Fitness trackers powered by sweat. Bluetooth earphones than run on atmospheric humidity. Stick-on insulin sensors fuelled by thin air.
These are some of the potential applications of what have been called "self-charging batteries", which harvest electrical energy from moisture.
The idea has been around for years, but early prototypes could not generate usable quantities of electricity.
Then earlier this year, a team at the University of New South Wales announced they had found a way to significantly boost electrical output — using atom-thick layers of graphene oxide in a pancake stack a fraction of a millimetre thick.
When several of these small, bendable moisture-electric generators (MEGs) were assembled in a series, they could power a pocket calculator.
Last month, Strategic Elements, an Australian company that had partnered with the UNSW researchers, said it was taking the tech to market.
It plans to have a demonstration version of a moisture-powered electronic skin patch by the end of the year.
So are self-charging batteries just around the corner — or are they too good to be true?
Where does the power come from?
Batteries that charge themselves may seem an impossible idea: like flying by lifting yourself up by the soles of your shoes.
Dewei Chu is the leader of the UNSW Nanoionic Materials Group.
His team co-authored the paper describing the new graphene-oxide MEGs, which was published in the journal Nano Energy in April.
MEGs, he says, work by separating positive and negatively charged particles.
"If we can have more protons on one side than the other, that is an ion gradient, and you can generate voltage and current."
To understand what he means, imagine a solution full of positively and negatively charged particles.
Now, imagine you introduce a membrane that only allows the positively charged particles through.
The unequal distribution of negatively and positively charged particles across the membrane causes an electrochemical gradient, which is what generates a voltage at the electrodes.
With the UNSW MEG design, these positively charged particles are hydrons, or hydrogen ions.
So the batteries aren't quite charging themselves — they're really converting chemical potential energy into electrical energy.
But where do the hydrons come from?
They come from the air — from water vapour.
Professor Chu and his team developed a way of printing very thin layers of graphene oxide nanomaterial.
One property of this material is that it's water-loving, or hydrophilic. Airborne water molecules easily bond with the surface.
On the surface, the water (H2O) breaks up, releasing hydrons.
The trick, says Professor Chu, has been to create a large enough reservoir of hydrons to generate usable amounts of electricity.
That's where the nanomaterial comes in. Because it's so thin, the stacked layers have a large combined surface area, meaning they can gather lots of hydrons from ambient moisture.
In the Nano Energy paper, a single unit with the surface dimensions of a strip of chewing gum, but a fraction of a millimetre thick, generated 0.85 volts, or about half a standard AAA battery.
Professor Chu and his team have developed better designs since the paper was submitted last year.
He says a matchbox-thick stack of these new MEGs is able to generate 4 volts.
"Now the performance is much, much better. It's at least 10 times higher performance."
Does it need to stay wet?
Yes, but they don't need much water and work fine in conditions of low humidity.
According to Professor Chu, the MEGs work with a minimum of 40 per cent humidity, which in weather terms is a non-humid day.
Once the whole MEG unit has the same level of saturation, it won't generate electricity — one side has to be drier than the other to get electric potential.
But as one side dries out, or gets wetter, it resumes generating electricity.
However, this process of getting wet and drying out gradually depletes the surface of the unit, Professor Chu said.
"We think it can last at least for three months."
What could it power?
Jingwei Hou is a chemical engineer and expert in nanomaterials at the University of Queensland.
He said the UNSW MEG design was "very interesting".
"I think this device is the best one so far," said Dr Hou, who was not involved with the research.
For the moment, those products are likely to be small, disposable devices, such as stick-on electronic sensors that can monitor vital signs 24/7.
"If you have a large enough surface, you could power a mobile phone, but that would be huge — on the square-metre level," Dr Hou said.
Later this year, Professor Chu's team plan to print a 3-square-metre MEG.
If MEGs prove scalable, they could be applied as a thin film to the windows of high-rise buildings and used to power emergency services, Dr Hou said.
"You could be creating enough electron density to power the fire alarm sensors for the whole building," he said.
'Demo version ready by spring'
The company Strategic Elements has been collaborating with Professor Chu's UNSW team since 2015.
The work has been partly funded by Commonwealth tax offsets and funding grants and help from the CSIRO.
"We've had millions of dollars of federal government support," managing director Charles Murphy said.
In the third quarter of this year, the company plans to release a commercial prototype, or what Mr Murphy called, "a pretty vamped-up demonstrator powering a commercial skin patch."
"The core benefit of the technology is that it's flexible," he said.
"Skin patches is a no-brainer market for us to focus on first."
