The economics of plutonium recycle vs. spent fuel storage

Frank von Hippel Program on Science and Global Security, Princeton University and

International Panel on Fissile Material Panel, New Diplomacy Initiative

Tokyo, 6 November 2015

Outline

•  The mistake that launched civilian plutonium separation

•  The high costs compared with spent fuel storage

•  The danger from dense-packed spent-fuel storage

•  The alternative: dry cask spent fuel storage and why it is safer.

Why do we have enough separated civilian (purple) plutonium for >30,000 Nagasaki weapons?���

(Source: Global Fissile Material Report 2015)

10,000 warheads,

(4-8 kg each)

0

500

1000

1500

2000

2500

1970 1980 1990 2000 2010 2020 2030 2040 2050

Glo

bal N

ucle

ar C

apac

ity (G

We)

1975

2015 High

2015 Low

Answer: Fears of uranium scarcity in the 1960s and 1970s led to proposals to develop “breeder” reactors.

Projected nuclear capacity (1975, IAEA)

Projected band for nuclear capacity (IAEA, 2015)

IAEA,  1975  

NEA-­‐IAEA,  2014  

Estimated Low-cost Uranium(40-year supply for LWRs)

High

Low

5

U-235 (0.7% of natural uranium) chain-reacts and provides most of the energy in current-generation (mostly water-cooled) reactors.

U-238 (99.3%) doesn’t chain-react but turns into chain-reacting Pu-239 after it absorbs a neutron.

The 3 grams of U-238 in an average ton of crustal rock has a releasable energy of 10 tons of coal.

The proposed solution: Move from U-235 to U-238 as a fuel and you can “burn the rocks”

Fresh ���LWR fuel

95.6 % U-238

 

4.4% U-235

92.6% U-238 & U-236

 

Spent Fuel

0.8% U-2351.2% plutonium5.4% fission products &

other radioisotopes

Plutonium in light water reactor spent fuel was to be separated as startup plutonium for breeder reactors

6

1974. India used first plutonium separated for its breeder program for a “peaceful nuclear explosion”

Crater from India’s 1974 underground nuclear test.

1977. President Carter: “Do we need breeders?” Answer: “Breeders will not be competitive.”

Cost of breeder electricity increased by unreliablity Average capacity factor for water-cooled reactors ~ 80%

for sodium-cooled “demonstration” reactors ~25%.

Country Demonstration Reactor Power (Mwe) Lifetime Capacity Factor* France Superphénix (1985-98) 1200 8% (sodium leaks) Germany SNR-300 (1991) 300 Did not receive safety license Japan Monju (1994-) 246 1% (sodium fire) UK PFR (1975-94) 500 19% (sodium leaks) USA Clinch River 300 Cancelled USSR BN-600 (1980-) 560 74% (but 15 sodium fires)

*IAEA Power Reactor Information System  

MOX Fuel

France decided to use its separated plutonium in water-cooled reactor fuel. Saves ~12% of natural uranium.

Spent LEU fuel storage

Reprocessing Plant

Spent MOX fuel storage

MOX Fuel fabrication plant

Plutonium & uranium

Water-cooled reactors

Radioactive waste

Spent MOX Fuel

Low-enriched uranium Fuel

Spent LEU Fuel

Separated Plutonium

US: once-throughFrance, Japan: plutonium recycle

Japan decided to do the same.

U.S.

France and Japan

0

20

40

60

80

100

120

1990 1995 2000 2005 2010 2015 2020 2025

Sepa

rate

d pl

uton

ium

(m

etric

tons

) In Japan if RRP operates and MOX delayed In Europe if MOX delayed Recycled In Japan In Europe

Japan’s MOX program has failed so far. Would have to become very successful to prevent explosive growth of Japan’s stockpile if

Rokkasho Reprocessing Plant operates as planned.

estimates of LWR unit reprocessing costs in constant dollars increased substantially from 1975 through 1983, as shown inFigure 6-2. Also shown in Figure 6-2 are the current estimates of the unit costs for reprocessing plants constructed in theUnited States, derived from estimated costs for contemporary plants in the United Kingdom, France, and Japan. Financialparameters were applied for a private venture in the United States, assuming optimistically that the financing would becharacteristic of a low-risk project in the chemical industry. The unit costs are expressed for an inflation-free economy. Theunit costs estimated for these three sources fall on extensions of the band of reprocessing costs shown by Wolfe and Judson,but at a level several-fold higher. It is clear that what may be financially valid for a government-owned European plantfinanced with customer prepayments for reprocessing services, and with relatively low annual charges on capital investment, isnot necessarily applicable to the same plant constructed by private industry in the United States.

