ML20040C805

Comments Opposing Exemption of Facility from Std Licensing Procedures

ML20040C805
Person / Time
Site:	Clinch River
Issue date:	01/22/1982
From:	Hippel F PRINCETON UNIV., PRINCETON, NJ
To:	NRC OFFICE OF THE SECRETARY (SECY)
References
ISSUANCES-E, NUDOCS 8201290257
Download: ML20040C805 (38)

Text

=

Princeton [Tniversity scNooL o, NcINaz.INo/AFPLIED SCIENCE ENTER FOR ENERGY AND ENVIRONMENTAL STUDIES g:!.

T2 JD122 Pa:C0 M

h NOINEERING QUADRANGLE gg ta roN, Naw Jamsar o8544 N gggy C

January 13, 1982

' "~

6 4

g W.

h7 Commissione u.

cJ alladino, Chairman G@['- ~

• ~ ~ ~ ~ -

Nuclear Regulato ommission 1777 H Street, N.W.

Washington, D.C.

20555

Dear Dr. Palladino,

This letter is in response to the NRC's invitation of December 24, 1981 for comments on the request by the Department of Energy that it be allowed to proceed with site preparation for the Clinch River Breeder Reactor plant without f irst satisfying the usual NRC construction permit procedures.

I would urge that tha NRC deny this request because the nation is just now in the afdst of a reconsideration of the necessity of this plant. As I will show below, the only logical conclusion which can be drawn from such a recon-sideration is that the plant is unnecessary, that its construction will be wasteful of our national reso arces, and would, in fact, undermine U.S. nonprolif eration obj ectives. Under these circumstances the attempt by the Department of Energy to try to rush the project past some point of no ret 2rn before it is possible to complete the painful process of cancelling it can hardly be seen as in the national interest.

The LVFBR Demonstration Program is No longer Necessary If you refer to the Proposed Final 1.MFBR Program Environmental Statement which was issued by the AEC in December 1974, you will find an argument there for the breeder that went as follows:

By the year 2000 U.S. nuclear generating capacity would be approxi-e mately 1200 Gw(e) and new capacity would be coming on line at a rate of approximately 100 Gw(e) per year;l If all nuclear capacity were light water reactors, the associated e

lifetime U308 commitments (30 year reactor lif etime) would be 5.5 million tons for 1200 Gw(e) in the year 2000 and 12.7 million tons for 3300 Gw(e) in the year 2020;2 only between 2.6 and 6 million tons of U 03 8 could be recovered from e

U.S. resources before it would be necegsary to turn to very costly low grade Chattanooga shale resources; and 503 9

s

//

8201290257 820122 DR ADOCK 05000537 PDR

The capital cost of breeder reactors and the associated plutonium fuel e

cycle facilities would be so low that LMFBRs and their symbiotic plutonium -

burning LWRs would generate electricity at a cost 20 percent lower than LWRs operating on a once-through fuel cycle even at a uranium price of

$27 per pound U 08 (equivalent to about $50 per pound in January 1982 3

dollars.)4 Contrast this situation with that projected by the Department of Energy today:

e In its most recent (1980) Annual Report to Congress the Department of Energy projects U.S. nuclear capacity in the year 2020 as between 4 and 380 Gw(e).5 The middle case, 290 Gw(e), is less than one tenth that used by the AEC to justify the breeder in 1974; e Assuming that this capacity is all LWRs, the same report estimates that the associated cumulative consumption plus 30 year lifetime commitments of uranium as of 2020 will be 2.3 million tonc U 038 e The DOE's estimate of U.S. U 03 8 available at a forward cost af less -

than $100 per pound has risen to 4.9 million cons (3.5 million tons at $50 per pound).

[The corresponding minimum estimates.(95 percent -

confidence) are only 20 percent lower.7]

e The DOE estimated in 1979 that, because of the high cost of the LMFBR and its associated fuel cycle, the cost of U % would have to go up 3

to $115-205 per pound before the LMFBR could compete with even a slightly improved (15 percent reduced uranium requirements.per kWh) LWR.8 Other DOE calculations indicate that additional cost-effective uranium

-fficiency improvements 9 in combination with enrichnent tails stripping ucin; advanced isotope separation systems would raise the breakeven range 1

for a mature LMFBR industry to $150-250 per pound U %

This is two 3

11 to three times the $78 per pound U 03 8 which DOE projects for 2020 6

(and still only the equivalent of oil costing $7.5_- 12.5 per barre 1 ) __ _

According te *.he DOE's own analyses, therefore, the LMFBR will not be economically cr aps.citive till far beyond 2020. This suggests that, instead of treating the construction of the CRBR as a critical national priority, the DOE might be looking for more pressing aspects of the nr. tion's energy problems to spotlight. Satayana's statement:

Fanaticism consists in redoubling your efforts when you have forgotten your aim.

never seemed more appropriate.

e

Congressional Support for the LMFBR is Weakening As the effects of the AEC's selling job wear off, Congressional support for the LMFBR program is steadily weakening -- following a pattern reminiscent of what occurred in the case of the U.S. supersonic transport program.12 Continuation of funding for the LRBR was approved by the Senate in November by only 2 votes.13 and the most recent vote by the responsible Committee in the House of Representatives was in fact in opposition to continued funding.14 The NRC cannot, of course, substitute its judgment for that of Congress on this matter. On the other hand, there is no reason why the NRC shold waive its own rules in order to speed a proj ect of ro detectable merit at a time when Congressional support for the project is obviously weakening.

The Breeder Program Complicates Our Nonproliferation Problems Finally, I would like to remind you that the CRBR became controversial long before the AEC's projected " uranium crisis" faded away. Many of us became concerned that the U.S., by promoting the plutonium fuel cycle, was also promoting the spread of nuclear weapons. Unfortunately, the reasons for our concern have not faded.

Indeed it would appear that the NRC has reason to share that concern.

According to the NRC's letter of November 27, 1981 to Senator Simpson; The NRC is concerned that the IAEA safeguards system will not detect a diversion in at least some types of facilities.

"Some types" of facilities would presumably include the reprocessing and plutonium fuel fabrication plants which would be required by the LMFBR.

The proliferation of nuclear weapons may have critical implications for the public interest well within the DOE's planning horizon of 2020.

I assume that, when it is relevant and you have the latitude, you will factor this concern into your decision-making process. You obviously have the occasion and opportunity to do so in this case.

I enclose a recent article, a piece of Congressional testimony, and same comments on the DOE's draf t supplement to the LMFBR Program Environmental Impact Statement. They contain additional detail on the way in which the breeder was originally sold and what that case looks like now.

Please feel free to contact me if you have any questions on this material.

Sincerely yours,

,c:n.L - i-4)"

J' n-Frank von Hipp'el FvH/zk

Enclosures:

~

~ -~ ' - - - - - - - - - -

1) Letter comment on the Draf t Supplementary Environmental Impact Statement on the Liquid Metal Fast Breeder Reactor Program (DOE /EIS-0085-D).

~

2) " Uranium, Electricity, and Economics." Invited Statement to the Subcoma.ittee on Energy Conservation and Power of the House Committee on Energy and Commerce.

October 5, 1981.

3) "Should Breeder Reactors Be Built in the United States? No!" Public Power, May-June 1981, pp.19, 21 and 24.

e i

5_

Ref erenc es

1) US AEC, Proposed Final Environmental Statement, Liquid Metal Fast Breeder Reactor Program (WASH-1535, December 1974), p.11.2 - 113._

2)

Ibid., p. 11.1 - 32.

3)

Ibid., p. 11. 2 - 10.

