The rise of renewable energy has exposed a new problem: energy storage.
Solar and wind can generate very cheap electricity, but they're intermittent. For entire grids to run on renewables, enormous amounts of storage are needed to avoid blackouts.
The two main options, pumped hydro and lithium-ion batteries, each have their drawbacks, such as high costs.
Fortunately, there may be a third option.
A type of battery invented by an Australian professor in the 1980s has been growing in prominence, and is now being touted as part of the solution to this storage problem.
Called a vanadium redox flow battery (VRFB), it's cheaper, safer and longer-lasting than lithium-ion cells.
Here's why they may be a big part of the future — and why you may never see one.
'We were 20 years too early'
In the 1970s, during an era of energy price shocks, NASA began designing a new type of liquid battery.
The iron-chromium redox flow battery contained no corrosive elements and was designed to be easily scalable, so it could store huge amounts of solar energy indefinitely.
Several years later, in Australia, a young chemical engineer at UNSW in Sydney named Maria Skyllas-Kazacos started studying these new kinds of flow batteries.
Within years, she and her research team developed another kind of flow battery, one that used vanadium instead of iron and chromium.
Like the NASA design, it was safe, reliable, long-lasting and easily scalable.
Unfortunately, there wasn't much of a market for energy storage.
"We understood at the time we were 20 years too early," recalled Professor Skyllas-Kazacos, who is still with UNSW.
"We always knew the big applications would be renewables and solar but it took a lot longer for the market to develop than we expected."
UNSW filed a patent in 1986 and a 200kW/800kWh system installed in Japan was the first large-scale implementation of the technology.
In the late 1990s, UNSW sold the patent to an Australian company, Pinnacle Renewable Energy, which failed to commercialise the product. That original patent passed through a series of corporate owners before expiring.
Still, the market for energy storage didn't exist.
Then, suddenly, everything changed.
One turning point, Professor Skyllas-Kazacos said, was the 2016 South Australian blackout.
"Elon Musk came along and said 'I can build a battery in 100 days'. And everyone realised you can build big batteries."
But what type of battery should be built? Lithium-ion batteries had a big head start as the US government had poured billions into funding Tesla "gigafactories" able to produce the batteries at relatively low cost.
"They had huge capacity available to manufacture lithium batteries," Professor Skyllas-Kazacos said.
And so, almost by default, lithium-ion became the technology of choice for grid energy storage.
Now, however, that's begun to change.
Cheaper, safer, more recyclable
When a commercial district in Trondheim, Norway, recently commissioned battery energy storage, it made an unusual choice.
Instead of ordering lithium-ion, it went with VRFB.
One of the main reasons for this was the lower cost, said Besart Olluri, co-founder of the Norwegian company that installed the battery, Bryte Batteries.
Another related reason was that, with proper maintenance, the battery could technically last forever.
"You get huge benefits both in terms of environment but also lifetime costs," Mr Olluri said, speaking from Trondheim.
"Even after 20 to 30 years of lifespan, you're able to easily recycle or refurbish the system to make it into a new one."
This remarkable property of VRFB has seen them being described as the next big technology for large-scale storage.
Dozens of companies around the world are now manufacturing and installing megawatt-scale VRFB.
Late last year, renewables developer North Harbour Clean Energy announced plans to build what would be Australia's largest VRFB — with 4 megawatts of power (the amount of energy that can flow in and out of the battery in any given instant) and 16 megawatt-hours of capacity.
Along with a joint venture partner, they also promised to build a VRFB assembly and manufacturing line in eastern Australia to "meet GWh demand for long-duration energy storage in the National Electricity Market".
How they work
To understand why VRFB have been getting this attention, we need to quickly brush up on how batteries work.
A battery is a device that stores chemical energy and converts it to electrical energy. It does this through chemical reactions that create a flow of electrons from one material to another.
This flow or "electric current" is what we call electricity.
Beyond this, different kinds of batteries work in different ways.
In a lithium-ion battery, energy (in the form of lithium ions) is stored in the solid anode and cathode. When you charge your phone, the charger passes current to the battery, and lithium ions move from the cathode to the anode. When you unplug, this process is reversed.
