Slashdot Mirror
World's First Molten-Salt Solar Plant Opens
An anonymous reader writes "Sicily has just announced the opening of the world's first concentrated solar power (CSP) facility that uses molten salt as a heat collection medium. Since molten salt is able to reach very high temperatures (over 1000 degrees Fahrenheit) and can hold more heat than the synthetic oil used in other CSP plants, the plant is able to continue to produce electricity long after the sun has gone down. The Archimede plant has a capacity of 5 megawatts with a field of 30,000 square meters of mirrors and more than 3 miles of heat collecting piping for the molten salt. The cost for this initial plant was around 60 million Euros."
Ok, so it can produce after the sun has gone down, but wouldn't the inverse be true, too, i.e. it'll take longer for it to reach a heat at which it can start producing in the morning? Anyone who didn't fail physics want to help an ignorant AC out?
I thought the headline said morton-salt.
LFTR's will render these things irrelevant. http://energyfromthorium.com/lftradsrisks.html
Oh, yeah! Wise guy, huh? Woob woob woob woob! Nyuk! Nyuk!
This is big news!
The larger the temperature difference, the more efficiently we can turn the heat into electricity. Superheated steam is just too difficult to manage over distances so this would make a great first step of collecting the solar energy and transporting it to a single location to make superheated steam.
The best part is that NaCl is non-toxic and doesn't need to be kept under pressure. If you have a natural gas Bunsen burner and good test tubes handy, it is just about possible to melt salt and prove to yourself how stable it is. Just be careful about spilling it because it is hot enough to get things like wood and paper to auto-ignite on contact. The hottest temperature you can expect to achieve with natural gas is around 700 degrees Celsius, if I remember correctly.
(as a side note, this is why low pressure nuclear power plants have such poor efficiency - because the water is only at 100 degrees Celsius after being heated by the nuclear fuel).
No sig. Move along - nothing to see here.
This one doesn't use salt, but we have had one very similar built here back in 2008. http://www.renewableenergyworld.com/rea/news/article/2010/07/areva-boosts-solar-supersteam-parameters-in-bakersfield
Life moves pretty fast; if you don't stop and look around once in a while, you could miss it. -FB
The article isn't specific about *what* salts they're using, but says "molten salts solidify at around 425 degrees F" - NaCl's melting point is about 800 C.
One of the articles they reference refers to another project that uses a mixture of sodium and potassium nitrates.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
The article is incorrect... Barstow had a molten salt plant in 1995 I believe. Excerpt from the Wiki - "1995 Solar One was converted into Solar Two, by adding a second ring of 108 larger 95 m (1,000 ft) heliostats around the existing Solar One, totaling 1926 heliostats with a total area of 82,750 m (891,000 ft). This gave Solar Two the ability to produce 10 megawatts. Solar Two used molten salt, a combination of 60% sodium nitrate and 40% potassium nitrate," - http://en.wikipedia.org/wiki/The_Solar_Project
30,000 square meters = 0.03 square kilometers
Could this technology be combined with desalinization, i.e. take salt water, pull the salt out to produce potable water, and use the salt to improve the plant's efficiency? Desalinization is a very energy-intensive process but I wonder if a lot of that could be offset using solar and redirecting the waste salt into the energy plant that powers the process in the first place.
Ahhh! Nuclear! Ahhh! It'll explode and kill us all and poison the planet for a bejillion years!
Just because you're paranoid doesn't mean there isn't an invisible demon about to eat your face
Most articles talking about power generation are talking about electrical power, so I'd guess that.
Is this thing really cost-effective? If it's mostly a proof of concept it doesn't have to be, of course. I did a back-of-the-envelope calculation:
So it's shiny and renewable (assuming the plant lasts a long time and doesn't break down into rusty mirrors encrusted with stray salt leaks in a year or two), and not *way* out of line compared to other power sources like coal plants, but it's not aggressively cheap either.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
5MW for $60M (euro).. really?
At 10c/kWh it can earn $500/hr. So it'll only take ~13.7 years to pay it off.. oh it's solar, right, well, with the seasons and everything I guess it's more like double that. Let's say ~27 years. How much is maintenance? Oh yeah, and the time value of money.
Another way of looking at it: it's $12B/GW + operations. Nuclear power plants take 5-10 years and cost $4-10 billion to build, and $4-6 billion for fuel and operation over their lifetime, so $8B/GW to $16B/GW. So the cheapest nuclear reactor beats this by at least 35% and the most expensive nuclear reactor probably beats it also.
But that fact that they've even made it into the right ballpark is impressive and perhaps once they scale it up to somewhere that is actually useful we'll have some idea how competitive it can be.
How we know is more important than what we know.
