Pages

August 05, 2008

What would it take ? New Nuclear electricity at less than two cents per kwh

New nuclear plants constructed in the USA can provide electricity at about 5-9 cents/kwh

Because of the large capital costs for nuclear power, and the relatively long construction period before revenue is returned, servicing the capital costs of a nuclear power plant is the most important factor determining the economic competitiveness of nuclear energy. The investment can contribute about 70% of costs of electricity, according to one 2005 OECD/NEA study (which assumed a 10% discount rate). The discount rate chosen to cost a nuclear power plant's capital over its lifetime is arguably the most sensitive parameter to overall costs.

Industry consensus is that a 5% discount rate is appropriate for plants operating in a regulated utility environment where revenues are guaranteed by captive markets, and 10% discount rate is appropriate for a competitive deregulated or merchant plant environment, however the independent MIT study which used a more sophisticated finance model distinguishing equity and debt capital had a higher 11.5% average discount rate.

Construction delays can add significantly to the cost of a plant. Because a power plant does not yield profits during construction, longer construction times translate directly into higher interest charges on borrowed construction funds. Modern nuclear power plants are planned for construction in four years or less (42 months for CANDU ACR-1000, 60 months from order to operation for an AP1000, 48 months from first concrete to operation for an EPR and 45 months for an ESBWR) as opposed to over a decade for some previous plants.


The first AP1000 has some delays but we will see how much that drops with the later ones which will be built, in particular China is ordering 40-100 AP1000s. China has put together what amounts to a production line for further build of AP1000 nuclear reactors. A 71000 square meter (710,000 square feet) factory specifically designed for nuclear power plant component modules. Shandong Nuclear Power Construction Group built the facility, which has the capacity to support the construction of two AP1000s each year, in just 11 months. [They will need four more factories to build ten AP1000s each year. Nine more to build twenty AP1000s per year.]


Shandong said the new 71,000 square metre factory includes a cutting workshop, a pipeline workshop, a paint shop and a workshop for containment vessels (the steel liners that lie within the overall reinforced concrete reactor containment).

Large components for the Haiyang units have already been contracted: Doosan Heavy Industries of Korea is making the reactor pressure vessels and steam generators, while Mitsubishi Heavy Industries of Japan and Harbin Boiler Works of China will supply the steam turbines. For Westinghouse's other pair of AP1000s at Sanmen the steam generators and reactor pressure vessels will be made in China by either Harbin, First Heavy Machinery Works or Shanghai Electric.

First concrete at Haiyang - the official start of construction - is expected in September 2009, with commissioning of the first unit about 36 months later.


Longer construction times add higher interest charges. (About 20-40% to the cost over 4-5 years.) Bringing construction time down to 2 years for China's HTR would help lower costs.The China HTR is starting at 6 cents per kwh but they need to get volume scaling to lower costs equal to large plant scaling.

$ per Megawatt-hour (10 $/Mw-hr = 1 cent/kw-hr):

                                           Old Nuclear      Coal  New Nuclear est
1 Fuel 5.0 11.0 5.0
2 Operating & Maintenance - Labor & Materials 6.0 5.0 8.0
3 Pensions, Insurance, Taxes 1.0 1.0 1.0
4 Regulatory Fees 1.0 0.1 1.0
5 Property Taxes 2.0 2.0 2.0
6 Capital 9.0 9.0 39.0
7 Decommissioning & DOE waste costs 5.0 0.0 5.0
8 Administrative / overheads 1.0 1.0 1.0
Total 30.0 29.1 60.0


Payoff capital costs (after say 30-40 years) and the it goes down to 2 cents per kwh.
for the remaining 20-30+ years on old nuclear. New nuclear has inflated capital costs because of currently higher commodity prices and other inflated costs.

Waste costs can be reduced if we have new deep burn capable reactors to burn it.

Nuclear fuel costs

Uranium: 8.9 kg U3O8 x $53 472
Conversion: 7.5 kg U x $12 90
Enrichment: 7.3 SWU x $135 985
Fuel fabrication: per kg 240
Total, approx: US$ 1787

At 45,000 MWd/t burn-up this gives 360,000 kWh electrical per kg, hence fuel cost: 0.50 c/kWh.


If MW days per ton is double then half volume of fuel is needed. This can be achieved with new annular fuel and slightly higher enrichment levels. 8% instead of 5%.

Enrichment from 2012 onwards will be 3-20 times cheaper with laser enrichment.

Half the fuel at three times cheaper enrichment is

Uranium $250
Conversion $45
Enrichment $165
Fuel fab $240 (halved for less fuel but doubled for more complex
fabrication)
Total approx $750

0.2 c/kwh instead of 0.5 cents for the fuel

The world nuclear link also factors in building a lot more units to get learning curve reductions in costs and the effect of 60 year life instead of 40 year (the 50% longer life reduces kwh costs by 1 penny when combined with 5 year instead of 7
year construction)

More units allow first of kind costs to be amortized.
$1200/kw capital costs, 5 year construction, 60 year life is 3.4 cents/kwh

$1000/kw capital costs, 2 year construction, 80 year life, reduced fuel costs 2 cents/kwh for new construction, plus less waste handling with deep burn.

