The bright future of solar powered factories

July 30, 2011 7:24 AM

Most of the talk about renewable energy is aimed at electricity production. However, most of the energy we need is heat, which solar panels and wind turbines cannot produce efficiently. To power industrial processes like the making of chemicals, the smelting of metals or the production of microchips, we need a renewable source of thermal energy. Direct use of solar energy can be the solution, and it creates the possibility to produce renewable energy plants using only renewable energy plants, paving the way for a truly sustainable industrial civilization.

Low-Tech and No-Tech Magazines previously: 1, 2, 3.

posted by Bangaioh (31 comments total)

Sounds like horsecrap. Electricity can be turned into thermal energy with 100% efficiency, but the reverse isn't true; the Second Law of Thermodynamics prevents thermal energy from being converted to work at anywhere close to 100% efficiency. We need low-entropy energy, not thermal energy. Maybe there are other arguments for avoiding heat generation from work, but this one doesn't pass the sniff test.
posted by Mapes at 7:31 AM on July 30, 2011

Maybe I'm wrong, but I always thought that the focus on solar electricity was because solar electricity is difficult and versatile, while solar heating is easy and already available to whoever can be bothered to set up their shit to make use of it.

I can see that it's not quite that cut and dried, and that there is low-hanging-fruit thermal fossil fuel use here that ought to be switched to solar, and perhaps there are some sectors that aren't getting the pressure to do so as some other sectors, so maybe it's a good drum to be beating, even if the sound of that drum is "duh".
posted by anonymisc at 7:37 AM on July 30, 2011

Sounds like horsecrap.

The majority of utility-scale solar electricity is done through heat and not PV. Most people are confused about this. PV is good for roof tops and satellites, but for large scale energy production heat is the cost efficient way to go.
posted by stbalbach at 8:26 AM on July 30, 2011

The limitation with solar thermal is that the energy has to be used close to where it is produced. We don't have a grid for heat energy that we can sell back into. Solar or wind electricity can be sold to utilities and consumed elsewhere.

I am a solar thermal designer in my day job. Solar thermal is more efficient and cost effective than solar photovoltaics, but you can't just cover a roof top with thermal collectors the way you can with electric panels because all of the energy has to be used on site. This means a careful matching of production, storage and consumption, and this adds much more complexity to projects as compared to PV.
posted by fzx101 at 8:28 AM on July 30, 2011 [7 favorites]

Solar furnaces can produce temperatures up to 3,500 °C (6,332 °F), enough to manufacture microchips, solar cells, carbon nanotubes, hydrogen and all metals (including tungsten which has a melting point of 3,400 °C)...

He's missing an obvious point - the objective is not to produce high temperatures per se, it's to produce high thermal energy output. These are two different things. A temperature by itself is meaningless without specifying how much material the furnace can process.

In other words, a solar furnace might be able to smelt tungsten at 3400 degrees but if it's only a thimbleful, well.....
posted by storybored at 8:29 AM on July 30, 2011 [4 favorites]

We don't have a grid for heat energy that we can sell back into.

Would it make sense to create one?
Areas of the eastern Bloc have this, and my expectation is that they have it because it's very efficient, but the kind of very efficient thing that doesn't tend to happen without a lot of central planning.
posted by anonymisc at 8:37 AM on July 30, 2011

The majority of utility-scale solar electricity is done through heat and not PV. Most people are confused about this. PV is good for roof tops and satellites, but for large scale energy production heat is the cost efficient way to go.

Did you look at the link? The person is arguing against using electricity and for using an assembly of mirrors and lenses to channel the sun's rays anywhere one needs a higher temperature.
posted by Mapes at 8:39 AM on July 30, 2011

Sounds like horsecrap. Electricity can be turned into thermal energy with 100% efficiency, but the reverse isn't true; the Second Law of Thermodynamics prevents thermal energy from being converted to work at anywhere close to 100% efficiency. We need low-entropy energy, not thermal energy. Maybe there are other arguments for avoiding heat generation from work, but this one doesn't pass the sniff test.

