Cleaning solar panels uses way more water than you might think – unless you were right on the money thinking 10 billion gallons of water per year.

That amount of water is almost unbelievable, says Professor of Mechanical Engineering Kripa Varanasi, who, along with graduate student Sreedath Panath and a team at the Massachusetts Institute of Technology, wants to create a water-free way to clear dust off of photovoltaic panels.

“The water footprint of the solar industry is mind boggling,” said Varanasi in a recent MIT news release , and it will be increasing as these installations continue to expand worldwide. “So, the industry has to be very careful and thoughtful about how to make this a sustainable solution.”

With simple components – a metal-bar “electrode,” a guide rail, and an electric motor – the MIT-developed sustainable system makes dust particles detach and virtually leap off the panel’s surface, without the need for water or brushes.

The achievement, described last week in the journal Science Advances, relies on a process called electrostatic repulsion, or what’s known as Coulombic forces.

To activate the system, a simple electrode (that metal bar) passes just above the solar panel’s surface, charging the dust particles, which are then repelled by a charge applied to the panel itself. The system can be operated automatically using an electric motor and guide rails along the panel’s sides.

The panel also has a transparent, nanometer-thick conductive layer deposited on the glass. By calculating the right voltage to apply to the panel, the researchers found a charge sufficient to overcome the pull of gravity and adhesion forces, and lift the dust away

Similar electrostatic-based dust-removal ideas have been tried on NASA’s rovers, but the Earth’s deserts, where many solar panels are placed, feature a natural element that Mars has much less of: moisture.

Moisture, it turns out, is a critical factor for the MIT-made prototype.

Varanasi and the MIT researchers tested their setup in various humidities. According to experiments, all dust can be removed when ambient humidity is greater than 30 percent. Dust removal with this method is more difficult when the humidity decreases.

There’s still good news, though, says Varanasi.

“Most deserts are in that range, so that's great,” the MIT professor told Tech Briefs.

The system can either be retrofit or built from scratch. To tack the setup onto existing panels, the idea is to fit a railing on each side, with the electrode metal bar spanning across the panel. A small electric motor, perhaps using a tiny portion of the output from the panel itself, would drive a belt system to move the electrode from one end of the panel to the other.

Alternatively, thin strips of conductive transparent material could be permanently arranged above the panel, eliminating the need for moving parts. The whole process could be automated or controlled remotely.

But there’s still more to prove.

In a short Q&A with Tech Briefs below, Prof. Varanasi explains what they’ll be trying to demonstrate next with their water-free way of removing dust.

Tech Briefs: What inspired this work?

Prof. Kripa Varanasi: You know, we all want more renewable energy, and solar will play a big role. There's a significant increase in the global photovoltaics capacity. There is impressive, significant research in improving the efficiency of these panels. But something as mundane as dust can bring all of that down. With the dust, you lose up to 30% efficiency. You get a shadowing effect.

Every time you wash the panel, you bring the efficiency back. For a small, 150-Megawatt plant, though. you're looking at almost 7,000,000 gallons of demineralized water per year to clean these. Water is such a precious commodity, and people need to be careful about how to make use of this resource that we have. The solar industry really needs to keep this in mind; we don’t want to be solving one problem and creating another. And if you go and just simply scrub these, you'll scratch all the glass and you're back to square one.

Tech Briefs: Did you try any other methods?

Prof. Varanasi: One idea was to collect the water from the plumes that go out of cooling towers. We inject charge and get those droplets to collect, and you can save a significant amount of water.

So, thinking along similar lines, we were trying to inject charge into the dust and use Coulombic forces. And we saw something interesting:

The dust particles are getting charged almost like a metal; they behave like a conductor. Because the dust is made up of silica, it can actually absorb moisture. So, you have a tens-of-nanometers-thick layer of water film that can make this conductive.

We did a bunch of systematic experiments, and we figured out what the charge is on these.

Tech Briefs: How did you do these experiments?

Prof. Varanasi: Believe it or not, there are people who sell dust for a living. And we were able to get systematic compositions and do a bunch of different experiments with a range of particle sizes. We were able to show that our physics, our modeling, works. Our system works in about 30% relative humidity. Most deserts are in that range, so that's great.

Most deserts are very cold in the night, so you get dew formation. Most desert animals actually survive that way. You could get the particles off the surface every morning, even in a dry environment where you have a small window to can get moisture in.

Tech Briefs: Have you tested this in the desert?

Prof. Varanasi: We haven't tested it out in the environment, but we did experiments in the lab. We changed the humidity to simulate these kinds of effects. But clearly, yes. there's a lot more work to be done in terms of taking it out, testing it in the field, and proving it out.

Tech Briefs: How much voltage is needed for this to work?

Prof. Varanasi: This is high voltage. It's typically in the kilovolts range – between 5 to 10 kilovolts.

Tech Briefs: What is the cost versus energy output?

Prof. Varanasi: We have not done detailed technical economics in that way. We look at this is in terms of the efficiency gain that you're seeing and the amount of water usage that you don't have to have. Any amount of water used has to be pure and has to be trucked in, because you’re in the middle of a desert.

The power consumption from our device is extremely low. Although it's high voltage, it's almost zero current, so there's no power consumption. The drive that's used to move this is a stepper motor. These are fairly well established and not significant.

Tech Briefs: And can this be retrofit onto existing panels?

Prof. Varanasi: Yes, that's the plan.

Tech Briefs: Can you tell me more about that prototype?

Prof. Varanasi: The prototype is a small, lab-scale panel with a robotic arm that can move back and forth. These kinds of robotic arms are already used in applications like cleaning, but they are all contact-based.

Tech Briefs: How would the transparent conductive film be applied to a rather large surface?

Prof. Varanasi: These are roll-to-roll-based films. The manufacturing of these transparent films is already well established because of the iPhone, iPad, and all of these markets. That's an established industry, and so it could be used to retrofit onto these panels. For new units, you can make it part of the manufacturing of the of the panels and the glass.

Tech Briefs: Can the direction of the dust and its removal be controlled?

Prof. Varanasi: As the dust is removed, one can adjust how the arm moves. You could adjust the way the electric field is applied, and you can definitely tailor where the dust falls.

Tech Briefs: Could that be a potential problem – guiding where the dust falls?

Prof. Varanasi: So, if you have a long stack of these panels, you could move the robotic arm forward and push it off the side, or you could go back and forth, and push it off. But I agree; one has to think about all of this when you're looking at the large-scale system, but that could all be engineered into the device.

Tech Briefs: And so, what's next?

Prof. Varanasi: I think the next step would be to do a full-scale pilot and prove this. We'll learn a bunch of things. We don't know what we don't know, right? I always like to do pilots like these and scale up the technology. We understand the science. We now know how to make a prototype and a proof of concept. Now we’ve got to do a proof of product in the in the actual settings. All signs show that we should be performing pretty well, but we learn what we learn and improve the product accordingly.

What do you think? Share your questions and comments below.