4)

Ibid., pp. 11.2 - 4, 11.2 - 10 and 11.2 - 30.

5)

US DOE, EIA Annual Report to Congress, 1980:

Vol. 3, Forecasts,

[ DOE /EIA-0173(80)/3 ], p.158.

6)

Ibid., p.177 using 170 million Btu primary energy released in LWRs per pound of U308 mined.

7) US DOE, An Assessment Report on Uranium La the United States of America (GJ0-lli (80), 1980), p. 1.

8) US DOE, Nuclear Proliferation and Civilian Nuclear Power: Report of the Nonproliferation Alternative Systems Assessment Program (Draf t DOE /NE-001,1979),

Figure 11.

D.F. Newman,e_t g., (PNL) " Assessment of Nonbackfittable Concepts for Improving 9) t Uranium Utilization in LWR's," paper presented to the American Nuclear Society, June 10, 1981.

10) Using the curve shown in figure 6 of ref. 9 for the economics of a 30 percent improved LWR and the estinate on p. 9 of ref. 9 that advanced isotope separa-235 tion systems could strip enrichment tails from 0.2 to 0.05 percent U at a cost equivalent to $43 per pound U 038

11) Ref. 5, p. 177.

12)

See e.g., Joel Primack and Frank von Hippel, Advice and Dissent: Scientists in the Political Arena (New York; Basic Books,1974; New American Library, 1975), Chapters 2 and 4.

13.

Steven V. Roberts, "Public Works Projects Squeak Through in Senate,"

New York Times, November 5,1981, p. A20.

14 Robert D. Hershey, Jr., " House Panel Opposes Reactor," New York Times, May 8, 1981, p. D3.

n -

Princeton University scNoot o, NciNx.iNcfa,,timosei,Ne, CENTER Foa ENERcY AND ENYiRONMENTAL STUDIES rNR ENGINEERING QUADRANGLE

' ~ ~ ~

~

patwearow, waw ysasar o8544

~

January 14,1985 Mr. Wallace R. Kornack, NE-6GTN Office of Nuclear Reactor Programs Office of the Assistant Secretary for Nuclear Energy U.S. Department of Energy Washington, D.C.

20545

Dear Mr. Kornack,

This letter is in response to DOE's request for comments on the Draf t Supplementary Environmental Impact Statement on the Liquid Metal. ____

Fast Breeder Reactor Program (DOE /EIS-0085-D).

I will not comment on the technical details of this draf t supplement at this time because it is missing an essential part which is required _to_.

make it meaningful - namely, a cost / benefit analysis of the proposed LMFBR Program.

As I will show below, the DOE has recently completed all the elements of such an analysis, and has concluded both that the U.S. has plenty of low cost uranium to support light water reactors for many decades and that the LMFBR will not be economically competitive with light water reactors for as far in the future as DOE has made projections (40 years). This is quite a different conclusion than that which was arrived at in the original EIS on the LMFBR Program where the AEC and ERDA argued that a uranium shortage was imminent in the U.S. and that the LMFBR would be economically competitive in the 1990's.

The DOE's failure to reveal in the Draf t Supplementary EIS the collapse of the basic rationale of the LMFBR demonstration program is, therefore, in eff ect if not by intention a coverup. For this reason I request that this Draf t Supplementary EIS be withdrawn and be replaced by one which contains the updated cost / benefit analysis. Below I will discuss in more detail the essential ingredients of this cost / benefit analysis and why it is critical to the reconsideration of the LMFBR Program at this time.

I will also comment on the reasons given by the DOE for not including such a cost / benefit analysis in this Draf t Supplement.

i

0 4

• The Cost / Benefit Analysis and its Importance to a Reconsideration of the LMFBR Program As the Draft Supplementary EIS states (p. 3):

Cost / Benefit Analyses of the LMFBR program were included in WASH-1535 and ERDA-1535.

WASH-1535 and ERDA-1535 are the AEC's proposed and ERDA's final LMFBR Program Environment Statement, respectively. These analyses were published in 1974 and 1975 and provided the basic rationale for the decisions made in that time period to proceed with an LMFBR program aimed at commercialization in the 1990's.

The basic argument presented in WASH-1535 was quite straightforward and can be summarized as follows:

In 1974 WASH-1535 projected U.S. nuclear capacity at 1200 Gw(e) in e

the year 2000 and 3300 Gw(e) in the year 2020;l e It also estimated that the U.S. resources of low cost uranium could support only about 1000 Gw(e) of LWR capacity; The AEC also believed at the time that the breeder would be economically e

competitive with LWRs fueled by even low cost uranium;2 e As a result the AEC concluded that it was necessary and cost-eff ective to commercialize LMFBRs as soon as possible.

By 1981, however, the picture had completely changed:

It had become quite clear that the historical decline of real electricity e

prices had ended and that in fact real prices could be expected to increase for at least a decade, o As a result it was clear that the period during which U.S. electricity demand doubled every decade had also passed and that in the future U.S. electricity demand would, like the demand for the products of most other mature industries, grow little or nc more rapidly than the economy as a whole. Accordingly, by 1981 the DOE's midrange projection for U.S. nuclear capacity had fallen to 175 Gw(e) for the year 2000 and to 290 Gw(e) for the year 20203 -- capacities which were respectively one seventh and one eleventh of those wnich had been projected by the AEC only seven years earlier; With these new projections the DOE found that, instead of predicting e

that the U.S. will be exhausting its uranium resources by about the year 2000, it was new estimating that even by 2020 U.S. LWRs will have consumed only about one quarter of the nation's resource of low cost U03 8 (less than $100 per pouad forward cost).4,5

e

• During this past seven years the DOE has also concluded that, e

even a large breeder system fully enjoying all the available economies of scale in the production of reactors and in fuel cycle facilities, wil3 not be able to compete economically with LWRs operating on a once-through fuel cycle until the cost of U 0g rises to extremely 3

high levels.

In 1979, in its report on the Nonprolif eration Alternative Systems Assessment Program, the DOE estimated that the LMFBR would become competitive with a once-through LWR system with 15 percent improved uranium efficiency only when the cost of U 03 8 rises to somewhere in the range of $115-205 per pound.6 Including nonretrofittable cost-effective improvements in the uranium efficiency to new LWRs and advanced isotope separation technology for enrichnent tails str per pound of U 08.ppping would raise this crossover range to $150-250 3

These numbers are 2-3 times DOE's 1981 estimate of the price of U 038 in 2020: $78 per pound.6 As a result of this changed situation, a revised cost analysis presented e

in the Supplementary EIS based on the most recent DOE analyses would show that DEBRs will not be economic until far beyond the DOE's furthest horizon - 2020.

Of course, the nation could decide to proceed with the program anyway.

The purpose of an Environmental Impact Statement, however, is to lay out tradeoff s involved so that they can be subjected to public and peer review.

DOE's Reasons for not Including a Cost / Benefit Analvsis in the Draf t Supplementary EIS On p. 3 of the Draf t Supplementary EIS it is stated that

... no such further [since ERDA-1535] cost / benefit analyses have been performed and none, therefore, are included in this supplement...

As my discussion above demonstrates, however, the DOE has performed all the essential parts of an updated cost / benefit analysis.

The EIS then continued on pages 3 and 4 to give three additional reasons why an updated cost / benefit analysis has not been included in the Draf t Supplemen-tary EIS:

1)

Cost / benefit analyses are not required in an EIS (see CEQ regulations, 40 CFR 1502.23)...

In the light of the description above of the conclusions which can be drawn from the analyses which the DOE has made, this legalistic statement gives the impression that the DOE finds the results of its updated cost / benefit analyses unwelcome and does not wish to bring them to public attention.