But there's a problem: the interaction of lithium ions and electrodes gradually degrades the battery.
As a result, your phone battery has an average lifespan of two to three years, or 300-500 charge cycles, and holds less charge as it ages.
VRFB systems sidestep this problem.
In theory, they can be charged and discharged an unlimited number of times with no capacity degradation, said Chris Menictas, head of the Energy Storage and Refrigeration Lab at UNSW and one of Professor Skyllas-Kazacos's former students.
"The electrolyte never degrades," he said.
"There's been demonstrations ... of flow battery systems that can operate for hundreds of thousands of cycles at high efficiencies."
VRFBs do this through taking advantage of a special property of vanadium: it has four different stages of oxidation, meaning the same element can have four different charges.
As with other batteries, charge is created through a chemical reaction.
But in this case, the reaction is between differently charged ions of the same element, so the electrolyte doesn't degrade.
Plus, there are some other advantages. Unlike lithium-ion batteries, VRFB can be completely discharged.
They can store energy for long periods with no ill effects.
Because of the liquid electrolyte, they're also less likely to catch fire.
Scaling up capacity is also easier than with a lithium-ion battery. Instead of having to connect together millions of small self-contained cells, you simply get a bigger tank of electrolyte.
Finally, vanadium is more abundant in the Earth's crust than lithium and therefore less vulnerable to supply bottlenecks.
"I think it's a very exciting time," Dr Menictas said.
"Large-scale batteries are required more and more and I think vanadium is one of the leading technologies."
But there's a catch ...
VRFB are less energy-dense than lithium-ion batteries, meaning they're generally too big and heavy to be useful for applications like phones, cars and home energy storage.
Unlike lithium-ion batteries, they also have moving parts: the pumps that produce the flow of electrolyte solution.
And although vanadium is more abundant than lithium, it's expensive to extract. Most of the world's supply is used in refining steel, so its price tends to be volatile, increasing in response to demand for steel.
As a result, vanadium batteries currently have a higher upfront cost than lithium-ion batteries with the same capacity.
Since they're big, heavy and expensive to buy, the use of vanadium batteries may be limited to industrial and grid applications.
According to Dr Menictas, VRFB batteries work out cheaper than lithium-ion for these applications.
"As you start increasing the storage time, vanadium becomes cheaper," he said.
"At more than three hours' storage, vanadium is cheaper than lithium-ion."
Storage time (or capacity) is a function of the amount of stored electrolyte, or the size of the tanks. Since VRFBs are most cost-efficient with size, they're probably going to be very big.
That's why you may never see one. They won't be in cars or phones, but probably housed out of sight, beside solar farms and within substations.
How much storage do we need?
The National Electricity Market (which supplies the grid for most of the country, except WA and the NT) has about 1.5GW of batteries and pumped hydro.
By 2050, the Australian Energy Market Operator says, it'll need about about 46GW/640GWh.
By comparison, the Victorian Big Battery is 300MW/450MWh. So the need for storage is roughly equivalent to adding more than five very large batteries per year for the next 27 years.
Andrew Blakers, director of the Australian National University Centre for Sustainable Energy Systems, estimates the need for storage to be even greater: about 50GW/1,000GWh of storage.
This is because electricity production will have to triple as the economy is decarbonised and sectors like ammonia production and steel-making are electrified.
"We have to do this inside 20 years," he added.
VRFB has the potential to store energy at a scale that would dwarf today's largest lithium-ion batteries, Professor Skyllas-Kazacos said.
"They are ideal for massive-scale energy storage," she said.
"They can be made in gigawatt-hours."
Almost 40 years since leading the team that invented the VRFB, Professor Skyllas-Kazacos can see the technology's potential is finally being realised.
And after failing to commercialise homegrown tech, Australia may soon be manufacturing VRFB in large quantities.
"It would have been wonderful if we could have developed it in Australia sooner," Professor Skyllas-Kazacos said.
"[But] all we know how to do is construction and mining.
"[The companies] couldn't see beyond that. They couldn't see the big picture."