I think this is missing the point, which is why convert solar energy into electricity and then back into use in heat applications? This is where the inefficiency comes in to the system. Systems to do this are expensive, while solar heating is still not economically competitive but is closer to being economic than PV. The article gives some stats but it is worth pointing out that pretty much every country on earth will use more energy for heating than it will for electricity. Typical European countries will use around as much energy for heatnig as they will electricity and transport energy combined. (I don't have th data for the US but I would be surprised if it was any different.) Heating demand accounts for about 40-50% of all climate change emissions in these countries. Yet nearly all renewable energy policy has so far focussed on renewable sources of electricity. Hardly any nation globally has a national policy to roll out greater renewable heat sources, and pretty much no-one has started to consider the more complex issues of regulation and market frameworks that increasing renewable heat sources will require if the many barriers to greater uptake are to be overcome.

while solar heating is easy and already available to whoever can be bothered to set up their shit to make use of it.

Not really, its not economic in most places without subsidy. The exception in in parts of China, and (IIRC) about 80% of global installed capacity is located there (and built there). In that respect, solar thermal is one of the more technologically mature renewable energy sources of heat (RES-H), and it is on a par with wind in terms of global installed capacity, with both these technologies massively outstrip other renewable technologies. However, solar thermal is not commercially mature in most of the western world and it will need some help to become so.

The majority of utility-scale solar electricity is done through heat and not PV.

Do you have some data to show this? Most of the utility owned solar in Spain and Germany is PV rather than concentrated solar power electrcity generation. I am only aware of a few CSP plants.

Regarding temperatures, there are lots of problems as regards the potential to substitute RES-H for electricity or other heating in industry. Firstly, it is very difficult to get good data on what the demand for heat energy is like at particualr temperatures. I have just finished a project looking at RES-H Policy in 6 EU Memeber States and we struggled to find good data for any of them. However, what we could find indicates that only a very small amount of total heat energy goes into achieving temperatures about 600C, I think we reckoned it was about 1%. Finding technology to apply in the 200-600C range might be more useful (and easier) in terms of displacing more carbon emissions, but there are also huge potential gains from using RES-H below 200C and this can be achieved or partly achieved through multiple RES-H technologies. The partly achieved is important, there is huge potential for partial heating of water for industrial processes, and since industry tends to have big flat roofs then there is an obvious place to install solar water heaters for this purpose.
posted by biffa at 8:40 AM on July 30, 2011 [5 favorites]

Transporting heat any significant distance is extremely difficult. Transporting high heat around a factory to where you need it is difficult. Setting up all your high-temp processes so that they are at the focus of a multi-acre solar concentrator is difficult. Making everything to have a non-solar backup is expensive and complicated. Diluting heat is easy, but re-concentrating it is hard. Direct solar heat use in industry has some niche applications; direct solar heat for low-heat use like housing is moderately useful.
posted by a robot made out of meat at 8:43 AM on July 30, 2011

We don't have a grid for heat energy that we can sell back into.

Would it make sense to create one?
Areas of the eastern Bloc have this, and my expectation is that they have it because it's very efficient, but the kind of very efficient thing that doesn't tend to happen without a lot of central planning.

It might do but it would require quite a lot of political will to do it in many places. Essentially, someone would have to make the investment in the grid structure. Putting in place the regulation so that the investor would be able to be able to this, and to do so with a financial return would require political support. It would effectively require the same level of govertnment buy-in that setting up gas or electricity networks requires, but without (at least currently) any political will that attaches to those networks. The free market energy provision now common in the US and EU then throws up all sorts of problems with regulatory oversight of the new markets created, third party access to the grid, customer freedom to choose suppliers, etc. And the grid would cost a lot, and it would need to be upfront which is always a problem. None of this is insurmountable though, but it is difficult.
posted by biffa at 8:46 AM on July 30, 2011

Sounds like horsecrap. Electricity can be turned into thermal energy with 100% efficiency, but the reverse isn't true; the Second Law of Thermodynamics prevents thermal energy from being converted to work at anywhere close to 100% efficiency. We need low-entropy energy, not thermal energy. Maybe there are other arguments for avoiding heat generation from work, but this one doesn't pass the sniff test.