. 2)

Cost / benefit information for alternative long-term technologies (fusion and solar electric) has not been developed to a degree that would make cost / benefit analyses of these alternatives meaningful.

This may be true, but it is also irrelevant.

If, as it appears from current DOE analyses, the LMFBR cannot even compete for many decades with other fission technologies such as the LWR, why should the nation move ahead now with a demonstration-commercialization program? This question can be answered without any information about the long-term prospects of nonfission technologies.

3) Parameters (e.g., discount rate (s). LMFBR introduction date(s),

future nuclear capacity, future cost of coal) used in complex cost / benefit analyses of the LMFBR are so uncertain at present that the value of such analyses would be questionable. It is the goal of the breeder research and development to reduce such uncertainties.

First of all, the principal focus of the LMFBR Program described in the Draf t Supplementary EIS is to demonstrate the hardware of LMFBR power plants. This program has very little resemblance to a research program on: the uncertainties in the discount rates used to determine the value of such a program, in the future of U.S. nuclear capacity, or even uncertainties in the future cost of coal!

Secondly, the uncertainties in the parameters which are critical to a cost / benefit analysis of the breeder - future U.S. nuclear capacity growth, the magnitude of U.S. uranium resources, and the capital and fuel cycle cost differentials between LMFBRs and LWRs -- have been significantly reduced since the AEC-ERDA cost / benefit analysis was published.

Indeed, it appears from the DOE's own analyses that they have been reduced enough so that the values of the key parameters used by the AEC and ERDA in their justification of the LMF1R demonstration program are now way outside the remaining uncertainty bonds and that, as a result, it is pointless to go ahead with an LMFBR demonstration program at this time.

On page 43 the Draf t Supplementary EIS states that:

the prudent course is to gear the development program toward possible commercialization of LMFBRs fairly early in the next century.

Yet, at the same time, the DOE has refused to present in this document its own analyses which support by a very wide margin a conclusion that the LMFBR will not be needed early in the next century, i

in the past the AEC, ERDA and DOE all accepted the basic assumption which led to the requirements of Environmental Impact Statements: the public has right to expect the government to present the rationale for its proposed programs for public and peer review. This was done in WASH-1535, and ERDA-1535. A number

{

of independent policy analysts took a great deal of trouble to critique these analyses 9 and, as I have demonstrated above, the DOE ultimately changed its own l

1

projections drastically. Yet now the DOE, like the tailors in Hans Christian Andersen's fairytale, demands that the public admire the invisible new clothes which it has produced in this Draf t Supplementary EIS and accept the bland recommendation that to proceed with the LMFBR demonstration program would be

" prudent."

The requirements that governmental agencies prepare Environmental Impact Statements on their major programs was a big step forward toward providing citizens with access to the information and analyses which they require if their rights as citizens are to be meaningful in an increasingly complex society.

In this context, acceptance of this Draf t Supplementary EIS would be a step backward s.

I therefore request, both in the interests of good public policy in this case and in the interests of good government more generally, that the DOE withdraw this Draf t Supplementary EIS and publish a new draf t which includes the results of DOE's updated cost /berafit analysis.

Sincerely yours,

~'

Frank von Hippel FvH/ k i

l l

References and Footnotes

1) US AEC, Proposed Final Environmental Statement, Liquid Metal Fast Breeder Reactor Program (WASH-1535, December 1974), p.11.2 - 11.3.

2) ref. 1. pp. 11.2-4, 11.2-10, and 11.2-30.

3) US DOE, EIA, Annual Report to Congress, 1980: Vol. 3 Forecasts,

{ DOE /EIA-0173 (80)/3), p. 158.

4) Compare Ref. 3, p.177 (converting primary energy released into pounds of U 0 at the rate of 170 million Btus per pound) with ref. 5.

38

5) US DOE, An Assessment Report on Uranium in the United States of America

[GJD-111(8), 1960), p. 1.

6) US DOE, Nuclear Proliferation and Civilian Nuclear Power: Report of the Nonproliferation Alternative Svstems Assessment Program (Draft DOE /NE-0001, 1979), Fig. 11.

7) Using the curve shown in ref. 6 fig. 6 for the economics of a 30 percent improved LWR and the estimate in ref. 6 (p. 9)that advanced iso-tope separation systems could strip enrichment tails from 0.2 to 0.05 percent U235 at a cost equivalent to $43 per pound U 0 38

8) ref. 3, p. 177.

9) See e.g., the report to ERDA by the following members of ERDA's LMFBR Review Stearing Committee; Thomas B. Cochran, Russell E. Train, Frank von Hippel and Robert H. Williams, Proliferstion Resistant Nuclear Power Technologies: Preferred Alternatives to the Plutonium Breeder (April 6, 1977) and the subsequent publication by Harold A. Feiveson, Frank von Hippel and Robert H. Williams, " Fission Power: An Evolutionary Strategy," Science. January 29, 1979, p. 330.

I Uranium Electricity and Economics Frank von Hippel Center for Energy and Environmental Studies Princeton University Princeton, New Jersey 08544 Invited Testimony before the Subcommittee on Energy Conservation and Power, of the House Committee on Energy and Commerce.

i October 5, 1981 l

_ _ _ _. _ =.

i d

It is my understanding that the purpose of these hearings is to explore one approach by which the utilities might reduce the contribution of uranium costs to the price of nuclear generated electricity. I would like to start, however, by putting this problem in perspective.

Uranium is Cheap Figure 1 shows the relative costs of the different fuels used by our utilities in 1980 and as projected in the mid-price scenario in the Energy Information Administration's 1980 Annual Report to Congress.

It shows that uranium is currently about one fif th as costly as coal per kilowatt-hour generated and that, according to the latest DOE proj ections, it will still be about a fifth the cost of coal in 2020. The reason is quite simple: uranium is such a concentrated source of energy that while it is necessary to mine approximately one thousand pounds of uranium ore to recover one pound of uranium, the uranium in one pound of average uranium ore suffices to generate more elect-ricity in current nuclear power plants than can be generated by the burning.of ten pounds of coal.

Whv Did the AEC Choose the Breeder?

If uranium is goir g to stay so cheap, the question naturally arises: Why has this nation for more than a decade been pouring such a large fraction of its energy research and development dollars into the breeder reactor program, a program which has as its only objective the further reduction of the already very small uranium costs being paid by the operators of nuclear power plants?

In order to answer this question 1* is necessary to go back and look at the assumptions about the future that the AEC was making when it committed the nation to the breeder program. The AEC made many assumptions which have turned

out not to be true about the economics of the breeder - assumptions which I will discuss later. The most important mistake that the AEC made, however, was in its assumption concerning the future growth of U.S. electricity consump-t ion.

Figure 2 shows the nuclear power growth projections which the AEC made in 1974 when it did its last cost-benefit analysis on the plutonium breeder.

The AEC s projecting that by 2020 U.S. nuclear power plants alone would be generating 10 times as much electric power as the entire U.S. electrical gene-ration system did in 1980! Since t.be AEC thought that the low cost uranium resources of the U.S. could supply the lif etime fuel requirements of only about 1000 Gigawatts of light water reactor capacity and that U.S. nuclear capacity would reach this level by about the year 2000, the Commission concluded that

~

af ter about the year 2000 all new U.S. nuclear power plants would have to be breeder reactors.

