80% solar cells

[raptor1956]

Yep, this has got to be click-bait, sure looks like it to me.

Actually, according to some research at Rice University, they appear to be indicating that they've found a way to convert the large percentage of wasted energy in typical solar cells by a process, if I understand it properly, that is able to convert the waste heat into photo-electrons that can come from any direction but are channeled in only one direction via carbon nano tubes and this channeling squeezes them into useful electricity.  OK, this has my bullshit detector going off but I'll link the article for your own perusal. 

https://www.sciencealert.com/device-that-channels-heat-into-light-could-boost-solar-efficiency-to-80-percent

https://news.rice.edu/2019/07/12/rice-device-channels-heat-into-light/?T=AU


The thing is, as improbable as this seems if there is any there-there to this and they are able to up the efficiency to anything near 80% its game over for all other energy source and the game isn't even close.  We've seen about 44% with triple junction cells that would never be commercially viable and consumer grade cells are now about 18% for the better ones, but getting anywhere near 80% would end the debate and put Texas out of business.


Brian

[Kleinstein]

There is a rather general thermodynamic based limit to the possible efficiency of a PV (or other solar light to power converter). Without concentration the limit is quite a bit lower than 50% AFAIR. For non concentrated PV it's more like a round 35%.

https://en.wikipedia.org/wiki/Shockley%E2%80%93Queisser_limit

So the BS detector goes to full over-unity.
Once converted to thermal energy, there is not much power to get back unless the temperature is really high.

AFAIK good commercial cells can reach some 22% under standard condition and maybe 20% real world (higher temperature).

The triple junction cells get there high efficiency with concentration. Under suitable conditions (AFIAK in Spain) such installations with concentrated PV were actually build at quite some scale. So it is not that far off to take this system into consideration for suitable places.
With concentration the actual cell is tiny (some 2% AFIAR) and may allow more exotic materials.

[CatalinaWOW]

As a Rice alumni, the BS comment forced me to read this closely.  It appears to me that the original Rice release has potential for credibility, saying that they could convert the heat from "very hot" sources to narrow band light.  They do claim 80% conversion, but again only for very hot sources.  With no other stipulations on the operating conditions or how efficiency is measured.

This process bypasses the Shockley limit mentioned earlier, as that involves direct interaction of the solar spectrum (photon energy distribution) with the band gap.  The Rice process changes that distribution and thus changes the Shockley calculation.

The science writer further removes the claim from any discussion of the circumstances and limitations, and thus into BS territory.

This whole discussion reminds me of one of my favorite SF stories where one protagonist proves something impossible by demonstrating instability of a (science fiction) force field for durations longer than a few microseconds.  Another protagonist says, "Sure, I found that experimentally."  And put a high speed switcher on the generator, resulting in a force field that never reaches its instability limit but is for all practical purposes always there. 

A real world example of the same thing is Lord Kelvin's analysis of heat flow from the Earth and the heat capacity and temperature of the material in the Earth's interior.  The conclusion was that the Earth could not possibly be more than a few tens of thousands of years old.  The discovery of radioactivity made those calculations irrelevant.

I don't know if the Rice work is a breakthrough or just a wild claim, but it is too soon to push the proven BS button.

                                                                                                                                                                                                                                                                                                                                                             

[raptor1956]

There is a rather general thermodynamic based limit to the possible efficiency of a PV (or other solar light to power converter). Without concentration the limit is quite a bit lower than 50% AFAIR. For non concentrated PV it's more like a round 35%.

https://en.wikipedia.org/wiki/Shockley%E2%80%93Queisser_limit

So the BS detector goes to full over-unity.
Once converted to thermal energy, there is not much power to get back unless the temperature is really high.

AFAIK good commercial cells can reach some 22% under standard condition and maybe 20% real world (higher temperature).

The triple junction cells get there high efficiency with concentration. Under suitable conditions (AFIAK in Spain) such installations with concentrated PV were actually build at quite some scale. So it is not that far off to take this system into consideration for suitable places.
With concentration the actual cell is tiny (some 2% AFIAR) and may allow more exotic materials.

There are triple junction cells in operation at over 43% so that's well above the 35% value you listed.  I may be mistaken but the 35% value may be the theoretical limit for single junction cells. 