What possesses people to write crap like this? The most efficient solar panels to pout at 43.5%. That's the world record. Ordinary commercial cells will get you 10-20% efficiency.

On the other hand, converting sunlight into heat is nearly 100% efficient. All you need is a black surface (minus heat loss to the environment before it does what you want)

We don't have a grid for heat energy that we can sell back into.

Interestingly, NYC has a steam tube network. You could conceivably come up with a bi-directional steam flow meter and let people sell heat energy directly onto the grid. Right now the stuff is sold by general electric, which they heat by burning natural gas or diesel, I think. They could add solar panels pretty easily.
posted by delmoi at 8:56 AM on July 30, 2011 [1 favorite]

We don't have a grid for heat energy that we can sell back into.


Common, just not common knowledge. However, using solar to power such systems might be difficult.
posted by Chuckles at 9:01 AM on July 30, 2011

I think this is missing the point, which is why convert solar energy into electricity and then back into use in heat applications? This is where the inefficiency comes in to the system.

Because I want controllable heat any time, anywhere with a power outlet? When I want to solder something, believe me, I don't funnel and focus sunlight onto a circuit board. I plug in a soldering iron.

I can appreciate that when no electrical power is available, when sunlight is readily available, and when the need for higher temperature is static, non-moving, predictable, and long-term, then it's a clever idea to use the sun to heat the site instead of constructing a power plant. But the authors didn't begin with this practical motivation with its obvious constraints. They went with misleading statements about how it's inefficient to turn electricity into heat (there's nothing more efficient when electricity is available!) and softpedaling the inconvenience and expense of of heat-transferring infrastructure while ignoring the tremendous existing electricity infrastructure. Later in the article, they "rediscover" electricity production by solar steam generation, perhaps forgetting that earlier they were bemoaning the resources required to build turbines.

I just think this idea could have been presented much better.*

*I'm writing my thesis. All I've been thinking about recently is how to propose new ideas without arousing scorn.
posted by Mapes at 9:05 AM on July 30, 2011 [1 favorite]

Because I want controllable heat any time, anywhere with a power outlet? When I want to solder something, believe me, I don't funnel and focus sunlight onto a circuit board. I plug in a soldering iron.

Uh, does your soldering iron use more power then your hot water heater or your oven?

I just think this idea could have been presented much better.*

I don't mean to be rude but your initial response was completely idiotic. People can't control how idiots are going to respond to what they write, and they shouldn't try. Solar cells are inefficient at converting sunlight into electricity, and you seem to be unaware of that. If you have to heat something up it makes more sense to use heat generated by sunlight rather then converting sunlight into electricity at a low efficiency, then into heat.
posted by delmoi at 9:09 AM on July 30, 2011

Maybe I'm being overly simplistic because I'm not a scientist, just an average joe who briefly wrote a few grant applications for businesses wanting to supplement with solar...I understand that solar cell generated electricity isn't the most efficient form of producing electricity, but wouldn't a 15%, or even more optimistically, a 25% replacement of electricity consumption by solar be a worthwhile goal to aspire to, especially when coupled by solar thermal, which is incredibly efficient? It is Germany's successful integration of solar-generated electricity into their marketplace that has enabled them to close seven nuclear power plants in the wake of Japan's disaster, and to announce the closing of all their nuclear power plants in the next decade? So what if solar's not 100% efficient or a 100% perfect solution? It is part of a renewable mix and has enabled at least one country to wean itself off of nuclear power. Isn't that a goal worth emulating worldwide?
posted by zagyzebra at 9:22 AM on July 30, 2011 [1 favorite]

Direct solar heat use in industry has some niche applications; direct solar heat for low-heat use like housing is moderately useful.

The latter is where the low-hanging fruit is. If we built houses with the sun in mind (with sun-facing windows or passive heating walls, skylights, and/or even hobbit-hole Earthship style homes, etc) we'd spend a lot less to live in them, and be more comfortable through most of the year.
posted by vorfeed at 9:27 AM on July 30, 2011 [1 favorite]

Uh, does your soldering iron use more power then your hot water heater or your oven?

What does it matter? Or is that a limitation of this approach, that it's only remotely practical for large-scale processes or small temperature increases?