Electricity Prices and the Growth of Electricity Demand Although the AEC's electricity demand growth projections may seem absurd now, at the time they were made they were simply an extrapolation of the pre-1970 exponential growth of U.S. electricity consumption. During the 40 year period 1930-1970 U.S. electricity consumption had approximately doubled each decade -

growing approximately twice as rapidly as the U.S. GNP (6.6% versus 3.2% per : rear) so that by 1970 the U.S. was consuming four times as much electricity per dollar of GNP as in 1930.

(See Figure 3.)

During this same period something else remarkable had been happenir.g as well, however, which we now realize was stimulating the enormous rate of growth in U.S. electricity demand:

the real price of electricity had been drepping steadily (except for a brief period in the Depression) so that by 1970 more than 4 kilowatt-hours of electric energy could be bought for the same 15 cents in 1980 dollars which would have bought only one in 1930.

(See Figure 4.)

The long period of declining electricity prices was made possible by the dramatic increase of the efficiency with which power plants converted fuel into electricity prior to 1960 and the dramatic increase in the size of central station coal and nuclear power plants between 1960 and 1970.

While many in the AEC expected that the introduction of nuclear power plants would make possible a continuing decline in the cost of electricity (some even predicted that nuclear electricity would become so cheap that it would be "too cheap to meter"), these expectations have not.been borne out.

Af ter 1970 both thermal efficiencies and the sizes of new central station electrical generating units plateaued and the increase in real fuel and capital costs began to drive electricity prices up.

As Figure 3 shows, following the reversal of the price signals to consumers, the growth rate in electricity consumption has slowed to about the same rate as that of the economy.

This does not mean that in the future electricity will not continue to become increasingly important in our economy relative to other energy forms.

Indeed, during the past decade the share of U.S. primary energy consumption going to the generation of electricity continued to grow: from 24 percent in 1970 to 33 percent in 1980. The reason for the growing relative importance of elec-tricity in the economy has changed, however, from being due to a very high rate of growth in electricity consumption to being due to a very low (perhaps in the future a negative) rate of growth in the consumption of primary energy for all other purposes.

Future Nuclear Power Growth The new conventional wisdom is that in the future U.S. electricity consumption will continue to grow at about the same rate as the GNP and the latest projec-tions by both the government and the electrical power industry are roughly 1

d,

c.unsistent with this expectation. Thus, for examole, in its most recent projec-tions,the Energy Information Administration assumed that both the size of the U.S. economy and U.S. electricity consumption in 2020 will be 2.5 times greater than today.

The {IA also assumed in its midcase projection that in the year 2020 about 30 percent of t,his electricity would be generated by nuclear power.5 view this is a reasonable midrange estimate for the fraction of U.S. electricity which niight be generated by nuclear power plants 40 years from now.

I believe that, even allowing for lots of electric cars and heat pumps, U.S. electricity consumption need not grow by as large a factor as the EIA projects, however.

Indeed, recent analyses show that the U.S. economy would be greatly strengthened if much of the money which would be spent on building new electrical generating capacity in the EIA scenario were invested instead in renovating our buildings a.

and industry. In the course of such a renovation the energy efficiency of these

+

facilities could be increased enough to eliminate the demand which the electricity generating capacity would have been built to serve.

From my yarspective, therefore, the DOE projection of future U.S. nuclear generating capacity is probably still too high.

In any case I show the 1980 DOE projection on Figure 2 along with the AEC's 1974 projection of U.S. nuclear capacity growth. The essential fact to note is that the 290 Gigawatts, shown there for 2020 are far below the 1000 Gigawatts that the AEC calculated could be supported by U.S. reserves of low cost uranium.

This is why the AEC's nightmare about a short. age of cheap uranium has receded into the future by many decades.

Of course, even if the problem of future uranium cost now looks relatively minor for a very long time into the future, there is no reason we shouldn't try to reduce it even further if that can be done in a cost eff ective manner.

. The Cost of the Breeder The breeder would solve the problem of rising uranium costs because it would ultimately reduce by one hundredfold the uranium requirements per kilowatt of nuclear electricity generated. As a result the contribution of the cost of uranium to the cost of nuclear power could be reduced from about 0.9 cents in 2020 p)r kilowatt-hour to essentially zero.

The breeder would also eliminate the need for enriching uranium which, according to the Department of Energy projection, will cost about 0.25 cents per kilowatt-hour in 2020.

Only if the extra costs associated with the breeder amount to less than one cent per kilowatt-hour, therefore, will it be competitive with current unimproved light water raactor technology.

(See Figure 5.)

As I will explain below,the breeder fails this test by a large margin.

While this technology would have been an effective " brute force" solution to the AEC's concern about the possibility of uranium costs climbing out of sight, it is too expensive to be competitive in a period of rising but still relatively very low uranium costs. The high costs of breeder generated electricity would be due firstly to the high cost of the reactor and secondly to the complexity and difficulty of its fuel cycle.

The Reactor l

i Six years ago ERDA projected that the first commercial breeder reactor would cost about 20 percent more than a 1980 vintage light water reactor of equivalent siz e.

Such a plant is now nearing completion in France.

It is well i

designed -- indeed many experts think that the French pool-type design is much superior to the loop-type design which has been pursued in the U.S.

The develop-ment program has been well organized and ef ficiently run. And no licensing delays have occurred in the commerciali2ation effort since non-governmental critics are

excluded from the French nuclear power licensing process. Yet the French now project that the construction cost of the " Super Phenix" will be more than twice that of a light water reactor of equivalent capacity.9 Of course, some of the extra cost of this plant stems from the fact that it is a first-of-a-kind plant. In routine production the French Atomic Energy CoImnissioner estimates that the cost of the Super Phenix could be brougli:

down to only 1.'75 times the cost of a light water reactor.

It is hoped that the cost may be brought down still further - perhaps to between 1.3 and 1.45 times the 04R cost if the safety margins in the reactor design are somewhat reduced. Even if the cost differential could be reduced to 40 percent, however, the extra capital charge for the breeder would all but eliminate the entire savings associated with the breeder's lower uranium and zero enri hment require-ments projected by the DOE.

(See Figure 5.)

The Riel Cycle Services Af ter a certain time breeder fuel would have to be reprocessed chemically l

l so that the plutonium and urana could be recycled.

(See Figure 6.)

Once again, however, the French, who : sve led the world in commercializing reprocessing technology, have found it to be much sore expensive than they and the U.S. AEC had expected. The official French estimates of reprocessing costs increased 1

tenfold in constant francs between 1970 and 1980 and have now reached a level where they would add another 0.45 cents per kilowatt-hour to the cost of the breeder-produced electricity.1 '

DOE estimates of breeder reprocessing costs i

based on paper studies are about half this large (Both numbers are shown on s

l Figure 5.)

There are also extra costs in the breeder fuel cycle when the fuel is being ref abricated because of the requirements for extra protections for both workers and materials whenever plutonium is being processed. Taking the 1979 DOE cost l

4 estimates and updating them only for general inflation, I find that the extra cost for the f abrication of fuel containing plutonium adds approximately another 0.2-0.4 cents per kilowatt-hour to the cost of breeder generated electricity.

The extra fuel cycle costs,of course, increase the margin by which the cost of breeder electricity can be expected to exceed that of light water reactor generated electricity in 2020.

(See Figure 5. 6)

There are additional smaller factors which bear on breeder economics and the cost assumptions used by different analysts differ, due to the inherent uncertainties associated with discussions of a technology which is not yet fully commercialized.

I am unaware, however, of any recent analysis which shows the plutonium breeder reactor becoming competitive with a light water reactor at a cost of uranium-oxide of less than $90 per pound - the cost which the DOE projects for 2020. The most recent DOE study of breeder economies, for example, showed a crossover point in the range of uranium costs of $150-270 per pound (1981 $).17 Considering this background I think that Congress might well ask itself whether the nation should continue to spend over one billion dollars a year

" commercializing" a technology which even its advocates don't expect to be commercially viable for at least 40 years. Congress dropped the U.S. supersonic i

transport commercialization program in 1970 and let the French show the world that it was the right decision.

