8 kW and storage of 32 kWh. SA Water will be using 128 kW of M32 flywheel batteries at the Hope Valley reservoir this year. According to SA Water, the project is on schedule for installation in coming months.

SolidEnergy

This MIT spinoff battery manufacturer promises the lightest, most energy-dense lithium batteries in the world with their redesigned fire-resistant lithium metal cells. So far they are only producing their Hermes cells at around 5000 units per month, so availability is limited to say the least. However, at a specific energy of around 450 Wh/kg and energy density of around 870 Wh/L, they are the most energy-dense cells we have seen. Let’s hope they follow through with their other cell designs in the next few years—they could double the range of the average EV, if the prices are low enough.

Concepts that failed

US startup LightSail planned to use their own compressor and carbon-fibre storage tank designs to provide above-ground large-scale compressed air energy storage. However, they failed to address the problems that previous companies such as SustainX had run up against, including the low energy efficiency of compressed air storage due to high thermal losses (air gets hot when you compress it), and ceased operations in December 2017 after burning through a cool US$80m.

As battery manufacture ramps up, the need for greater quantities of raw materials is steadily increasing. Lithium batteries are normally made from new (virgin) materials, but there is movement in the recycling of existing lithium batteries to recover the raw materials they contain for reuse in new batteries.

Traditionally, the recovery methods have been expensive and energy-intensive, hampering the lithium battery recycling industry. However, some interesting work is being done in this area, with one university team taking an old-school approach.

Michigan Technological University researchers are using century-old mining techniques to recycle lithium-ion batteries at low cost. Research leader Lei Pan, an assistant professor of chemical engineering, also has a graduate degree in mining engineering. He thought that methods that are normally used to separate metals from ore could be applied to lithium battery recycling, and suggested to his students to look at this further.

The team used these techniques, such as simple gravity separation and froth flotation, to separate everything in the battery, including the casing, metal foils, the anode and cathode, and the lithium-metal oxide, with considerable success. The end result was a range of materials, including graphite, lithium and cobalt, which were cheaper to produce than virgin materials, but just as suitable for new battery manufacturing. The processes used are the cheapest and most energy-efficient currently available, and the processing equipment is readily available. www.bit.ly/2AKWUvs

Closer to home, battery recycling is starting to ramp up. Envirostream Australia is a company created in 2017 by PF Metals to tackle the increasing waste problem of discarded batteries of all types. The factory takes batteries from EVs, phones and computers, power tools, energy storage systems, almost anything that uses a battery, and can deal with lithium, nickel metal hydride and nickel cadmium, alkaline single use and lead-acid battery types. The batteries are sorted into type and then processed, with up to 80 tonnes of batteries per month being recycled into their base elements for reuse. All batteries are processed in Victoria at the New Gisborne plant, with up to 95% of the materials in the batteries being recovered. Battery drop-off locations can be found on the Envirostream website.

• ‘The lowdown on battery technology’ in Renew 137
• Energy Storage Buyers Guide in Renew 141
• Clean Energy Council storage system FAQ and guide
• Solar Quotes battery storage comparison table
• batteryuniversity.com