I didn't mention in my first post but it appears this may work in co-generation situations by converting waste heat into electricity.  OTH, if it really is in the 80% range, and that point just seems impossible, but if it is true why do co-generation -- just use whatever heat source you have and use that heat directly with this technique. 

As for commercial cells ... the range is about 13%-16% for CIGS based really cheap cells to the 16%-20% for lower cost silicon cells to 35% or so for space grade cells at much higher price on up to the 43% cells that I don't think have made it out of the lab just yet.

It appears that solar can now produce at around $0.02/KWHr though that is still limited to daylight hours, obviously.  One of the advantages of molten salts based concentrating solar is that the energy can be stored and then produce electricity at night, generally in the hours after sunset when demand is still high but the Sun is no longer shining.  I believe concentrating solar is now about $0.06/KWHr or there abouts.   The person or company that cracks the storage problem and can utilize solar 24/7/365 will be an instant billionaire and hugely impact the climate debate in a favorable way.


Brian

[NiHaoMike]

One easy way to boost solar panel efficiency is to make the back side out of heatsink. Problem is that would increase the cost per watt by a lot, so it's only used in special applications.

It appears that solar can now produce at around $0.02/KWHr though that is still limited to daylight hours, obviously.  One of the advantages of molten salts based concentrating solar is that the energy can be stored and then produce electricity at night, generally in the hours after sunset when demand is still high but the Sun is no longer shining.  I believe concentrating solar is now about $0.06/KWHr or there abouts.   The person or company that cracks the storage problem and can utilize solar 24/7/365 will be an instant billionaire and hugely impact the climate debate in a favorable way.

Considering that the majority of home energy use is thermal in nature - HVAC and hot water, thermal storage can work very well. What's lacking is incentive to invest.

[raptor1956]

One easy way to boost solar panel efficiency is to make the back side out of heatsink. Problem is that would increase the cost per watt by a lot, so it's only used in special applications.

It appears that solar can now produce at around $0.02/KWHr though that is still limited to daylight hours, obviously.  One of the advantages of molten salts based concentrating solar is that the energy can be stored and then produce electricity at night, generally in the hours after sunset when demand is still high but the Sun is no longer shining.  I believe concentrating solar is now about $0.06/KWHr or there abouts.   The person or company that cracks the storage problem and can utilize solar 24/7/365 will be an instant billionaire and hugely impact the climate debate in a favorable way.

Considering that the majority of home energy use is thermal in nature - HVAC and hot water, thermal storage can work very well. What's lacking is incentive to invest.

In the USA, the average daily electric consumption is about 30KWHrs with southern climes a bit higher do to AC needs and northern climes a bit lower.  Thermal storage is certainly a significant part of the total energy needs and if all thermal regulation were done by electricity then the daily average would no doubt be higher -- many places up north use fuel oil to heat there home so that doesn't figure into the electric bill other than the relatively small requirements for pumping/circulation.

I think the big thing is home builders have other priorities when they build and home buyers tend to fixate on kitchens and bathrooms with far less focus on energy.  But, going forward, as more people drive electric vehicles the ability to recharge them from your own energy supply at $0.02/KWhr could be a huge money saver.  A typical Tesla consumes about 275WHrs to about 350WHrs per mile so recharing at $0.02/KWHr would translate to a cost per mile of less than $0.0055 or about $1 for as much as 181 miles.  Imagine filling up the tank/battery in you SUV enough for, say, 500 miles and doing so for $3.50 -- about the cost of a mochachino.


Brian

[Kleinstein]

The waste heat from solar cells is at a low temperature, the lower the better. For the normal Si cells, as expected the efficiency goes down if they run hotter. There is very little energy that can be taken out from this heat, even with an thermodynamical ideal conversion - in real world it's probably better to just improve the cooling, if there is a better heat sink.

The temperature effect is causing quite some confusion, as the efficiencies cited for lab samples are at 20 C cell temperature, while real world conditions usually call for a much higher temperature when the sun is shining.

The second article is a little better, but not much.
The articles look a little like someone misinterpreted the intense heat part. Starting at a few 1000 F down to just uncomfortably warm.