In the case of the breeder the French seem once again to have done us the same favor. Indeed, if we do not move soon, the French may cancel their breeder commercialization program before we cancel ours.

An Alternative Evolutionarv Approach I would now like to return to what I understand is the primary purpose of these hearings: the investigation of an alternative approach to uranium conservation involving evolutionary improvements in the current once-through syst em.

I am very much in favor of such an approach and would like to discuss it in the' remainder of my statement.

The current once-through fuel cycle has a number of advantages of which I will mention three:

e It is simple. The contrast with the breeder fuel cycle is striking in this regard. Both reprocessing and plutonium fuel fabrication plants have proven extremely difficult to operate reliably. The reprocessing plants at Windscale, England; La Hague, France; and West Valley, New York all had lengthy shutdowns because of accidents or unsafe working conditions and on average have only worked at a fraction of their design capacities for reprocessing light water reactor fuel, a

The reprocessing of breeder fuel would be even more difficult. Since each reprocessing plant would recycle the fuel for about 50 reactors, a lengthy breakdown could be economically catastrophic if it shut down the associated reactors. This problem could be mitigated by building extra reprocessing plants or by stockpiling a year or more extra breeder fuel but the large expenses involved would make the economics of the breeder even worse.

l e The safeguarding of plutonium in the spent fuel is relativelv easy.

The highly radioactive fission products in spent fuel are relatively eff ective in protecting the plutonium from diversion and, because the fuel rods are countable, one can keep track of the plutonium which they l

s I

contain with no measurement error. In contrast it has proven impossible in the Department of Energy's weapon's program to verify that plutonium has not been diverted.

In 1977 ERDA announced, for example, that the cununulative inventory difference in AEC-ERDA facilities through September 30, 1976 was about 1.5 metric tonnes of plutonium - enough to make about 200 Nagasaki bombs.

e The radioactive waste disposal problem is relatively uncomplicated for the once-through fuel cycle. In the past, various arguments have been raised to the effect that recycling and consuming plutonium significantly reduces the long term hazard of radioactive waste.

By now, however, it is generally accepted that this is not the case and that considerations related to radioactive waste dis-posal do not favor plutonium recycle.

Indeed history so far suggests just the opposite. The military reprocessing program lias shown reprocessing and plutenium f abrication facilities multiply the number and difficulty of waste forms and greatly increase the volume of contaminated material which requires disposal. The Department of Energy has created horrendous radioactive vaste disposal problems at Hanford, Oak Ridge and elsewhere as a result of its reprocessing and plutonium operations. We should, therefore, be in no hurry to break open the metal cladding which today separates from the outside world the even greater quantities of radioactivity contained in our spent power reactor fuel.

The Potential for Uranium Efficiency improvements in Once-Through Systems The advantages of the once-through fuel cycle suggest that we should do what we can to increase it's viability and longevity.

Despite all the concern that has been expressed in the past by the U.S. nuclear industry about the

. 1!nited U.S. resource of low cost uranium, however, the reality has been that uranium has been extremely cheap and U.S. utilities have not been particularly interested in uranium efficiency. It should not therefore be surprising that,

~

when the reactor manufacturers and national laboratories were asked by the Department of Energy to explore the possibilities for making uranium efficiency improvemeits in the once-through fuel cycle, they found that the current systems are " uranium gu'zzlers." Apparently cost-effective retrofittable improvements were identified which could increase the amount of electricity that can be 21 generated from a pound of uranium by a factor 1.25 and additional cost-ef f ective improvements were identified which, if they were incorporated into new reactors, could increase this resource extension f actor to 1.5.2

Making High Cost Uranium Economic The payoff from uranium efficiency improvements could be considerably greater than a factor of 1.5 extension in the amount of energy which can be generated by light water reactors, because these Laprovements, in addition to extending a fixed resource base would make it economical to exploit lower grade r

ores than before. Consider, for example, the DOE's 1977 estimate of the uranium supply curve shown in Figure 7.

Assume also that we are willing to epend up to 1.5 cents per kilowatt-hour on uranium. With current reactors this would make

(

uranium costing up to $180 per pound of oxide affordable. With a one third reduction in uranium requirements per kilowatt-hour, uranium costing 50 percent l

more or $270 per pound would become affordable on the same baais.

In the example shown in Figure 7, at least 1.5 times as much uranium is available at prices of $270 per pound as at $180 per pound.

. This extension of the economically exploitable resource base compounds with the improved system's ability to extract 1.5 times much energy out of each pound of uranium so that we can more than double the number of kilowatt-hours t hat we can economically generate with a once-through fuel cycle (1.5 x 1.5 = 2.25).*

Reducing the Spent Fuel Problem One oT the approaches which would be used to increase the uranium efficiency of the once-through fuel cycle would involve increasing the percentage of the atoms fissioned in the fuel from about 3 to 5 percent.

An important side benefit of this would be the reduction of the rate af discharge of spent fuel from the reactor by about 40 percent. This would probably also reduce spent fuel storage and disposal costs by a sinilar percentage.

A Federal Program to Laprove the Once-Through Fuel Cvele Since the light water reactor improvements being discussed are incre-mental, the associated research and development costs would be very small in compar!4on to what would be required to bring on line a whole new reactor and fuel cycle.

A federally funded improvement program for light water reactors would have a number of other advantages as well, including the following:

e It would be of real current interest to the utilities A strategy of incremental improvements would therefore be much more eff ectively disciplined by the " marketplace" than a development pregram for a whole new reactor-fuel cycle system which may never be deployed i

commercially.

• In this simplified discussion I have assumed that there will not be increased costs associated with light water reactor efficiency improvements which will signi-l ficantly off set the uranium cost savings. This assumption appears to be approximately valid, however, since, according to the DOE analysis, the efficiency improvements are already economic at :he current low uranium cost of about $30 per pound.

o There would be saf ety and reliability benefits. Shif ting some of the nation's nuclear energy R&D talent back to work on light water reactors might help solve the safety and reliability problems which are f cereasingly plaguing these reactors. Indeed ! think that one of the principal reasons for our current troubles was the fact that the AEC diverted its best people to work on breeder R&D just when the safety problems of large light water reactors were beginning to be recognized.

e The U.S. would still have a breeder " option".

Some breeder advocar.ss are concerned that the U.S. may forget everything it ever knew about liquid metal technology and f ast neutron reactors and that the time may come

- even if it is 100 years from now - when we need more uranium efficiency than can be achieved by light water reactors on a once-through fuel cycle.

These people may be reassured to learn that we will still have a breeder option as 1 0re as we use light water reactors. Admiral Rickover's group has shown that, if reprocessing is allowed, it is possible to increase the conversion ratio of light water reactors up to any level including that of a "breakeven" breeder reactor.* The cost of electricity from thecc light water. breeders would probably be about the same as that from liquid metal breeders.**

• A " break even" breeder would produce about as much " fissile" (chain reacting) material as it consumed. While a true breeder such as the liquid metal fast reactor could produce enough excess fissile material to start up new reactors at a certain rate without any requirements for inputs into the system of the only naturally occurring fissile material, uranium-235, the U.S. resource base is adequate, to start up a system of break even breeders, of any reasonable size.

The AEC excluded break even breeders from its program on the basis of unreasonable nuclear power growth projections such as that shown in Figure 2.

