Thursday, June 11, 2009

Greenpeace and CSP Costs to 2050

Greenpeace recently spoke of a glowing future for Concentrated Solar Power:
With advanced industry development and high levels of energy efficiency, concentrated solar power could meet up to 7 percent of the world's power needs by 2030 and fully one quarter by 2050.
The 3rd joint report from Greenpeace International, the European Solar Thermal Electricity Association (ESTELA) and IEA SolarPACES estimated that investment in CSP technology would increase to to $29 billion a year by 2015 and $243 billion a year by 2050. An investment increase at that rate would lead to installed CSP plant capacity reaching 1,500 GW by 2050.

The report states that "during the 1990's infrastructure was around €158-186 billion each year . . ." The report states that "the cost of CSP electricity is coming down and many developers say it will soon be cost-competitive with report estimates that with an all out investment program in CPS costs would drop 3,060 € ($4290) per kW by 2015 and 2,280 € ($3200) per kW by 2050.

The report, however, offers a confusing account of CSP costs that obscures important information about the CPS costs. The most conspicuous evidence for dropping CSP prices was based on a series of 7 small CSP plants built in California between 1986 and 1992. But rather than offer us real world information about construction costs the report offers us the information that operating costs are dropping. The report also offers an extremely obscure account of the cost of CSP with energy storage, noting for example the cost of Molten Salt heat storage, but failing to note the added cost of expanded heat gathering capacity required to provide electricity over a longer period of time. Thus for a 16 hour a day CSP with heat storage, the solar gathering capacity would need to be twice that of a no storage plant with the same rated electrical output. The report stated that molten salt heat storage cost costs $30 per kWh. So we not have all of the information we need to calculate the cost of 16 hour a day dispatchable CSP. Assume 2X the 2015 CSP cost of $4300 Per kW or $8600. Add to that $30 per kWh X 16 = 480. So we end up with a 2015 cost of $9100 for 16 hour a day concentrated solar power that is dispatchable and capable of generating electricity 16 hours a day.

In fact this cost seems to correspond to the cost of the Starwood 1 CSP facility.

As usual advocates of CSP are keeping quiet about the cost of Starwood 1, a CSP facility with molten salt storage. But it appears to be $2.7 billion for a 290 MW facility, or about $9300 a kW. This cost tracks closely with my estimates of the cost of CSP with storage, and while it gives me a certain satisfaction to have accurately predicted CSP costs, the magnitude of those costs give me no satisfaction. The bad news, and it is very bad news, is that CSP will be significantly more expensive than conventional nuclear power,

As I noted the Greenpeace report on solar power claimed without evidence that the cost of solar power was dropping, and claimed that it would drop even more by 2050. Let us examine the cost of CSP if the price of CSP does not drop between 2015 and 2050, and then look at costs if CSP drops as projected by Greenpeace. Greenpeace estimates that CSP could be responsible for as much as 25% of global energy output by 2050 or 1500 GWs of generating capacity. Given the no price drop assumption the cost of the 25% assumption would be around $13.5 trillion. If the average cost of the CSP generated power dropped by 1/3 as Greenpeace assumed, the total cost for the 25% CSP system could be as low as $9 trillion. Remember that $9 trillion is a low cost, based on the most unlikely of assumptions. There is by the way no assurance that the cost of CSP will not be higher than the Starwood 1 costs.

Now lets look at some nuclear costs. Indian LMFBRs are expected to cost $1200 per kW in serial production. 1500 worth of indian LMFBRs would run $1.8 trillion. Chinese LWRs have an estimated cost of $1750 per kW or $2.624 trillion for 1500 GWs. American factory built modular LFTRs could run as low as $1200 per kW or $1.8 trillion for 1500 GWs. The maximum estimated costs for Westinghouse AP-1000s in the United States is $7000 per kW or 10.5 trillion for 1500 GWs generating capacity which would be the lowest end of the CSP price range to 2050.

The conclusion is that AP-1000s may have a cost advantage over CSP facilities until 2050, and will remain at least cost competitive. Generation IV nuclear technology could cost as little as 20% of the cost of CSP facilities at least until 2050. Chinese reactors are likely to cost more than Indian or American Generation 4 reactors, but will still be inexpensive compared to European or American CSP.


donb said...

Greenpeace said:
With advanced industry development and high levels of energy efficiency, concentrated solar power could meet up to 7 percent of the world's power needs by 2030 and fully one quarter by 2050.

Taking what Greenpeace said (which I think is optimistic), we still need to get the other 3/4 of our power from somewhere. Hydro, geothermal and wind will be part of the mix, but I doubt they will be able to fill out completely that 3/4. So that leaves fossil fuels and nuclear as big contributors. Fossil fuels are dirty and increasingly expensive. So nuclear is what we need. Thus full speed ahead in developing and fielding advanced reactors such as the LFTR (in a supportive regulatory environment).

With the deployment of advanced reactors, we will know their cost. My bet is that they will show concentrated solar power to be so much foolishness.

soylent said...

The service economy is vacuous crap and it's reaching the end of line. The US cannot keep borrowing to maintain living standards, it's got to actually start making more stuff than it consumes again so it can pay off all that debt.

Making stuff means using energy and raw materials. Green peace no doubt presume the US is going to keep up the phony service economy for another 40 years. They're probably also assuming energy efficiency above and beyond what will happen in a free market; but that's going to make the US hopelessly uncompetitive(in a free market only the efficiency improvements that make financial sense will happen).

The only way to make that happen is for every country to agree to enforce the same efficiency rules so that no one gains an advantage. That may happen with carbon emissions, and it probably should, but it's not going to happen with efficiency because is a trade off and not an inherent good(it's not clear to me why wasting insulation is better than wasting uranium, say).


Blog Archive

Some neat videos

Nuclear Advocacy Webring
Ring Owner: Nuclear is Our Future Site: Nuclear is Our Future
Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet
Get Your Free Web Ring
Dr. Joe Bonometti speaking on thorium/LFTR technology at Georgia Tech David LeBlanc on LFTR/MSR technology Robert Hargraves on AIM High