Solar cells are inefficient at converting sunlight into electricity, and you seem to be unaware of that. If you have to heat something up it makes more sense to use heat generated by sunlight rather then converting sunlight into electricity at a low efficiency, then into heat.

I'll take that inefficiency for the ability to connect the solar cell to the load with a wire instead of some type of sunlight tube. It may make sense to you, but I'll pass on the idea that I'll walk into my lab Monday morning and propose moving all the ovens up to the roof so we can shine mirrors at them.
posted by Mapes at 9:33 AM on July 30, 2011

I'll take that inefficiency for the ability to connect the solar cell to the load with a wire instead of some type of sunlight tube. It may make sense to you, but I'll pass on the idea that I'll walk into my lab Monday morning and propose moving all the ovens up to the roof so we can shine mirrors at them.

This clearly makes the most sense for new buildings -- you know, the kind which don't already exist, and could therefore conceivably be designed around a central system for distributing the heat of the sun. How about we discuss that possibility, instead of pretending as if people are actually suggesting dragging all our existing ovens onto existing roofs so we can bake roofcakes?
posted by vorfeed at 9:41 AM on July 30, 2011 [1 favorite]

Solar cells are inefficient at converting sunlight into electricity, and you seem to be unaware of that.

delmoi, we get it. You've completely stomped on two people who made statements completely orthogonal to the efficiency of PV cells, specifically pointing out that that the conversion from electricity to heat, and including nothing more than that specific part of the energy generation and use process, is very efficient. Maybe there's some other implicit assumption there that set you off, but I think that could be handled with "Yes, but there are also problems X,Y,Z" or "No that's not really a problem, because Y", rather than "That's idiotic, here's another statement about the efficiency of a different process". If your point is about the efficiency of the entire generation-transmission-use chain (I think? I really can't tell.), then please make that point.
posted by kiltedtaco at 9:47 AM on July 30, 2011

Doesn't solar thermal usually refer to using the sun's heat to boil water and power a steam turbine or power a Carnot engine to generate an electricity?

Another thing to consider is thermodynamics. It is useless to talk about heat energy without talking about how it is distributed. It takes energy to concentrate heat, to take heat from something colder and make it warmer. So, for instance, if you needed heat for an oven, or industrial process, you would have to supply a substance at that temperature, or you would end up having to use electricity anyway. That's not to say there isn't a lot of room for making use of lower density waste heat from, say, a power plant or industry, and somehow using that in a productive fashion, i.e. warming homes.
posted by Zalzidrax at 10:02 AM on July 30, 2011

Direct solar heat use in industry has some niche applications; direct solar heat for low-heat use like housing is moderately useful.

No, large scale usage of solar thermal is possible and effective in both applications. There is huge demand for low grade heat - it represents the majority of heat use in industry and ~100% in the domestic sector, and this can be addressed by solar in a large number of cases. Solar thermal seems likely to become the dominant renewable heat source is proper policy mechanisms are provided across the sector. The models we produced in our recent project suggest solar could easily deliver 50+ TWh by 2030 in the UK alone.
posted by biffa at 10:15 AM on July 30, 2011

That's not to say there isn't a lot of room for making use of lower density waste heat from, say, a power plant or industry, and somehow using that in a productive fashion, i.e. warming homes.

Yeah, I think this is really worth noting, a couple of people have mentioned district heating and while this could be used to supply renewable heat, including from solar thermal, the first stage in any sane world would be to use the 60-65% of heat from fossil fuels that goes up into the atmosphere from our major power stations.
posted by biffa at 10:18 AM on July 30, 2011

120 W soldering iron needs (@$1.80/watt PV cost) a $216 PV panel.
5,000 watt home water heater needs $9000 worth of PV.

In 1980, the Knesset passed a law requiring the installation of solar water heaters in all new homes in Israel. These hot water heaters account for 4% of the total energy used; Israel now leads the world in the amount of energy derived from solar per capita.
posted by 445supermag at 10:58 AM on July 30, 2011

[few comments removed - if you can't comment without calling other people names, Do Not Bother. This is last warning. Thanks.]
posted by jessamyn at 12:02 PM on July 30, 2011