30,000 square meters = 3 hectares = 7.41 acres = 0.012 square miles.
File under 'M' for 'Manic ranting'
These numbers really don't add up. The article cites 2,100 tonnes of oil equivalent (which works out to about 3 MWth) and another cites 10 million kw-hrs of electricity per year (which works out to about 1.1 MWe). This would seem to imply the plant is about 3 MW thermal on average (and perhaps the extra 2 are only counted during the day). 30,000 square meters of reflectors perfectly aligned would generate about 30 MW thermal maximum at the best time of the year. Counting for night, seasons, etc., perhaps it could be diluted to 3 MWth.
Why are they citing all of the different numbers in CO2 equivalent, oil equivalent, and equivalent kw-hrs instead of actually saying what the actual electrical output is going to be on average?
So in other words, they are seeing if this design is worth it's salt?
My webcomic
5MW? Its sad that one HV pump on a process plant will use all of this. Miners should really have to purchase some of their power from renewable energy. It will stop them(us) from blatantly wasting power because its cheap.
The costs for this plant are very high of course because its a new thing.
This simple power point PDF reallyshows the numbers of the solar thermal salt plant in spain that is run as a research plant.
http://www.dlr.de/tt/Portaldata/41/Resources/dokumente/institut/thermischept/Solar_Thermal_Energy_Storage_Technologies_Hannover2008.pdf
They actually concluded that Salt is Not the only option. The problem with salt is rust, and so you have to use carbon coating on all the steel parts, which makes it expensive.
Simple using concrete was a very attractive option also.
And then that means that hemp concrete is also possible which is much cheaper again.
What's this Fahrenheit rubbish?
A 5MW windmill can be up and running for about 1.5M euros, but a pilot plant such as the one in TFA does not have the same goals as a commercial plant.
And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
Do you mean the one mentioned in the middle of this article
"Molten salts have been used in many industries as a high temperature heat transfer medium. The 'highest profile' use of molten salts in this regard is the Solar Power Tower near Dagget, California (excuse the pun). It uses a Sodium Nitrite/Nitrate mixture to absorb and store the sun's heat from the focus of many mirrors in the desert upon a central tower. The heat from the salt is then transfered via a heat exchanger to produce steam to drive a conventional steam turbine and generator to produce electricity from the sun for Southern California.3a"
"Last modified, 20 Nov 97"
--= Isn't it surprising how badly I spell ?
Try estimating what the basic maintenance costs are for 3 miles of piping that can handle molten salt.
Molten salt is rely, really *corrosive*. Either they're spending tons of money up front on miles of stainless steel, or even more every year replacing the pipes as they corrode away.
Either way it's hard to even break even-- 5MW of electricity is only about $2 million a year wholesale, far less than the interest cost on a $60M plant, and likely less than the cost to maintain 3 mmiles of molten salt piping and collectors.
You forgot that
a) nuclear power plants are the only industrial plants in the world which do not need to be insured to the full extent of possible damages they might cause. The insurance industry made politics cap the max at a mere 5 billion Dollar which may sound like a lot, but it's not. The population at large would shoulder those costs.
b) the countries in which the plants operate are charged with long-term storage. So the population at large shells out for that.
A prime example of privatizing earnings and socializing losses if there ever was one.
It's high time we got rid of fission (other than what we need for medical & research reasons). The claimed cost-efficiency _does not exist_. Period.
Having said that, it's been known for a long time that certain austenitic high-chrome alloys resist molten alkali nitrates very well. I would imagine that the designers of this plant have optimised the piping for the salt mixture in use, using the usual lifetime/costs tradeoffs in corrosion engineering. (The same tradeoffs that make it much cheaper, for instance, to make a boat out of steel with sacrificial anodes than out of stainless steel or aluminum)
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
"when it's produced in bursts, you will have to find a way to store it, which means a loss in efficiency."
Yes, however you're only looking at energy loss in one particular circumstance rather than looking at the overall efficientcy of the system in dollar terms.
Currently coal plants produce too much at night and not enough during the day. This means they waste fuel at nightly lows and have to be supplemented by "busrts" from gas turbines during daily peaks. Therefore (if it was possible**) there's much more value in producing energy that matches the peaks and troughs of consumption rather than trying to produce it at a constant rate capable of handling the peaks, especially if you have to pay for fuel.
The fact is that producing electricity at a constant rate capable of handling the peaks is not how electricity is generated on a commercial scale. All methods of generating electricity are intermittent. The idea that we currently have an efficient steady stream of "base load" power provided by constantly running coal plants is largely a myth created by the coal industry.
