March 06, 2008
Prospects For Solar Thermal Power
A new solar thermal electric power installation in Boulder City Nevada uses arrays of mirrors to concentrate sun light to drive electric power generation. The cost of electricity for this plant is estimated at 15-20 cents per kilowatt-hour (kwh).
Many states, including California, are imposing mandates for renewable energy. All of that is reviving interest in solar thermal plants.
The power they produce is still relatively expensive. Industry experts say the plant here produces power at a cost per kilowatt- hour of 15 to 20 cents. With a little more experience and some economies of scale, that could fall to about 10 cents, according to a recent report by Emerging Energy Research, a consulting firm in Cambridge, Mass. Newly built coal-fired plants are expected to produce power at about 7 cents per kilowatt-hour or more if carbon is taxed.
That is at least double what cheaper sources of electricity cost in the United States. Can the costs really go down substantially with a bigger market?
While solar thermal still costs more than wind power predictable daylight hours and the ability to store the heat allows solar thermal to provide a more reliable power source.
According to the U.S. Department of Energy, wind power costs about 8 cents per kilowatt, while solar thermal power costs 13 to 17 cents. But power from wind farms fluctuates with every gust and lull; solar thermal plants, on the other hand, capture solar energy as heat, which is much easier to store than electricity. Utilities can dispatch this stored solar energy when they need it--whether or not the sun happens to be shining.
Solar thermal doesn't have to be able to provide electric power 24 hours per day to be useful. If its cost could drop in half then solar thermal would greatly reduce the use of coal and natural gas and allow limited fossil fuels to last longer and pollute less..
One solar thermal facility in Nevada is claimed to use 400 acres for enough electricity to power 14,000 homes.
Acciona's plant, which began operation last year, produces 64 megawatts of electricity for the utility company Nevada Power, enough to light up 14,000 homes. The company's Spanish competitor Abengoa just announced a plan to build a 280-megawatt solar thermal plant outside Phoenix, which would be the largest such project in the world.
All you need is a lot of sun, a lot of space and a lot of mirrors — and NS1 has all of the above. 182,000 parabolic mirrors are spread over 400 acres of flat desert, creating a glistening sea of glass visible from miles away.
That's 35 homes worth of electric power per acre of land. Mind you, this is an area of the United States that gets above average amounts of sunlight. But this result suggests that use of solar thermal to power all homes would not use an inordinate amount of land - at least not in countries with lower population densities.
Solar thermal looks cheaper than solar photovoltaics and the heat from solar thermal can be stored to stretch into evening hours. But solar photovoltaics might have better prospects for lower cost reductions and it lends itself more easily to decentralized use and smaller installations on homes and other buildings.
The most important thing is the price trends, while coal is cheaper now it has not been dropping in price as fast as wind and solar.
All this makes me thing of Total Annihilation!
The deserts in the southern states are perfect for solar power generation. This sounds too far from civilization in the north, but this electricity can be made transportable because recently, thanks to nano-technology, hydrogen production by means of electrolysis became 85 % efficient"
Hydrogen is a very fine gas, and a thin pipeline can be used to carry hydrogen to northern cities for heat. Then this hydrogen can be used to regenerate electricity.
Alternatively, by using trains, we can carry millions of charged batteries back and forth from the south to north for electric cars. The empty batteries would be shipped by train back south to get charged.
With much less than half the money wasted in Iraq, we could have built hundreds of gigawatt thermal solar power plants, as well as nuclear power plants.
solar thermal plants, on the other hand, capture solar energy as heat, which is much easier to store than electricity. Utilities can dispatch this stored solar energy when they need it--whether or not the sun happens to be shining.
How is this possible? Heat is very expensive to store. One can build a large insulated tank to store heated water, but that's a very expensive endeavor. The problem of heat storage is encountered by people interested in passive solar homes that try to collect enough heat during the summer to last throughout the winter. It turns out to be very expensive to store heat. One can build a large water tank beneath the house, as one of the early prototype MIT solar homes did, but that is usually not economically viable. One can also store heat in soil, as earth-sheltered homes do, or as the Drake Landing Solar community do (in a thermal borehole, http://www.dlsc.ca), but these are also expensive options.
I don't know what technology they are referring to, when they say that heat "is much easier to store than electricity".
If the claim that electrolysis can be made 85 % efficient is true (this is claimed by QuantumSphere mentioned in the article I have pasted above), then it follows that energy can be stored for a very long time after it is captured by solar thermal plants.
Additionally, this energy can be transported in many ways:
1) Thin pipelines that can carry this hydrogen (generated by electrolysis at only 15 % loss of energy) to northern states
2) Millions of electric batteries can be charged in southern states and then these can be economically shipped by train to northern states. Empty batteries would be carried back by the same trains from northern states to southern states.
