Spray-On Coating Could Make Solar Panels Snow-Resistant

Edward Brown, Associate Editor

A University of Michigan team led by Professor Anish Tuteja has demonstrated an inexpensive, clear coating that reduced snow and ice accumulation on solar panels, enabling them to generate up to 85% more energy in early testing.

Tech Briefs: What applications for ice-phobic coatings have you worked on previously?

Professor Anish Tuteja: Ice-phobic coatings have a lot of different applications — they're all over the place. If we stick to the renewable energy field, certainly, coatings for wind turbines. You probably remember from last year, in Texas, where they had all those winter storms, the wind turbines essentially froze, and they had shut them down for a while. They were initially blaming that for all the challenges they faced. So, wind turbines is a big one. Then, there’s de-icing of airplanes and lots of other applications, like power lines, which can actually collapse due to the buildup of ice. There are applications in refrigeration where you don't want to have ice buildup on refrigerator coils. Then, there are car windshields, of course. Ships, as they're starting to travel across the Arctic — there, icing is a big concern — all sorts of transportation applications.

That’s what first got us interested — trying to develop surfaces where ice doesn't adhere or can be easily shed by its own weight.

Tech Briefs: How have your de-icing solutions differed from what came before?

Tuteja: We've been working in this area for about six to seven years now. As we all experience, ice likes to stick to basically everything and so for decades, people have been trying to research solutions that have some sort of coating or paint with low adhesion, or low attachment, of ice on surfaces. You may be familiar with skis, where wax is applied to try and prevent the attachment of snow and ice.

Those kinds of solutions have been explored over the last 50-plus years. The challenge has always been their durability — those coatings don't last very long.

If you're going to put it, say, on a wind turbine, or a solar panel, or a power line — those things need to have lifetimes in decades. But no solution for that has existed, so that's where we got involved.

We figured out a new mechanism by which you can design such coatings, essentially chemistry-agnostic, not just dependent on one type of chemistry. We tried them on a huge range of different surfaces for which we developed very, very different chemistries. All of these showed a very low ice adhesion, very low attachment of ice. So, that that was our starting point.

For about five to six years, we published a number of papers on new mechanisms of ice shedding. This one is different, however. Ice is probably the simplest one you can solve, and even that was very complicated. In snow, you have solidified water, of course, but also a lot of porosity. Many times, you also have liquid water trapped within it, and so the density and the structure can be quite variable. Depending on the environmental conditions, the range can be huge over which we structure the modulus — the hardness or softness of the snow.

We were certainly not convinced that the things that work well for ice were going to work well for snow — they’re not the same. So, that's where this was really exciting. We made an educated guess on what might be useful, and it turned out to actually be useful — it was a happy surprise.

The other thing that excites me about this is I like to see real-world applications. We applied it to solar panels in Alaska — we shipped the material, and our collaborators at the University of Alaska applied it. You don't have any control when someone else is putting the coating on, so it was nice that even without that control it actually worked. So, both on the scientific level and on the practical level, it was satisfying that it all came together.

Tech Briefs: Have your previous de-icing solutions been applied in the real world?

Tuteja: They have been licensed out at the university through a startup, and it's being commercialized in different fields. There's no large-scale application just yet, but there have been many trials on things like buildings that have a lot of snow and ice buildup in winter. For example, if you have a high rise in New York, things falling off can be a huge issue. They've also done trials on dams and such. In winter, dams can get frozen over, and some of the gates don't move well because of snow and ice buildup.

There’s also been some work on food packaging, so that, for example, your wrapper won’t stick to a frozen pizza or burger or something along those lines. Although, there are different sorts of applications being tested with the goal of commercializing them in different fields, nothing large-scale just yet.

Tech Briefs: What does low interfacial toughness mean?

Tuteja: That was one of the new mechanisms we figured out for ice shedding. For the last 50-plus years, people have talked about ice-phobic surfaces. These surfaces have all been defined by an ice adhesion strength, which is force per unit area. Imagine you have a surface — a piece of aluminum or whatever — and a cube of ice is sitting on top of it. The force per unit area required to move or break the ice from the surface divided by the area of the cube that's in contact with the surface is the definition of force per unit area.

