New materials could deliver ultrathin solar panels | Hacker News Reader

New materials could deliver ultrathin solar panels

news.stanford.edu

57 pointsstareatgoats31 comments

Thin/flexible cells are often called "eco", because "less material", but then I find the composition and they are made of Arsenic, Cadmium or similar toxic non-degradable materials that will end up in the environment.

This one seems actually seems very ecological, or at least not toxic:

> tungsten diselenide and contacts of gold spanned by a layer of conducting graphene that is just a single atom thick

They don't mention Cadmium Sulfide surface layer as seems standard in CIGS. Hope they don't feel the need to add it in the efficiency optimization process.

EDIT: Doing some calculation on CIGS with 700Å (70nm) CdS layer [1], 1m² of solar panel would have about 1m²·70nm·5g/cm³ = 1m²·70e-9m·5e6g/m³ ~ 0.4g of CdS.

[1] https://www.nrel.gov/pv/copper-indium-gallium-diselenide-sol...

There was an interesting post shared here recently on low-tech solar panels, which might be able to trade off efficiency for recyclability:

https://solar.lowtechmagazine.com/2021/10/how-to-build-a-low...

> The plug was an alloy of zinc and antimony – which we now know is a semiconductor. It was alternately capped by German silver (a nickel, copper, and zinc alloy) and copper on opposite ends.

While the design may be possible to make and repair in artisanal manner, it uses a lot of antimony and any repairs would likely shed more Sb than there is Cd in an entire CIGS panel (though In and Ga may be a concern).

If the design was resource-optimized, it would still have a potent monk-killer (Sb) and then it wouldn't be worth it to recycle all of the panels.

Thin isn't as useful as it sounds. For a few application (drones) light weight is important, and thin probably does mean light weight. However for most solar application a bit of extra thickness is just something you design into your mounting system, and since thicker is stronger (think hail and other weather phenomenons) you typically design panels far thicker than needed to support the cell anyway.

Transportation at the least. Got a place in the mountains - moving a few panels was a insanely costly and annoying.

The best thing about being very thin is that they could be folded up into smaller spaces.

You could unfold 100m of panels from the back of your EV when you park. Or carry enough in your backpack to power your lappy.

I get your point, but you can already carry enough panels in your backup to power your laptop. I have a 100W folding panel that folds down to about 10x8x2.5” I bought it for $200 about a year back, you can pick it up today for $140.

Crazy how fast prices are dropping.

Link to model? Any recommendations?

Look up the Topsolar SolarFairy 100W it’s a Chinese brand you can order on Amazon. I also have a Dokio 220W folding panel that’s my main one for camping, but it’s way bigger. Though it’s really thin and easy to find a place to store without taking up much space, it’s also easier to keep directed at the sun.

The new Stanford prototype achieves 5.1 percent power conversion efficiency, but the authors project they could practically reach 27 percent efficiency upon optical and electrical optimizations. That figure would be on par with the best solar panels on the market today, silicon included.

I'll believe it when I see it. Solar panels and batteries are both subjects that have generated far more in terms of improvements on paper than they delivered in real life.

To add to that you simply can’t escape physics…

A 200W solar panel with 20% conversion efficiency needs a 1000W of solar radiation hitting it to produce that much, most of it would be converted into heat, that’s roughly the average for an area just under 1 sq/m in size.

Thermals are a big problem for solar panels and I’m not sure how that would work with thinner panels.

I also don’t see how you’ll power a drone with solar panels regarding how much they weight the surface area seems not to be sufficient even if the panels would be 80-90% efficient.

On the plus side, every little bit that can be exported as electrons does not need to be shed as IR. The problem is that quite a bit of that 1000W incident is not in a part of the spectrum that lends itself readily to conversion to electricity.

Couldn't one build a dielectric mirror onto the surface of the panel to selectively reflect those frequencies that aren't converted well, to avoid the thermal load of absorbing them? Is the issue that dielectric mirrors are too fragile?

No, the problem is that your dielectric mirror would have to be a perfect reflector for one set of wavelengths and a perfect passthrough for the remainder. In practice though it will always be gradients so you'll end up reflecting some of what you want and passing some of what you don't want.

They are not too fragile; they can coat them onto the solar cell's surface.

Panel thickness is not a positive factor for cooling, it's only good for structural support. Unless your panels are tiny, all the heat escapes through the front/back surfaces, thinner being better, though surface treatment and airflow is a bigger factor.

If they meant winged drones - those already exist.

You haven't been paying attention, I suppose. The cost per watt of solar panels continues to drop exponentially, as optimizations continue and production continues to scale up. Breakthrough research announcements like this happen often, and only a handful of such announcements will ever make it to production. But improvements certainly /are/ making it into production, and lowering LCOE steadily.

Here's an example graph, with both solar and battery costs: https://greentechlead.com/renewable-energy/cost-for-onshore-...

And compared to fossil fuels: https://en.wikipedia.org/wiki/Levelized_cost_of_energy#/medi...

(LOTS of people think about renewable prices as they were ten years ago instead of where they are now or where they may be in five years. This means lots of people simply have wrong priors about the relative cost of different energy sources. This is expected: People don't intuitively adapt to exponential change.)