Also shown in Figure 6-2 are recent estimates by the ALMR project of unit costs for privately reprocessing LWR fuelfinanced by U.S. construction (Taylor et al., 1992; Chang, 1993). Each plant has a throughput of 2,700 Mg/yr, with high-yieldrecovery of all actinides and volatile fission products. The estimated unit costs are about $500/kg for aqueous reprocessing and$350/kg for pyrochemical reprocessing. The latter is about six-to eightfold below the estimated unit cost of a 800- to 900-Mg/yr U.S. plant, based on contemporary plant costs in the United Kingdom, France, and Japan.

FIGURE 6-2 Current estimates of the unit costs for reprocessing plants constructed in the United States.

Also shown in Figure 6-2 are the unit costs estimated in 1991 by S.M. Stoller Co. (Gingold et al., 1991) for aqueousreprocessing and pyrochemical reprocessing, with U.S. private financing. The Stoller estimates, in a study financed by theElectric Power Research Institute (EPRI), are close to those made by GE, but far below the estimates derived herein from thenew United Kingdom, French, and Japanese plants.

Comparisons with published reprocessing prices and with other estimates of reprocessing costs are given in Appendix J.

SummaryCosts of contemporary aqueous reprocessing plants in the United Kingdom, France, and Japan are important benchmarks

to compare with U.S. estimates of reprocessing. For the purpose of this report, we adopt the OECD/NEA estimates of thecapital and operating costs of a plant with 900-Mg/yr throughput. These are based largely on the U.K.'s THORP data, withinput from France's COGEMA. We have translated these costs to U.S. construction as described in Appendix J: 10-28 and J:37-60. We estimate interest during construction and calculate the unit reprocessing costs for a similar U.S. reprocessing plantfor three forms of U.S. financing: government, $810/kg; utility, $1,330/kg; and unregulated private industry, $2,110/kg. Eachof these costs is so high that there would be no financial incentive for operating a transmutation system that would requirereprocessing spent fuel from LWRs, unless it were subsidized by the government for possible benefits to waste disposal. Toobtain the high recoveries to recycle all the TRUs (not only Pu), as proposed by DOE laboratories and contractors, the cost ofaqueous reprocessing would be even greater. Even higher costs for a U.S. reprocessing plant would occur if the delaysexperienced by previous reprocessing ventures were again encountered.

We have compared these unit costs derived from contemporary plant data with costs of aqueous reprocessing projected instudies by DOE laboratories and contractors for the purpose of transmutation economics. The latter are so much lower thanthose estimated in the present study that there is good reason to question the validity of all the recent U.S. estimates for the costof reprocessing LWR spent fuel. Given that those estimates for aqueous plants are so far below the costs inferred from theEuropean and Japanese benchmarks, it is questionable that reliable estimates could now be made of the pyrochemical processfor LWR spent fuel, which is in a relatively early stage of development.

ANALYSIS OF THE ISSUES 117

Abo

ut th

is P

DF

file:

Thi

s ne

w d

igita

l rep

rese

ntat

ion

of th

e or

igin

al w

ork

has

been

reco

mpo

sed

from

XM

L fil

es c

reat

ed fr

om th

e or

igin

al p

aper

boo

k, n

ot fr

om th

e or

igin

al ty

pese

tting

file

s. P

age

brea

ks a

re tr

ue to

the

orig

inal

; lin

e le

ngth

s, w

ord

brea

ks, h

eadi

ng s

tyle

s, a

nd o

ther

type

setti

ng-s

peci

fic fo

rmat

ting,

how

ever

, can

not b

e re

tain

ed, a

nd s

ome

typo

grap

hic

erro

rs m

ay h

ave

been

acc

iden

tally

inse

rted.

Ple

ase

use

the

prin

t ver

sion

of t

his

publ

icat

ion

as th

e au

thor

itativ

e ve

rsio

n fo

r attr

ibut

ion.

Copyright © National Academy of Sciences. All rights reserved.

Nuclear Wastes: Technologies for Separations and Transmutationhttp://www.nap.edu/catalog/4912.html

Costs of French and UK reprocessing plants as of 1992.

Estimates on the basis of which France, Japan and UK built their reprocessing plants

Estimates by U.S. reprocessing advocates in early 1990s.

Reprocessing costs were grossly underestimated by advocates

Nuclear Wastes: Technologies for Separations and Transmutation (National Academy Press, 1996) p. 117.

X(5-­‐10)  

}

Conclusions of Economic Reviews

France (2000). Plutonium and uranium recycle costs five times more than the savings in LEU fuel costs.*

Today, the cost may be ~10 times because of loss of foreign customers. Reprocessing costs have become a major issue between Électricité de

France (EDF) and AREVA. EDF refused to renew UK reprocessing contract.

Japan (2011). Plutonium and uranium recycle cost ten times more than the savings in LEU fuel costs.**

Yet both countries have chosen to continue with reprocessing because change is judged too disruptive.