• • While the capital cost of a light water breeder would be lower than that of a liquid metal breeder, its fuel cycle cost would be higher - principally because the fuel would have to be recycled at least twice as frequently.

A Saf er, Less Costly Alternative In summary, it would appear that:

e With any reasonable nuclear power projection, the U.S. has enough uranium to fuel for generations a system made up exclusively of reactors operating on a once-through fuel cycle; o This system would p oduce electricity at lower cost and more reliably than the breeder; and The more money that the Department of Energy has put into nuclear R&D e

- billions for the breeder and millions on once-through systems the wider has the gap between breeder and light water reactor economics become. We have 1e.arned that the breeder system will be much more expensive than had been hoped and that the once-through system can be r

made significantly cheaper and more uranium efficient.

1

{

i l

=. - _. _ _ _ - _ -

_. - =

l<

1 Footnotes and References 1)

U.S. DOE Energy Information Administration, Monthly Energy Review, August 1981, p. 86.

2)

U.S. DOE Energy Information Administration,1980 Annual Report to Congress, Vol 3: Forecasts, pp. 126, 177. I assume that one 197 96 = 1 JZ l'El S.

Eleven thousand Btus is approximately the amount of heat required in a steam-electyicpowerplanttogenerateonekilowatt-hourofelectricenergy.

3)

U.S. AEC, Proposed Final Environmental Statement on Tha Liquid Metal Fast Breeder Reactor Proaram, December 1974, p. 11.2-113.

I

4) Robert H. Williams, " Industrial Cogeneration," Annual Reviews of Energy, 1978, p. 313.

5) Ref. 2, p. 158.

6) See e.g., Marc H. Ross and Robert H. Williams,.Our Energy: Retaining l

Control (McGraw-Hill, 1981), and Solar Energy Research Institute Report on Building a Sustainable Energy Future, U.S. House of Representatives Committee on Energy and Commerce, Committee Print, 1981).

4

7) Ref. 2, p. 159.

a C

8)

U.S. Energy Research and Development Administration, Liouid Metal Fast Breeder

}

Reactor Program: Final Environmental Statement (1975), p. III, F-12.

2 i

9)

M. Hug, Revue de l'Eneraie, February 1980, p. 71 (Dr. Hug is a director of q_

}

Electricite de France, France's electrical utility.)

10)

C. Pierre Zaleski:, " Breeder Reactors in France," Science, April' ll,1980, p.137.

11) According to current Department of Energy estimates, a new U.S. light water reactor will cost about $1500 per kilowatt generating capacity in 1981 dollars (ref. 2, p. 274). Forty percent of $1500 equals $600 which, in constant dollars, would have associated with it an annual capital charge of $60 (compare ref. 2, j

pp. 262 and 274), which would amount, at a 65 percent average capacity utili-l zation factor, to a capital charge of 1 cent /kilowett-hour.

12) The official estimates of reprocessing costs are made in France by the PEON Commission, a majority of whose aanbers come from the French Atomic Energy Commission, Electriciti de France and the French' Treasury. Between 1970 and 4

1980, PEON Commission estimates for the costs of reprocessing' LWR fuel have risen from $100 to approximately $1000 per -kilogram of heavy metal in the fuel (1980 S).

In most analyses it is assumed that the reprocessing of breeder fuel will cost somewhat more than the reprocessing of light water reactor tuel, but I have assumed $1000 per kilogram.

[These numbers were kindly 1

provided by Professor Claude Henry of the Ecole Polytechnique in a letter to my colleague, Robert Williams, dated June 24, 1981. I have assumed that one French Franc.= 0.23 S.]

}

-. ~..... -..---.-

. -- =~

i R-2 (12 cont.)

A 1000 Megawatt (electric LMFBR with a hoacgeneous core, operating.

on a plutonium fuel cycle at a 70 percent average capacity factor, would discharge about 26 metric tonnes of heavy metal per year (ref. 13).

This corresponds to a gvaaration of 0.24 x 106 kilowatt-hours per kg. of heavy metal.

)

13)

U.S. DOE, Nuclear Proliferation and Civilian Nuclear Power (Draf t report of i

the Nonprolif erstion Alternative Systems Assessment Program), Vol. IX. Reactor and Puel Cycle Descriptions. Tables B.2 and 5.6.

i l

14) The 11.ff erential estimated in ref.15, between the costs of fabricating breeder oxide fuel (including axial blanken) for operation ou a plutonium fuel cycle and LWR fuel for the current once-through fuel cycle ranges $540 to $870 per kilogram of heavy metal in January 1,1978 dollars. One of these dollars was wocch about $1.35 in 1981 dollars. The differential in overall fabrication costs is reduced somewhat by the fact that the fabrication of the radial blan-ket in a breeder (which I assume to be 25 percent of the heavy metal flow through 4

the reactor) would cost little more than the fabrication of once-through LWR fuel.

.I

)

15)

U.S. DOE, Nuclear-Prolif eration and Civilian Nuclear Power. Vol. V: Economic s i

and Systems Analysis, Tables A-5 and A-6.

16)

In Figure 5 I have assumed breeder-LWR cost differentials of $600 and $1125 per kw(e) capacity corresponding respectively to 40-75% increases over a

$1500 LWR base cost.

I have assumed reprocessing costs of $600 and $1100 per kilogram heavy metal based respectively on DOE estimates of $450 - $800 in ref. 16 and PEON estimates for LWR fuel reprocessing costs (see foot-I note 12). Finally, I have assumed cost differentials between breeder and LWR fuel fabrication costs of $400 and $900 per kilogram heavy metal based on ref. 15.

All costs have been converted to 1981 S.

Uranium costs would have to increase to $120 and $240 per pound respectively for the breeder savings to equal the two estimates of breeder total incremental costs shown on Figure 5.

17) Ref. 15, p. 44.

I

18) ERDA Press Release, "ERDA Issues Report on Inventory Diff erences for Strategic Wclear Materials," August 4,1977.

19) Hartmut Krugman and Frank von Hippel, " Radioactive Waste: the Problem of Plutonium," Science October 17, 1980, p. 319.

20)

U.S. ERDA, Proliferation Resistant Nuclear Power Technoloates: Propo sed Alternatives to the Plutonium Breeder, report by Thomas B. Cochran, Russel E. Train, Frank von Hippel and Robert H, Williams, April 6,1977; Harold A. Feiveson, Frank von Hippel and Robert H. ' Williams, " Fission Power:

4 An Evolutionary Strategy," Science, January 26,197 9, p. 33 0.

8 21)

Kef. 13, pp. 13, 14 22)

S. Newman, S. Goldsmith, and R.M. Fleisciutan, " Assessment of Nonbackfittable Concepts for Improving Uranium Utilization in LWR's."

Abstract of Paper i

presented to the American Nuclear Society, June 10, 1981.

a I

Figure 1 i

i i

PROJECTED FUEL COSTS

)

FOR US ELECTRICAL POWER PLANTS (DOE-EIA,1980, MIDRANGE )

d

~,

?

2 RESIDUAL. OIL m

15 8

O,.

=

8x i

H g

10 S

x cc GAS E

Wz O

5<

^

COAL r

q URANIUM

=

i 1980 1990 2000 20!O 2020 i

Figure 2 3300 3000 m

f.