Operation and maintenance can be reduced with more automation and design efficiency.

                                   Volume Improved Nuclear estimate
1 Fuel 2.0 (laser enrichment, 50% uprate)
2 Operating & Maint - Labor&Materials 2.0 (automation, improved designs)
3 Pensions, Insurance, Taxes 1.0
4 Regulatory Fees 1.0
5 Property Taxes 2.0
6 Capital 7.0 (loan guarantees,factories, high volume, longer 80 year operation so more amortization)
7 Decommissioning & DOE waste costs 1.0 (Deep burn the waste, more time until decommissioning with longer life,more time for interest to build for decommissioning fund)
8 Administrative / overheads 1.0
Total 17.0 equal to 1.7 cents per kwh

China could get labor costs down with lower salaries and capital costs could be reduced because of lower construction costs. China could get to 1.3-1.5 cents per kwh.

UPDATE FURTHER READING [Added after the above analysis was made]
This study shows an increased cost for current new power build.
There is a new 12 page whitepaper on current nuclear, coal and natural gas

In the base case, new nuclear capacity produced a levelized cost of $83.40/megawatt-hour. Supercritical coal was at $86.50/MWh; supercritical coal with CCS at $141.90/MWh; IGCC at $92.20/MWh; IGCC with CCS at $124.50.MWh; gas combined-cycle (CC) at $76.00/MWh; and gas CC with CCS at 103.10/MWh. Although it had the highest capital cost, the new nuclear capacity produced the lowest-cost electricity, except for gas-fired CC capacity without CCS.

Natural gas had $76MWh assuming $6.26/mmBtu.

CBO also found that new nuclear capacity could be competitive at even lower carbon dioxide charges if the price of natural gas rose above the price assumed in the reference scenario ($6.26/mmBtu) or if the construction cost reductions predicted by the reactor designers were accurate. CBO found that in a high gas price environment of $12/mmBtu (near present-day prices), natural gas would no longer be competitive with new nuclear.

Nuclear Energy Institute modeling shows that a merchant nuclear plant with an 80 percent debt/20 percent equity capital structure, supported by a federal loan guarantee, will produce electricity in the range of $64/MWh to $76/MWh. (The range reflects EPC costs from $3,500/kWe to $4,500/kWe) A high-cost ($4,500/kWe EPC cost) nuclear plant producing electricity at $76/MWh is competitive with a gas-fired combined-cycle plant burning $6-8/mmBtu gas or an SCPC plant.

Similar results are found using the same capital cost range for a regulated plant. Assuming a 50 percent debt/50 percent equity capital structure typical of a regulated electric company, and assuming the company is permitted to recover the cost of capital during construction (CWIP), NEI’s financial model shows the levelized cost of electricity from the plant ranges from $74/MWh to $88/MWh – competitive with a gas-fired combined cycle plant burning gas at $8-10/mmBtu or an IGCC plant (without carbon capture and sequestration).

5 comments:

Brock said...

The nuclear industry wants security from government-induced construction delays. Everyone wants to get off oil and stop burning coal. Solution?

The government pays the interest on construction loans until electricity starts flowing. From there it's a purely commercial enterprise.

It's a bit of a subsidy to the nuclear industry, for sure, but a good idea overall, I think. Unless we get some really good news from a Polywell or Blacklight Power type company, nuclear is the only long term solution. Subsidizing a quick and painless transition to sustainable energy is no bad thing in my book.

10ksnooker said...

What I would like to know is why can't the U.S. build a production line for new nuclear reactors? Wouldn't this solve a whole lot of problems with delays? While at the same time reducing costs? Make the U.S. into France, which gets about 80% of their electricity from nuclear power.

An assembly line, as your article states is one of the best ways to get costs down. Wouldn't it also produce safer units, since it is far easier to maintain the quality of work on an assembly line than in the field where everything is more or less custom built -- By varying skill level teams I might add. Why not go safe and chaep, like China? Just do it ...

If we did this, we could then use the coal for coal to liquids manufacturing plants which would produce transport fuels.

Noah said...

10ksnooker asked: Why can't the U.S. build a production line for new nuclear reactors?

From what I understand, this was tried by Combustion-Engineering in the early seventies. It was a great idea, but suffered from bad timing. First, the Nuclear Industry went more or less belly up after Three Mile Island, and second, C-E was hit by crushing asbestos lawsuits. The company went more or less bankrupt, with other companies buying their assets.

bw said...

AP1000 module factories will be built in the USA as well

18+ module factories around the world planned. 4 more at least in China

Cyril R said...

There's quite a bit of variance in nuclear levelized cost estimates for the latest projects in the US. Most are in the 9-15 cents/kWh range. Learning curve analysis suggests 2020-2030 costs to be 8-30 cents in the US.

This increase in uncertainty is a real problem for investors, who want to know what they're up against. IMHO this is one of the key reasons why there's so little private investment in new nuclear projects.