*Report to the Prime Minister [of France]: Economic Forecast Study of the Nuclear Power Option, 2000. **JAEC, Technical Subcommittee on Nuclear Power, Nuclear Fuel Cycle, etc. Estimation of Nuclear Fuel Cycle Cost, 2011, slide 28.

Operating Rokkasho will cost ~ ¥200 billion/yr. (¥250,000/kg) ~7x cost of buying dry-cask spent fuel storage

¥200 billion/yr.

31 countries with nuclear power plants: 6 reprocess

•  France (62 tons of separated plutonium) and Japan (48 tons) for use in water-cooled reactors

•  China (0.025 tons), India (3.6 tons) and Russia (85 tons) for breeder reactor prototypes.

•  UK (103 tons). Disposal path not yet decided.

250 tons total is enough for 30,000 Nagasaki weapons

Most countries manage older spent fuel with safe onsite dry cask storage. (Japan has dry cask storage at Fukushima-Daiichi and Tokai.)

Tokai

U.S. Connecticut Yankee (old picture)

Lingen NPP, Germany

At Fukushima Daiichi after the tsunami

15

Tokai

Dry-cask storage also alternative to dense-packed storage pools. A spent-fuel fire in Fukushima Daiichi Pool #4 could have forced

evacuation of Tokyo (JAEC chair Kondo to Prime Minister Kan)

Thermal analysis of pool heatupand boil off

� Models of spent fuel pools developed to predict pool boil off time and to understand hydrogen production

� Used to perform analysis of pool leakage scenarios� Calculations based on several codes and models to provide range in

turn-around time and fidelity

0 9

8642

1 3 5 7

0 9

8642

1 3 5 7

0 9

8642

1 3 5 7CR

CR

CR CR CR

CR CRCRCR

CR30

0

2

4

6

8

9

1

3

5

7

CR

CRCR

CR

CR

CR

CR

CR

C

E F P M W F P M

UNIT 4 SFP HEAT GENERATION RATE DISTRIBUTION POOL LEVEL FOR VARIOUS SCENARIOS FOR UNIT 4

16 0.19 kW

24 0.16 kW

14 0.20 kW

10 0.22 kW

12 0.21 kW

9 0.23 kW

5 0.30 kW

8 0.24 kW

2 0.55 kW

4 0.40 kW

1 1.12 kW

IF 3.60 kW

Fuel discharged end Nov 2010

Fresh fuel BWR fuel assembly contains ~ 170 kg U

3/25

1. SF at Fukushima NPS(1/2)-(1) In the reactor buildings-

CRIEPI

Unit  1 2 3 4 5 6

FA in Core (No.)400 548 548

(MOX 32)0 548 764

SF in Pool (No.) 292 587 514 1331 946 876

FF in Pool (No.) 100 28 52 204 48 64

Decay Heat in Pool(MW)

March 11,2010

0.18 0.62 0.54 2.26 1.00 0.87

June 112010

0.16 0.52 0.46 1.58 0.76 0.73

Ref.1

Condition of Unit 4‘s spent fuel pool 

U.S. Nuclear Regulatory Commission estimates release from fire in a high-density pool 100 times worse than Fukushima

(average consequences for the Peach Bottom site in Pennsylvania)

Sources. http://pbadupws.nrc.gov/docs/ML1328/ML13282A564.pdf;http://pbadupws.nrc.gov/docs/ML1328/ML13282A563.pdf , http://pubs.rsc.org/en/Content/ArticleLanding/2013/EE/c2ee24183h#!divAbstract

Explanation: Enough hydrogen from zirconium-steam reaction in high-density pool to destroy reactor building.

High-density pool

Low-density pool

Fukushima-Daiichi release

Release (PBq) 925 4 6-20 Cancer deaths 43,100 1,100 ~1000 Area evacuated 46,600 221 ~650 Population displaced

10.9 million

72,000

~100,000

Some of the Commissioners on Japan’s Nuclear Regulation Authority (NRA) understand the danger

On 19 September 2012, in his first press conference, NRA Chairman, Shunichi Tanaka urged “Spent fuel not requiring active cooling should be put into dry casks … for five years or so cooling by water is necessary…I would like to ask utilities to go along those lines…”

On 29 October 2014, Chairman Tanaka and Commissioner Fuketa urged the president of Kyushu Electric Power Company, to introduce dry-cask storage.

If dense-packed pools are dangerous, why does the NRA not order nuclear power plant operators to follow Chairman Tanaka’s advice?

Summary

1.  Using a nuclear-weapon material as a commercial nuclear fuel is a very bad idea.

2.  Sodium-cooled reactors are much more costly and much less reliable than water-cooled reactors. After 50 years, no country has commercialized them.

3.  Plutonium recycle in light water reactors is about 10 times more costly than storage.

4.  Reprocessing persists in 4 weapon states and Japan. South Korea is pressing for the right to reprocess.

5.  Spent fuel pools are safer if fuel that has cooled more than 5 years is downloaded to casks.