The graphs you mentioned are about cost per watt, an important metric, but not panle conversion efficiency.

I've been paying excellent attention, the panel efficiency for the various crystal lattices and layering combinations has hardly moved in the last decade. A little bit here, a little bit there.

Sure the price has come down, but that wasn't what this was about.

Also, I think the tone of your comment could be a lot better than it is.

Here's some progression of panel efficiency over time: https://news.energysage.com/solar-panel-efficiency-cost-over...

I would suggest it's a matter of "nothing changes in a year but everything changes in a decade."

Partial apologies for the tone; I'm personally kinda sick of Internet comments crapping on people's work by default.

The part that matters is that 'run of the mill' monocrystalline solar panels, not exotic ones such as triple junction and other very expensive cells have been roughly steady over the last decade or so at 20% give or take, with their polycrystalline brethren doing a slightly worse.

There are prototypes and special application cells that do much better than that, but they are typically priced in such a way that you would never use them in a regular domestic application. But for a satellite or something else where price isn't your first consideration (but for instance weight is) they may well make good sense.

FWIW I have been following the renewables scene very closely for two decades, have built a house on solar and wind power in Canada and am in the process of converting a house here in NL to as close to self sufficient as I can make it without rebuilding it from the ground up.

I, for one, would be very interested in any write up on your experiences, esp wrt solar. I'm still trying to sort out solar hot water vs panels. We have a fairly high hot water consumption rate.

An interesting experiment in Alberta - http://www.dlsc.ca/

> I'm still trying to sort out solar hot water vs panel. We have a fairly high hot water consumption rate.

Every situation is different, but generally speaking it makes more sense these days to install ever cheaper PV rather than solar water heating and pair that with a heat pump water heater, at least in typical grid connected scenarios.

Perhaps the considerations are different for full off grid setups though.

This is correct and roughly the installation that I'm planning.

A few more bits of useful info for this setup:

- Most Heat pump water heaters have a 3000W-6000W backup resistance element used when conditions are too cold to run the heat pump. If you live in an area with long stretches of near or below freezing weather, make sure you install it in a conditioned or semi-conditioned space.

- The exception is the Sanden C02 system which can heat water even in sub-zero temps.

- Heat pump water heaters draw about 800W / 4A when using their heat pump, although probably more for the Sanden. As with most fixed speed compressors, the inrush/surge current can be as much as 4x the steady current, which is important for sizing the wiring and circuit breakers, as well as sizing max current output of any batteries that might be installed.

- Heat pump water heaters have slower recovery times than gas fired water heaters.

In a nutshell: panels were still quite expensive when I built my installation, so we choose a double axis movable mount for the panels, two blades of 8 panels each. This allowed for both day-to-day and seasonal adjustment and ensured perfect incidence. In practice that system was the weakest point in the chain, and my new installation will be stationary, I'll spend the extra money from the moving mounts on more panels and end up with a surplus in energy compared to the movable mounts and have a more reliable system.

The system I built was totally off the grid, in my new installation, which has neutral net metering I will use the grid as a fake storage device saving a bunch of money as well as space for the installation.

The inverters I used were two stacked Xantrex 5 KW each for 240V, in my new installation I will use a Victron 12 KW inverter.

The panels will be Samsung monocrystalline, there will be 32 of them, there will be no other source of power.

Heating will change from natural gas to heatpump, which will be a major affair because there is only the possibility of a shallow sink/source here, deep ones are not allowed.

If you have any specific questions feel free to mail me: jacques@modularcompany.com. As for hot water: I will definitely use all the space for electrical panels and use some electricity for hot water, there are excellent 'instant' hot water sources that are electrically powered now. The hassle of dealing with a system with fluids in it on a rooftop is not something that is maintenance free enough to add value to the house. Those systems tend to be quite fragile and need frequent repairs as well as not being available 'on demand' which with a family can get you to the point where there simply isn't enough hot water.

> converting a house here in NL to as close to self sufficient as I can make it without rebuilding it from the ground up

Do you have any helpful resources geared towards beginners on self sufficiency for either ground up construction or renovation? I have seen videos on YouTube but feel like watching the final result is not the same as a comprehensive guide on how to go about it.

Hm, interesting idea. I Have a lot of projects on the go and my vow for this year is not to start new ones until the ones that are in progress are done. But I've renovated 4 houses and built one from the foundation up so there is a ton of experience accumulated that might be useful to someone else.

This is a major project though, probably book sized by the time it is done.

Agreed that Youtube videos are not the right medium for this (they aren't for lots of stuff that is on youtube).

Another entry in this already very long list:

https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...

This is good, but shouldn't the priority maximising power output from X area of solar panel?

I would have thought the industry needs customers to get excited about covering their electricity bills (especially now), let alone just getting to a point where residential customers can enjoy a day of full power without dipping into the grid.

(It's likely I don't know what I'm talking about with the above, but I know what I feel as a potential solar power customer. I imagine this is addressing a different problem that others are interested in, but I wish the industry would solve the most pressing problem).

Reply on news.ycombinator.com