The challenge with this is how you can keep reducing force per unit area. Some of the earlier work that we did was all about that. The difficulty is when you have a really large-scale application. For example, for a solar panel, or a wind turbine, or an airplane wing, the interfacial area is so large that even if you only need a very small force per unit area, you require a huge total force to get rid of the ice.

Two or three years ago we came up with the idea that, instead of worrying about force per unit area to dislodge an entire cube of ice, what if we instead focused on crack propagation. We could induce a crack at the interface — not through the ice, but between the ice and the coating. Once we formed that crack, we wouldn’t need any additional force — it would just keep propagating across the interface, delaminate the whole thing, and detach the ice. The only force required would just be to initiate the crack. After that, even a small additional amount of force will cause the crack to propagate. Then, it wouldn’t matter how big the surface was — it would be exactly the same force.

So, interfacial toughness is essentially a measure of the energy required to create that crack. Low interfacial toughness means you only need very low energy to create the crack — making it easier to shed the ice.

Tech Briefs: How do you initiate a crack?

Tuteja: Literally by an application of force. We showed that you can even use the weight of the ice itself. Imagine you have ice building up on a vertical surface. There is a lot of stress being generated at the edge, so the weight of the ice itself can create a crack that will go up and down and get the entire thing to delaminate. You can use that weight or any additional forces that might be present. The ideal is using what we call passive forces so that you don't have to apply any additional energy.

Tech Briefs: How do you initiate the crack versus moving the whole block?

Tuteja: All we’re doing in both cases is applying a force at the edge. What we explain in our paper is when you have smaller areas, the entire thing will break as one. Whereas if you have a larger area, you can initiate crack propagation, even on conventional materials. When we make special materials, they do better, but this is a universal phenomenon — it works for all things.

As we had discussed in our paper back in 2019, there are different material design parameters for combining things to get to ice-phobic surfaces. In this case, for the solar panels, we developed something that was both ice-phobic and had low interfacial toughness, because we were dealing with both small-scale and large-scale aggregations. There might be only a pound of snow covering the panels, or a much larger amount, and you want both to have low forces required for detachment. So, it was pretty exciting that our initial material worked. We developed something that had low isolation strength and low interfacial toughness. But all of that was just with ice. At that point, we didn't have any way of making artificial snow.

Tech Briefs: I read that you coat the surface with rigid PVC. How do you do that?

Tuteja: We modified the PVC with some plasticizers, which are oils that are miscible with PVC. In some of our papers we talk about the idea of slip or slippage. Essentially what it does, it allows ice to slide more easily and allows for lower detachment forces.

Tech Briefs: It also mentioned that you used PDMS.

Tuteja: Yes, those were two different materials that we tried.

Tech Briefs: From what I read, PDMS is inherently kind of slippery.

Tuteja: Yes, by itself it has relatively low attachment to ice. And we modified it further by adding silicone oil. PDMS by itself is a silicone that's crosslinked to form a rubber and we added some silicone oil within it to make it even more slippery. In the end, though, it turned out the PVC system worked better than the PDMS system.

Tech Briefs: If this is so slippery, how does it adhere to the panel?

Tuteja: That's a great question. That was one of the main things in our first paper in 2016 when we were working on this. If you add too much oil, it forms a continuous layer on the surface. If it's above the miscibility limit between the oil and the rubber matrix, it will phase-separate to form a layer of oil on top. That’s terrible for any sort of mechanical durability. That's why we determined it had to be a miscible oil so that it was fully imbibed within the matrix — there's not a free layer of oil on it. If you touch the surface, you can't tell that there is any oil within the material.

Tech Briefs: So, you already tested this in Alaska. Do you have any sense of when this might be developing further?

Tuteja: We're hoping so. Since this has come out, there's been a lot of interest from different sorts of manufacturers, different PV operators across the northern climates to try this out — so that’s in planning.

This particular coating is certainly not our most durable one. We expect it probably has a lifetime of about a year or so — you would have to reapply it every year. We've now been working on generation 2, which is much more durable.

The next step is to apply these in the field and see how well they work. For next winter, I think there will be multiple installations for us to test our coatings. But at least in internal testing, the new types of coatings are more durable than what we had previously.

As I was saying earlier, we had no idea what would actually work. Now at least we have some idea, so it’s a good starting point.

Edward Brown is Associate Editor.

---