Holographic Solar Collectors

An anonymous reader writes "The MIT Technology Review is reporting that Prism Solar Technologies has developed a technique to use holograms to concentrate light onto photovoltaic (PV) cells. While the implementation is only about a 10x increase over PV cells without collectors such as mirrors/lenses (mirror/lens approaches can do 100-1000x), it is a great deal simpler, more compact, and cheaper. Also because of the concentration, there is less need for physical PV cell real estate compared to crystalline PV silicon cells of similar output."

Comments

Holograms?

[exp(pi*sqrt(163))]

These sound like good old fashioned diffraction gratings to me. 'Hologram' sounds like nothing more than a marketing term. One disadvantage of using diffraction gratings is that the amount of bending is wavelength dependent. And it seems like the marketing department managed to put a spin on that too.

Solar collecting is good.

[crazyjeremy]

When you can concentrate the suns energy the collector is more efficient. This is a VERY good thing, especially considering the amount of cloudy / rainy days most places have. Lots of people do not go solar because it simply does not draw enough power for the amount of money they have to use to build the system.

Re:Solar collecting is good.

[shmlco]

"Lots of people do not go solar because it simply does not draw enough power for the amount of money they have to use to build the system."

I can see solar as a potential option for some businesses, but for home use you still have the small problem of no power output during the night. And that's usually just when you want some lights, television, heat, and so forth.

If they want solar to REALLY catch on someone is going to need to develop not just a cost-effective solar cell, but also a cost-effective way to store and reuse the energy collected during the day.

So many stories but where are they?

[Zaai]

Pardon me for being sceptical about the actual commercial feasability of this.

Over the last decade quite a few of these wonderful improvements have been announced yet the commercially available solar-cell still has an efficiency of less than 15% and the price hasn't changed that much either.

I wonder if these announcements are more motivated by an upcoming investment round...

God knows we could use them, but when do we get to see them?

10x input != 10x output

[EmbeddedJanitor]

Silicon solar cells are still too damn expensive and power hungry to manufacture to be a useful mainstream generation tool. The only places they're really being used is where mainstream supply is not available/practical or where they are heavily subsidised for political/marketing ends. Increasing concentration to reduce the silicon are does reduce the amount of silicon and therefore potentially reduces the $ per W. However...

PV efficiency reduces significantly with increase in temperature (which is why you see solar racer folk pouring water on the PV panels). Thus just cranking up the sunlight by concentration does not give a linear increase in output. PV cells for concentration thus need to be made thicker and differently (to code with the extra current, heat sinking etc.) but hopefully the payback is still there.

Personally I think the PV quest is being approached incorrectly. There's too much emphasis on efficiency. Labs try to out % eachother and the big solar showcase is the solar race which is all about high efficiency cells.

What they should target is $ per Watt because that is the real hurdle to making PV viable. Who cares if it's only 5% efficient, so long as it is cheap? Tile your house with the stuff to get the area.

Re:10x input != 10x output

[syphax]

Who cares if it's only 5% efficient, so long as it is cheap? Tile your house with the stuff to get the area.

You are almost absolutely correct. Except for two things:

1. Low efficiency leads to higher indirect costs- specifically, the infrastructure that holds the cells and connects them to the grid. As you get down to lower efficiencies, these costs become significant.

2. Even at 10% efficiency, you need a huge area to produce a significant amount of juice. Sure, we could in theory generate all the energy we need in the U.S. by covering "only" around 1% of the U.S. land area with 10% efficiency PV (practical issues aside), but that still works out to be a huge area. Like, say, Maine. So even if we had a nice, cheap, low-efficiency solar technology, it's usefulness would ultimately be limited by land use constraints.

Re:Only 10x? That's huge!

[Ungrounded Lightning]

It sounds [like the improvement is] to make it cheaper by only having to use ten percent the amount of PV cells in the same area of solar panel.

Yep.

a one square yard panel of naked PV cells shouldn't get any more energy than one square yard with holographic cells... right?

A square yard of naked cells (or cells imbedded in a classic panel), a square yard of focusing concentrator onto a smaller area of cells, and a square yard of holographic panel containing some smaller area of cells, would all potentially collect the same power (neglecting concentrator inefficiencies).

The point is that:
  - doing a square yard of collection with a square yard of cells costs.
  - A normal focusing concentrator focuses not just the useful light, but the non-useful far-infrared, so you need serious cooling of the cells to run at a high degree of contentration, and the concentrator is bulky, heavy, and may need to track the sun.
  - This thing is WAY cheap to make, doesn't focus useless infrared below the cells' bandgap frequency, and doesn't need to track. It loses some of the light, so you may need a little extra area to make up for that. But you use only 10% of the cells compared to a classic panel for a given amount of power.

As I read the drawings this is basically a glass plate with solar cells glued to 10% or so of the back and the remainder covered with a holographic coating.

The holographic coating diffracts the desired frequencies so they become trapped between the faces of the glass plate by total internal reflection (as light is trapped in a fiber optic light pipe) and it bounces back and forth between the surfaces until it hits a place where a cell is glued to the back. At that point the glue's index of refraction is high enough that the light can escape into the cell. So you just need enough cells that most of the light encounters one before it gets to an edge or leaks out where a dirt speck sits on the glass. (I'm not clear how they keep the holographic coating from diffracting it back out toward the sun but I presume they've got that covered.)

Far infrared doesn't bend enough to get trapped so it escapes out the front or back of the panel.

This is VERY nice. With maybe 90% of the infrared passing through the panel or bouncing out the front of it you don't get the massive greenhouse effect of a classic panel.