It's a little like those "Amarican scientist found ..." jokes.
Sadly one regularly finds these extremely exaggerated claims about potential benefits in applications for research money. This often goes to the level of promising nearly everything they think is not obviously wrong (or not easy to prove wrong). So the less they know the better the promises.
The real jokes come out if a journalist is involved making it even more spectacular.

[raptor1956]

Yeah, the 80% claim just jumps out at you as impossible and the conversion of otherwise waste heat into electricity at anything like 80% efficiency just can't be.  This does have some of the feel of 'cold fusion' of the late 80's early 90's.  Still, if they actually can convert some of the otherwise waste heat into electricity after first extracting 20% to 34% that would be a great thing. The reports are from about 3 years ago and nothing much new appears to have developed so it does appear closer to cold fusion than reality.


Brian

[Kleinstein]

The waste heat from PV is really cold side waste heat with essentially not temperature difference to a useful heat sink. So more like think about some 1 K temperature difference and thus 0.3% theoretical efficiency to use is. Lowering the cell temperature by this amount could well be more efficient and is definitely more cost efficient.

[Marco]

Considering that the majority of home energy use is thermal in nature - HVAC and hot water, thermal storage can work very well. What's lacking is incentive to invest.

If you have to carry heat over for a substantial amount of time (months say) you'll need to benefit from cubed laws and use district heating ... and that's a bit pricey.

[raptor1956]

Considering that the majority of home energy use is thermal in nature - HVAC and hot water, thermal storage can work very well. What's lacking is incentive to invest.

If you have to carry heat over for a substantial amount of time (months say) you'll need to benefit from cubed laws and use district heating ... and that's a bit pricey.

One potential residential co-generation concept would employ solar cells cooled by water and the waste heat used for domestic heating and hot water.  If the cells were 18-20% and you can reclaim waste heat you could up the net efficiency to, perhaps, 25%.  You would need some storage for that but it wouldn't be storage over months, just hours.  Of course, in a residential solar system the solar cells will usually be locked at a fixed angle so the true net would be perhaps 70% of the max or 13ish% + a few percent for heat.  Still, for a typical roof with 100 m^2 of Sun facing roof you'd still be pulling in about 13KW over 4.5-9 hours per days depending on location.  As mentioned before the average daily electric usage is about 30KWHrs so such a system would provide all the home needs and also provide daily commute energy for a couple electric cars.


Brian

[Someone]

One potential residential co-generation concept would employ solar cells cooled by water and the waste heat used for domestic heating and hot water.  If the cells were 18-20% and you can reclaim waste heat you could up the net efficiency to, perhaps, 25%.  You would need some storage for that but it wouldn't be storage over months, just hours.  Of course, in a residential solar system the solar cells will usually be locked at a fixed angle so the true net would be perhaps 70% of the max or 13ish% + a few percent for heat.  Still, for a typical roof with 100 m^2 of Sun facing roof you'd still be pulling in about 13KW over 4.5-9 hours per days depending on location.  As mentioned before the average daily electric usage is about 30KWHrs so such a system would provide all the home needs and also provide daily commute energy for a couple electric cars.

You're out by quite a margin for the energy needed to run multiple cars, average transport use is surprisingly high:
http://www.withouthotair.com/c3/page_29.shtml
that same publication even has an example of a two person house that went full solar:
http://www.withouthotair.com/c6/page_40.shtml
268m²

There are dual function solar+thermal panels available from several manufacturers, but mostly at "low" temperatures so you can't run a heat pump or high temperature storage from the thermal loop. It was a more promising option with concentrator systems but the price of silicon cells is so low that those are a historical footnote now. Very few buildings make good use of their already existing thermal resources so a few well placed fans and/or louvres can be a better investment.

[NiHaoMike]

If you have to carry heat over for a substantial amount of time (months say) you'll need to benefit from cubed laws and use district heating ... and that's a bit pricey.

A few days of storage will help a lot for leveling the load. More than that goes well into diminishing returns.

[james_s]

Using water to cool the solar panels and then using that water to heat something relatively low temperature like a swimming pool might be an effective setup although obviously the number of people who can make use of such an arrangement is going to be rather small. It might also work well to heat a water tank that then feeds the warm water into a conventional hot water heater or tankless unit, reducing the need for fuel without needing the solar panels to be as hot as you want the water.