<R s

<0 AEC(1974)

I 6

m 2000 o

i 1>

~

I

>-b Y

l LMFBR s

W l

1000 e

m

~'y e'

l s' '"(

~~. /

4

,/

H EQUIV. TOTAL U.S CAPACITY'

\\o LWR +HTGR 290 DOE (1980 O

1970 1980 2000 2020 l

US NUCLEAR GROWTH PROJECTIONS

'USING THE 80 PERCENT AVERAGE CAPACITY FACTOR ASSUMED BY l

TE AEC IN 1974 l

Figure 3 O.9 O.8 ELECTRIC ENERGY CONSUMED PER UNIT OF GNP d

(kwh/I980$)

0.7 0.6 0.5 O.4 0.3 0.2 0.1 t

l t

t I

i 1920 1930 1940 1950 1960 1970 1980 W

0

l Figure 4 25 i

i i

i i

~.

20 AVERAGE PRICE OF ELECTRICITY (1980 ( PER KwH )

15 10 5

O 1920 1940 1960 1980 l

Figure 5 DOMINANT FACTORS IN BREEDER ECONOMICS IN 2020 (1981 cents /kwh) 3 SAVINGS COSTS (DOE,HIGH)

BREEDER-LWR FUEL-FABRICATION COST DIFFERENTIAL

( FRENCH,)

i m

LWR

• ag2 I

9s m

E (DOE, LOW)

Q d

(DOE)

REPROCESSNG l

ENRICHMENT l

yl (DOE)

(PRODUCTION 8

~

BREEDER-LWR PHE

)

CAPITAL

( FRENCH

' COST

/

i l

LOW COST DIFFERENTIAL I

URANIUM 4590/1b.,)

DOE

• Based en French experience with reprocessing LWR fuel.

Figure 6 t.DW ENRICHED t

URANIUM BREEDER SYSTEMWf1H ONCE-THROUGH PLUTONIUM-URANIUM RECYCLE f

d URANIUM URANIUM MILL MILL U

4 1 t URANIUM ENRICHMENT U

LEU 4

FUEL FABRICATION FUEL FABRICATION Pu, U LEU h

Pu,U CONVERTER REACTORS FUEL C

V REPROCESSING Pu,U,FP p

CTORS v

v SPENT RADIOACTIVE FUEL WASTE STORAGE STORAGE M

(b)

NUCLEAR FUEL CYCLES

J

..s,.

Figure 7 7

i i

i i

6 -

^

m o

m o

5 U0 SUPPLY ESTIMATE 3

b

( DOE,1977 )

u)

ZOH 4

u.

O u)

Z9 3

d 5

~

KNOWN DEPOSITS h2 s

u>

22 l -

l k

i O

I I

I I

i 0

50 10 0 15 0 200 250 PRICE (1981 iPER POUND OF U 0 )

3 8

i

~

W i

public

~

power i

M AY-3UNE 1981 VOLUME 39 NUMBER 3 pp. 19,21,24 ll I

ShouldbrDederreaclar8he bUlltlalheUniledState8?

by FR ANK YON HIPPEL Chairman of the Federation of American Pr acet a n we s ty s e t r or nerg>-

i ain"e'"a' ";TLlL";is " "t?

n

': r&'e",,"'s":M';i"b,'ef:

ment Administration.

i L

I sources of high grade uranium ore are i large enough only to fuel about one million mw oflight water reactors over their expected 30-year lifetimes. If the United States wishes to build much more nuclear capacity than that it will have to switch to more uranium effi-cient reactors.

Over the short run relatively small efficiency improvements could greatly

" Fanaticism consists in redoublins your er.

mitigate the uranium supply problem.

forts whenyouhavetorsortenyouraim. "

tonium from the fuel that we had de-The nuclear power pioneers were big veloped within the U.S. nuclear weap-thinkers, however, and they realized

-George Santayana,1905 ons program. The AEC had trained the I BECAME INTERESTED in the proposal nationah of many countries in thi that over a period of millenia a breeder to commercialize the plutonium fuel technique in the " Atoms for Peace} low grade ore that,wou I

in principal at cycle in 1974 when I learned about the program because it was advocating the least., of the earth Since the breed rocks,it would be possible to, burn the enormous flows of materials usable in plutonium breeder reactor as the prime nuclear weapons which would be in-mom of future m rgy systems r

volved. The plutonium discharged an-worldwide.

reactor was W uhnate soMon to &

uranium resource problem, no m, ier-nually from just 100,000 mw of The Indian bomb woke us up to the mediate solution seemed worth the breeder capacity would be enough for fact that the interest of a numper ofbother.

the construction of 10,000 Nagasaki-a ternments in nuclear technology was In 1974 the AEC projected the United size nuclear weapons.

decidedly ambiguous. Indeed, the am. States would have 1.2 million mw of in 1974 the U.S. Atomic Energy biguity of the mierest of Israel, India, Commission (AEC), predecessor of Pakistin, South Korea, Argentina, nuclear capacity on im, e m the year the Nuclear Regulatory Commission Brazil, the shah's Iran, Iraq and other 2000 and approximately twice that (NRC), was projecting that the united

"'t' "5 m purch*5fa8 *ad coa 5tructias much by 2010. (See Figure 1, p.21.)

T**"tY-I', Y'*r5 * "'d be * *ery States would be bringing this much facilities for the separation of pluto-short time to create a breeder reacto breeder capacity on line each year by naum fr m nuclear fuels soon became construction industry which cou the year 2000. I knew that a panel of the cause of some of our government's experts set up to advise the State De-worst foreign policy headaches.

bring on line 100,000 mw of capacity a partment had concluded already in The expectation that a plutonium-year. A major government demonstra-tion and commercialization effort 1946 that "in the real world" interna-fueled future would become a reahty therefore appeared to be justified.

ti:nal safeguards could not be effective was already providing a convenient t

in preventing a nation from diverting c ver f r nations mterested in develop-At the time the AEC made these nu-plutonium or highly enriched uranium ing a nuclear wcapons option. A group clear growth projections it was also from commercial to weapons use.

I three physicists and a pohtscal scien-projecting that total U.S. electricity t st therefore organized themselves production would t'e 10.6 trillion kwh When I inquired whether the situation Prmeeton University to analyze in the year 2000 and 27.6 trillion kwh had clianged since 1946 I was told it at had: The experts were now worried whether the plutonium breeder reactor in the year 2020-up from 1.9 trillion that it would be impossible to prevent was an essential part of the world's kwh in 1974. About three quarters of diversions of materials usable for nu-energy future.

all electricity production after the year cler weapons from a plutonium econ.

2000 was to come from nuclear power cmy even by terrorist groups.

Demand and the Breeder plants operating at 80 percent average The problems of a breeder reactor capacity factors. The overall electricity economy seemed to be problems of the At the time we undertook this effort,in production growth projections were next century, however, only a few peo-1975, the United States was the world based on the assumption that use of pl2 had time to worry about them. This leader in breeder technology. The latest electricity, which had been doubimg word in U.S. policy analysis with re-every 10 years during the 1920-1970 pe-state of complacency did not last very gard to the breeder was contained in riod, would continue to grow at almost 1:ng.

In May 1974 India exploded a nu. the " Proposed Final Environmental the same rate.

cler bomb using plutonium obtained Statement on the Liquid Metal Fast n:t from a plutonium breeder reactor Breeder Reactor Program," a state.