Coal plants are shut down for regular maintenance for ~45 days/year. Meaning one redundant coal plant needs to be built for (roughly) every seven coal plants in use. Plus to handle peaks you still need to build gas turbines that will sit idle for 20 or more hrs/day (or "inefficiently" pump water uphill). The advantage with wind, solar, etc, over fossil fuels is that; when it comes to handling the unavoidable peaks you can pump water uphill, (melt salt, whatever), during "bursts" and it will cost you some percentage of nothing in fuel costs.
Sure, windfarms also require maintenance but you can do it one windmill at a time, the whole farm very rarely needs to be shut down all at once.
** = Regardless of how you produce the electricity the most economically efficient answer to the inherent problems of peaks, troughs, bursts and breakdowns is a large well managed grid with built in generation/transmission redundancy and plenty of pump storage capacity.
And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
Someone needs to explain me how you can create 5 megawatts with only 30.000 square meters. That would make the plant produce 160 Watts/sq. meter on average.
Theoretical maximum efficiency for any kind of solar plant (on the equator) is less than 200 Watts per square meter (to give you an idea, in southern florida it drops below 150, and this is north of florida). That would make this plant over 120% efficient (at least).
Unless, of course, you know, they're lying and it's like 5 megawatts peak capacity at 12h noon at that optimal day in spring when the sun is directly overhead for its longest period, and only counting the total energy circulating in the plant, not what's actually coming out to the grid, which should be a bit under 2/3rd of that, or, say 3.8 megawatts. And 3, at best, during winter.
Isn't the idea to spread dependence away from one source ?
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
I guess you could have (redundant) electric heaters for that
Or change the salt mixture, maybe something that goes "sludge" instead of becoming solid
how long until
TLDR: Molten salt has zero benefit as a nighttime storage system. Ordinary boiling water is a better choice by a factor of >500.
I can't find good data on the heat capacity of the particular salt used in this system, but heat capacities for salts in general are around 1 J/kg-K.. If you're dealing with a temperature change of 700 K, that means each kg of salt can store around 700 J of heat. To store enough heat to power a typical American household overnight (1 kw x 12 hours), you'd need 61 tonnes of salt.
Now, most power plants use water as the working fluid. The latent heat of vaporization of water means that steam stores *at least* 330,000 J per kg of water in the phase change alone, plus additional specific heat if the steam is stored above the boiling point, which I'm too lazy to calculate.
That means that plain old ordinary water, already used in every thermodynamic power plant ever made, is at least 500 times better at storing heat than salt is.
Oh god dammit. Units failure, I'm off by a factor of 1000, and boiling water and high-temperature salt are actually about equal in terms of heat storage.
Mod parent down.
Then you must know that solar plants are often stated to have a rating which is at peak output. And by peak I mean maximum you'll ever see on the ideal day in the ideal weather. For that matter, so is any other power plant -- the nameplate output is peak, not average.
BZZZT. The solar constant is 1360 W/m2 (minus atmospheric effects). With this, the efficiency is a reasonable 12%.
I suspect the number you have quoted there is the TMI of solar cells. This plant depends on thermal conversion, not liberating electrons across a silicon band-gap. I do respect your effort, and so will not request that you turn in your geek card.
"I guess the moral of the story is, don't paint your airship with rocket fuel." -- Addison Bain
Excellent point. Except, your numbers are wrong. The peak insolation in Sicily is over 1 kW/M^2 at high noon on the equinox. So, that would mean they are horribly inefficient if 5 MW was their peak. At any rate, 250 Watts is the average for the entire world, for the entire year. Sicily isn't that far north...I believe its average insolation (again averaged over an entire year, not just during daylight) is around 180-220 Watts. (Italy is farther north than Florida, you're right, but latitude isn't everything. The average insolation in Florida is the same as the average insolation in the southern half of Oregon, maybe lower).
So, if they found a particularly sunny bit of Sicily, which I think is rarely cloudy as it is, an average insolation of 220 is probably believable. That puts it at 70% efficiency. That's kind of high, considering that they are doing solar energy to heat salt to heat water to turn steam turbine. But, it's not impossible. Plus, if their 5 megawatt figure is only the average when it's active, and it's inactive between midnight and dawn, that puts it more like 50% efficient, which is totally believable.
ASCII stupid question, get a stupid ANSI
A 5MW windmill can be up and running for about 1.5M euros
Do you happen to have a source on that? I know that at the moment it's like $1.3 per , but last I heard wind turbines were running $2/watt and up.
So I might believe 5M, but not less than a third of that.
60M for 5MW is 12($15.60) per watt, which is kinda, sorta, acceptable for a test plant. But I'd say costs would have to come down nearly an order of magnitude for this to be truly economical.
I'd also want to know if that 30k m^2 can actually RUN that plant at 5MW all day and night, on average. What sort of capacity factor are we looking at?
I don't read AC A human right