3)Superconducting wires can carry electricity to northern states with lo loss of power.
Solardude: Think carefully. What is the best way to store heat? Molten salts. Phase change materials. Use very good insulators. An Australian company is using huge insulated blocks of high-grade graphite they manufacture themselves. Course once they get redox flow cells working to scale, large-scale PV will worth doing. Since solar radiation ranges from UV through IR wavelengths, the best approach is to try to grab the whole spectrum. Nano PV plus regular PV plus solar thermal.
Wolf-Dog: Hydrogen is not an energy-dense fluid unless liquified. Shipping charged batteries is a no-starter--much too expensive. But the idea of superconductors plus HVDC transmission lines makes a lot of sense.
Piping hydrogen would be a problem. H2 is such a small molecule that it tends to diffuse out of even tight containers. I imagine transport losses would be huge in a pipeline with thousands of joints over long distances.
And that's if the electrolosis is really 85% efficient, and scalable.
Rather than complex schemes to move energy north we should first maximize solar thermal where it works well. That releases fuel for use farther north and for use at night in the south.
I advocate much more nuclear myself. If not, burnable fuels will remain primary at high latitudes.
Heat storage may require a compromise. Perhaps, heat water during the day and feed it to conventional generators at night? Those generators would burn some fuel to heat the water to steam; but much less because the water would be hot coming in. The heat storage is an engineering problem; numbers will select the best method.
The energy problem will be solved by continous improvements and invention rather than by a total economic revolution. A lot of traditional carbon fuel will be used for a long time.
Yes, until solar can compete in the most intense sunshine states there's no point in worrying about transporting electricity great distances.
Closer to the poles our choices are coal or nuclear and to a lesser extent wind.
The price trend for coal electric is upward. The construction material costs have skyrocketed and coal prices have risen as well.
Nuclear faces the same problem as coal with construction materials costs but even more so since nuclear is more capital intensive.
Think carefully. What is the best way to store heat? Molten salts. Phase change materials. Use very good insulators.
Big monkey: Like other solar builders, I have looked at PCM materials such as paraffin and eutectic salts (namely, Glauber's salt) for storing heat. Unless there has been a breakthrough I am not familiar with, these materials are fraught with problems. They are highly corrosive, they often loose their phase-changing capability, and building a large insulated stainless steel that can resist corrosion but yet store that much heat is not economically viable. The only PCM material that comes close to being economically viable is Micronal, the PCM drywall developed by BASF, but there has not been enough long-term experience to determine whether Micronal drywall is safe or stable enough for human occupancy.
Think about it. The concept of solar heating and solar homes have been around since the days of cob and adobe, yet except for some pleasant climates with low heating and cooling demands, most homes built depend upon fuel for heating requirements. While super-insulation goes a long way towards reducing energy usage, builders and manufacturers are still far from designing homes that can satisfy all of their heating requirements from solar. The ones that do not require supplemental heating are either heavily bermed so that most of the house is covered in soil, or devote so much of their southern exposure to solar collection, that they give away all of their views of the outdoors. Neither of these options are attractive to homeowners, thus we are still waiting for a breakthrough in solar thermal storage.
You are mixing up the economics of what makes sense in individual houses with what makes sense at large solar plants. As a home solar installer your orientation is toward the home. But think about the large plants (which is what I'm writing about here).
Concentrators are cheaper than photovoltaics at large solar plants. But concentrators do not work on houses. At the same time, large solar plants can manage the complexities of thermal storage even though that does not work for houses.
Each type of tech has its own best scale of operation. Molten salts need insulation. Well, surface area of insulating material does not go up as fast as volume. So the bigger the thermal storage unit the cheaper - at least up to some size way bigger than a single home's thermal storage unit.
The same thing pertains to batteries and cars versus stationary storage. There are types of batteries that are too heavy and too dangerous for use in cars that can be used in stationary plants operated by electric distribution networks.
this electricity can be made transportable because recently, thanks to nano-technology, hydrogen production by means of electrolysis became 85 % efficient
Which does nothing for the pumping losses in transmission or the inefficiency of re-conversion to electricity.
China is building an 800 kV HVDC line with transmission losses of just 7%, and of course there are no chemical conversions involved. If solar thermal is used with e.g. hot-water storage, there is no reason to consider hydrogen.
You are mixing up the economics of what makes sense in individual houses with what makes sense at large solar plants.
Very good points. All of your points are well taken.
Interestingly, at another site I occasionally read, this development of using molten salt was also introduced recently. One of the commenters (Abe Lincoln) made the points we've just gone over: how PCM is not economically feasible for homes, but may be for large power plants.