Real Price of kwh Declined but from a research reactor. The In-ment published in Decemb r 1974 by In retrospect it is easy to see that the the AEC.

diin nuclear technologists had used the The case for the breeder at that time historically sustained rapid growth of same technique for separating the plu-was the same as it is now: U.S. re-a

growth projections had been unrealis-tically high, so had its estimates '

WE*8Mbck MD I).S. uranium resources. Their conclu-U.S. electricity consumption prior to stor was that the effect of the two er-M8COOR 1970 was possible only because the real rors cancelled and that therefore the price of electricity was declining rap.

need for the breeder was as urgent as idly during the same period. Figure 2 ever. It appears now however that, if shns that between 1920 and 1970 elec-anything, the AEC 5 uranium resource tricity prices fell in constant dollars at assessment was pessimistic.

an average annual rate of 3.2 per-in 1974 the AEC s median citimate of cent-almost the same percentage by U.S. uranium resources mineable at a which ilectricity consumption was ex.

marginal cost less than about $50 per ceedis.g gross national product (GNP) pound (in 1980 do!!ars) was 2.8 million gr:rteth. Consequently it was possible t ns of uranium oxide. By 1980, as a for thc :.ation to increase its rate of result of the Department of Energy electric 4y consumption at the rapid (DOE) National Uranium Resource historical rate without increasing the

'#'"#'i?"

O' E'#

'h'

'I' "#

share of the GNP devoted to the pur-probability that U.S. uranium re-2 chisc o/ electricity (about 2.5 percent).

sources m this cost category was less Figure 2 also shows, however, that than 2.8 million tons had been reduced real electricity prices bottomed out in hom 50 to 5 percent.

1970 and have in fact since risen by about 50 percent. As a result, in order I".the same period, it was realized that it would be economic to mine for electricity consumption to grow much higher-cost uranium to fuellight-even as rapidly as the GNP since 1970 it water reactors (LWRs). The Department i

has been necessary for funds to be Ji-f Energy published a first estimate verted from other parts of the econ-that about five million tons of U 0s

,,-I..,,,',,

3 omy. Naturally, there has been con- (urankm oxide) would be mineable m sumer resistance to such a shift and the the Un,ited States at a marginal cost of yg result has been a dramatically slower less than $100 per pound. ( At this price g

m growth in electricity consumption.

With the passage of the '70s it has

."I# " "" E

I become clear that the causes of slower

• # . ejectnc3ty generated by

,...I I

light-water is still equivalent to a cost electrical demand growth are not tran-of oil at only $4 per barrel.)

M N N

M sient. The projections of long. term fu-Fise mittion tons of U 0s would 3

ture growth trends have therefore als amount to about 5.000 tons of U 0, 3

been coming down. The September for each of a thousand I,000.m w 1980 ELECTRICAL WORLD forecast for electrical utility generation in the year continued un page.N 2000 was 4.5 trillion kwh-about twice the 1980 level but less than one half the les which the AEC was projecting in y

As projections of future demand 198o cents per KwH growth have fallen, so naturally have projec6cni of future generating capac-ity-most notably nuclear generating c preity. The most recent ELECTRICAL WORLD projection for nuclear capacity on line in the year 2000 is only 150,000 mw. This is still a very large capabil-ity-at 65 percent average capacity factsr it would generate about as much electric energy annually as s!! U.S.

coal-fired plants do today-but it is cnly ene-eighth of frie 1.2 milhon mw which the AEC was projecting in 1974.

It is also far less than the one million mw of light water reactor capacity which the.*EC was projecting in 1974

%w as supportable with the U.S. resource b:se of high. grade uramum ore.

U.S. Ursnium Resources Frr a brief period in 1977 and 1978 ad-e vocates af the breeder reactor argued

gee wee wee tsee that, if the AEC's estimates of nuclear a

~D the construction by the U.S. govern-the DOE that the United States must go ment of a demonstration breeder re-ahead with a breeder demonstration LWR's-enough even at current utiliza-actor "is crucial to the nation's ability program if we are to continue to lead tion efficiencies (which according to to keep pace with foreign breeder tech-the world in nuclear pow er technology.

the Department of Energy, can be in-nology developments."

The final decision on the breeder, how-creased by 15 to 40 percent) to run each The implication is that the United ever, will be made by the market.

of these reactors for about 30 yens at States, by not keeping up with the There too the French will ultimately an average capacity factor of 65 breeder reactor demonstration pro. learn the answer to the question asked P"" ' "

grams of some other nations, is falling by LE MONDE, France's leading news-

~

Breeder Economics behind in some kind ofimportant race. paper, just four years ago: "Is the in 1974, when the AEC was proposing a Unfortunately Behnke does not explain Super Phenix a Nuclear Concorde?"

why it is important for the United The Breeder Legacy breeder reactor commercialization pro-gram, the agency's cost figures indi-States to keep pace with other nations Perhaps a hundred years from now, cared +at U.S. utilities might want a in what looks increasingly like a race to just as we today are dusting off the de-breeder even if the nation had unlim-develop a white elephant.

signs of old windmills and are rediscos-ited supplies of high. grade uranium When I first heard expressions of cring how the Greeks proportioned ore. The AEC projected the cost of the concern from the U.S. nuclear research buildings to Ict in direct sunlight in the first commercial scale breeder would and development establishment that winter and exclude it during the sum-be only 25 percent greater than that of the United States was falling behind mer, our descendants may dust off the an LWR of the same capacity, and that the French and the Soviets in an impor-plans of today's prototype breeder re-thereafter the cost difference would tant area of technology, the warnings actors. I am afraid, howeser, that long rapidly drop to zero. The cost of the had a familiar ring. Then I remem-before that time the current burst of breeder fuel cycle per kwh was pro-bered the great debate over the U.S. su-enthusiasm for this technology will jected to be one-tenth that of the LWR.

personic transport (SST) demonstration have helped spread another industry Thus, even with uranium costing as lit-program, around the earth-the manufacture of tle as $30 per pound (today's approxi-President Nixon commissioned two the most barbarie weapons ever per-mate price) breeder-generated electric-major reviews of this controversial fected by man, ity was projected to cost 25 percent less program just after he came into office.

It often is argued by breeder propo-ihan LWR. generated electricity.

When they were completed, both re-nents that the " genie" of nuclear Things have turned out much differ. ports expressed doubt that either the weapons is already "out of the bottle."

ently, howeser. Recently the French re.

U.S. ssT or the French British Con-If one thing is certain about nuclear vealed that the bus bar cost of electric-corde would be able to compete eco-weapons, howeser, it is that things can ity from the world's first and only nomically with subsonic aircraft. Yet always get worse. There will always be commercial-scale breeder reactor-the the U.S. aircraft industry and the De-another country or terrorist group much. touted Super. Phenix-is almost partment of Transportation persuaded interested in obtaining a nuclear Iwice the cost of electricity from their the president to go ahead. In his expla- " device."

LwRs. Both the breeder reactor itself nation of his decision to the nation the Our primary responsibility to our and the fuel. reprocessing service which president adopted their principal argu-descendants must therefore be to con-it requires have proven very expensive. ment: "I want the United States to tain or at least slow the spread of this Electricitd de France is resisting pres-continue to lead the world in air dread menace. In this context the pro-sure from the French Atomic Energy transport."

moters of the plutonium econ'omy Commission to make commitments to it appears President Reagan has must be recognized for what they purchase further breeders unless the been persuaded by the breeder advo-are-the typhoid Marys of the nuclear breeder cost can be brought down to cates within the nuclear industry and era.4lt within 25 percent of the cost of pres-surized water reactors. According to a French news report, Novatome, the builder of Super-Phenix, has proposed as a cost saving measure the removal of one of the safety barriers-the con-tainment vessel and the dome-in the next generation of French breeder reac-tors.

The Breeder Reactor Gap As the need for the breeder reactor has faded into the mists of the future and its economics have come to seem in-creasingly doubtful, its advocates have been left uith one last argument. h was made by Wallace Behnke in the fore-word of the 1980 annual report of Project Management Corp., which manages utility interests in the Clinch River breeder reactor. Behnke argued 3