[Kleinstein]

Cooling the cells with water would only make sense if there is a relatively low temperature heat sink. So something like the swimming pool, or initial warming (to some 20-25 C) of warm water.  A heat pump system could be another case, either direct, or possibly a underground loop that takes heat from the ground in winter.

The gain in efficiency should not be very high. I have not looked up the numbers, but it should be in the order of 0.3 % per degree (about the Carnot limit for the extra temperature step). Also take into account that claimed efficiencies for solar cells are usually already for low temperature operation. So its more than a nominally 20% cell would give you 20% maybe 21% with cooling instead of some 16 % if they run some 10-20C hotter.

Combined heat and PV to get the heat out at a higher temperature would reduce the PV efficiency quite a bit.

Given the relatively low costs for the cells, extra effort in cooling is likely not worth it, unless one has a good use for the low temperature heat. Water pipes and anti-freeze measures can be quite a hassle.

[Someone]

Given the relatively low costs for the cells, extra effort in cooling is likely not worth it, unless one has a good use for the low temperature heat. Water pipes and anti-freeze measures can be quite a hassle.

Well said, peak junction temperatures of typical panels are 60 degrees or so at full insolation and most of the time well below that so its low grade heat. The gain in efficiency of the PV cells can't cover the costs of a complex cooling system.

[thm_w]

One potential residential co-generation concept would employ solar cells cooled by water and the waste heat used for domestic heating and hot water.  If the cells were 18-20% and you can reclaim waste heat you could up the net efficiency to, perhaps, 25%.  You would need some storage for that but it wouldn't be storage over months, just hours.  Of course, in a residential solar system the solar cells will usually be locked at a fixed angle so the true net would be perhaps 70% of the max or 13ish% + a few percent for heat.  Still, for a typical roof with 100 m^2 of Sun facing roof you'd still be pulling in about 13KW over 4.5-9 hours per days depending on location.  As mentioned before the average daily electric usage is about 30KWHrs so such a system would provide all the home needs and also provide daily commute energy for a couple electric cars.

You're out by quite a margin for the energy needed to run multiple cars, average transport use is surprisingly high:
http://www.withouthotair.com/c3/page_29.shtml
that same publication even has an example of a two person house that went full solar:
http://www.withouthotair.com/c6/page_40.shtml
268m²

Depends on the person and location, Leaf can do 19.1 kW⋅h/100 km so two people driving 15km commute * 2 ways = 11.5 kWh. 20kWh left over, not quite enough for the average home (30kWh) but enough for an energy efficient household.
The median commute here is 9km. I'm sure in Aus its higher.

[Someone]

One potential residential co-generation concept would employ solar cells cooled by water and the waste heat used for domestic heating and hot water.  If the cells were 18-20% and you can reclaim waste heat you could up the net efficiency to, perhaps, 25%.  You would need some storage for that but it wouldn't be storage over months, just hours.  Of course, in a residential solar system the solar cells will usually be locked at a fixed angle so the true net would be perhaps 70% of the max or 13ish% + a few percent for heat.  Still, for a typical roof with 100 m^2 of Sun facing roof you'd still be pulling in about 13KW over 4.5-9 hours per days depending on location.  As mentioned before the average daily electric usage is about 30KWHrs so such a system would provide all the home needs and also provide daily commute energy for a couple electric cars.

You're out by quite a margin for the energy needed to run multiple cars, average transport use is surprisingly high:
http://www.withouthotair.com/c3/page_29.shtml
that same publication even has an example of a two person house that went full solar:
http://www.withouthotair.com/c6/page_40.shtml
268m²

Depends on the person and location, Leaf can do 19.1 kW⋅h/100 km so two people driving 15km commute * 2 ways = 11.5 kWh. 20kWh left over, not quite enough for the average home (30kWh) but enough for an energy efficient household.
The median commute here is 9km. I'm sure in Aus its higher.

Median and average distances are not dissimilar across Australia and Canada. But commuting is only one part of the transport energy use. Household energy use for the houses I've lived in averages under 10kWh a day (2-4 people) so its possible to get right down but thats not matching the actual lifestyles the general population lives. 100m² of solar is around the usable space on the average house, but doesn't meet the average demands.