This isn't really a very viable thermal battery for small scale applications . If you are using solar thermal, the point of contact of the beam has to be greater than 1000 deg F, which requires quite a few reflectors depending on the contact area. Otherwise you need a very high temperature heating coil (expensive) and are converting between 2 or more modes of energy transfer.
Since it is stored at such high temperatures you need really really good insulation, and even then the losses due to thermal gradient will be quite large. The molten salt thermal battery only really works efficiently at a large scale.
"The construction material costs have skyrocketed"
Yes, but this applies to all energy conversion, storage and transmission systems. Carpeting the Sonoran Desert with solar plants, building Gigawatts of energy storage facilities and thousands of miles of high voltage, long-distance transmission lines, will also be frightfully capital intensive. There are no cheap ways out. I have no clear idea of what will be the most cost efficient way to proceed, although I suspect that it is nuclear, which needs far less in the way of storage and transmission. Anyone who can provide us with sufficient cost information will be doing a real service to the discussion.
Randall: Although there is a lot of enthusiasm for solar thermal power, there is one issue that I have not seen discussed. Cooling is an important part of any thermal generation system. The efficiency of such systems is limited by the availability of cooling. Most fixed thermal systems use water for cooling. Of course deserts are characterized by the lack of water*. So how will solar thermal power systems be cooled? One possibility is to heat the working fluid during the day and use the cold of the night sky to do the cooling.
*places like the Peru, Chile, Argentina, Namibia, and Australia have deserts adjacent to oceans, which might serve as sources of cooling. But other areas like the Sonoran Desert and Central Asia are too high and far from water.
Construction costs: I want to know more about where these cost increases are coming from. What makes the cost of a coal plant go up by 50% to 100%? Steel? Copper? Skilled laborer shortages? Bottlenecks in manufacturing turbines? What exactly?
If the cost increases are coming from higher steel costs due to Asian demand then that'll impact nukes more than solar.
Also, solar has much more potential to drop in cost because it is my impression the bulk of the cost is in fabrication. Get away from silicon crystals and use some thin films and self-assembling nanodevices and the actual amount of mass could go way down and the amount of labor could go way down too.
By contrast, how are nukes going to use less than massive amounts of concrete and steel? But then how much of a nuke's costs come from concrete and steel? I'd really like to understand this better.
I'd like to know the ratio of steel and/or aluminum to megawatt of capacity for various electric power generation methods.
To amplify on my original reply about small versus large facilities: A sphere's surface area goes up at 4*pi*r^2 whereas its volume goes up at (4/3)*pi*r^3. That difference between the squared and the cubed means that make a sphere big enough and your surface area is way smaller than your volume. Your insulation material needs aren't as great with a big object.
This makes intuitive sense. Take two rectangular homes and make them share a wall. Suddenly that shared wall area ceases to be a surface that leaks heat or cold. That's why multi-unit dwellings are more energy efficient.
"If the cost increases are coming from higher steel costs due to Asian demand then that'll impact nukes more than solar."
I do not see that this intuitively obvious. Yes, nuclear plants use lots of steel, concrete and copper, but they cover acres not square miles. The solar plant you refer to covers 400 acres (0.625 sq. mi.) with mirrors. The cost of just the frames to hold the mirrors and the motors to tilt them will be large. Of course every solar installation needs to be paired with a storage or fast rise generating system (hydro or gas turbine) which is another capital cost that must be accounted for.
"Also, solar has much more potential to drop in cost because it is my impression the bulk of the cost is in fabrication."
That is true for the PV panels, but it does not hold for solar thermal. Further, even PV has to be mounted, connected to the grid etc. If the PV cost no more than asphalt roofing tiles, the cost of thousands of square miles of them would be enormous, further there would still be a storage and transmission problem.
"By contrast, how are nukes going to use less than massive amounts of concrete and steel? But then how much of a nuke's costs come from concrete and steel? I'd really like to understand this better."
So would I, but remember that nuclear is a young technology, one whose development has been retarded by malign political influences. I have read that pebble bed reactors are far less demanding of containment than current boiling and pressurized water reactors. Further, other construction methods can substitute for some concrete and steel.
I once proposed digging a tunnel in the bedrock under New York City from New Jersey on one end to Long Island on the other. Galleries on the sides of the tunnel could be fitted with nuclear reactors, which would need limited containment. An ample supply of cooling water could be drained from the Hudson and East Rivers. If one of the reactors suffered a core meltdown or other adverse event, its gallery could be sealed off, ditto for decommissioning. The tunneling alone would cost billions of dollars, but it might be worthwhile if the alternative is not having enough electricity to run the city. Further, with the reactors only a few hundred feet from the city, they could serve as a source of live steam for the existing municipal heating system and to allow for its expansion to replace oil fired boilers.
Nah, some things are just so obviously right that many people discover them independently.