445: This is likely to be the model for renewable heat, including soalr thermal, for Europe. Spain and Germany have policies along this line, Austria is likely to follow and this line of policy is mandated by the 2009 EU Renewables Directive as the default instrument for any nation which does not mandate an alternative.
posted by biffa at 5:59 PM on July 30, 2011

delmoi, we get it. You've completely stomped on two people who made statements completely orthogonal to the efficiency of PV cells, specifically pointing out that that the conversion from electricity to heat, and including nothing more than that specific part of the energy generation and use process,

No, read Mapes first comment. It makes no sense unless you assume that he doesn't understand that PV is much less efficient then directly heating things with the sun.
posted by delmoi at 8:11 PM on July 30, 2011

How easy is it to store heat energy? Say you ran water through black pipes on the roof, through a heat exchanger, and used the heat exchanger to heat up water in an insulated container. Would that let you balance summer cooling and winter heating?
posted by Joe in Australia at 2:14 AM on July 31, 2011

I believe ground-source heat pumps do something like that, using the soil as reservoir. David MacKay's book runs the numbers from page 301 onwards.
posted by Bangaioh at 3:25 AM on July 31, 2011 [1 favorite]

How easy is it to store heat energy?
Tricky. The utility grade solar heat farms (like the ones they are building in Chinchilla and Moree) struggle to find a suitable way to store power for when it is dark or cloudy. My reading is that they can produce electricity commercially, or close to it, when it is sunny, but the economics get killed by coal burners who can predictably generate whatever the weather. That the Chinchilla plant is combined with a gas burning turbine for off days and other similar plants use a molten sodium based heat well gives some idea of the challenges.
In terms of the idea to store summer heat insulated for winter use, it is a logistical problem.
Consider you stored the heat generated on your roof on a hot summer day. It might easily heat 1000 liters or more of water to 90 degrees (based on 2 square meter solar hot water system that heat a 300l tank). You could then pump this hot water through a hydronic radiator come a winter day to get a toasty room. The trouble is you probably need 1000L for each winter day, and assuming 90 summer days and 90 winter days, the 90 kiloliters you would need to store in an insulated tank is pretty sizeable (several backyard pool fulls), and not cost effective with current low fossil fuel energy prices.
The good news, of course, is the sun shines in winter as well as summer. You could probably store 1000L with a good size solar heat panel and capture enough energy to provide useful heat. A quick cost estimate:
- 1000L insulated storage and accompanying solar evacuated tubes = $4k
- hydronic heaters installed in two rooms - $1k
- labour to plumb it all in = $1k

So for about $6k you can have a heat system that warms two rooms in winter as long as there isn't a string of cloudy days (the third cloudy day probably cuts your heat by about 50%).
Worth it? Borderline, I think. I can heat a couple of rooms for $5 a day with natural gas with a $700 heater, so a 10yr+ pay back, not counting the odd days you need to supplement your solar heat because it is cloudy.
posted by bystander at 4:00 AM on July 31, 2011

No, read Mapes first comment. It makes no sense unless you assume that he doesn't understand that PV is much less efficient then directly heating things with the sun.

I assure you, the subject's been thoroughly covered in my training. kiltedtaco got it. I'm talking about the load location, comparing resistive heat generation vs. solar heating. If you want to include energy accumulation and transport, you should also acknowledge the inefficiencies of the solar heating approach. You're blithely talking about laying out a strip of optically black material (in a deleted comment), but it's more complicated than that, isn't it? If I expand into a new facility across the street that requires higher-temperature processing, I can string a wire across the street and use longer, narrower, or higher-resistivity heating elements to obtain these temperatures—if I'm using electric current. If I'm working with solar heating, however, I need to buy and build steam generation and energy reclamation units and lay steam pipes that are less efficient than copper wires. And if I want temperatures of hundreds or thousands of degrees, I need space for an entirely new assembly of mirrors and lenses at the new site.
posted by Mapes at 8:49 AM on August 1, 2011

I don't mean to be rude but ...

Fail.
posted by IAmBroom at 11:45 AM on August 1, 2011