chapter 1 — the future of nuclear power — overview … · chapter 1 — the future of nuclear...

1C h a p t e r 1 — T h e F u t u r e o f N u c l e a r P o w e r — O v e r v i e w a n d C o n c l u s i o n s

The generation of electricity from fossil fuels, notably natural gas and coal, isa major and growing contributor to the emission of carbon dioxide – a green-house gas that contributes significantly to global warming. We share the sci-entific consensus that these emissions must be reduced and believe that theU.S. will eventually join with other nations in the effort to do so.

At least for the next few decades, there are only a few realistic options forreducing carbon dioxide emissions from electricity generation:

increase efficiency in electricity generation and use;

expand use of renewable energy sources such as wind, solar, biomass, andgeothermal;

capture carbon dioxide emissions at fossil-fueled (especially coal) electricgenerating plants and permanently sequester the carbon; and

increase use of nuclear power.

The goal of this interdisciplinary MIT study is not to predict which of theseoptions will prevail or to argue for their comparative advantages. In our view,it is likely that we shall need all of these options and accordingly it would be amistake at this time to exclude any of these four options from an overall carbonemissions management strategy. Rather we seek to explore and evaluate actionsthat could be taken to maintain nuclear power as one of the significantoptions for meeting future world energy needs at low cost and in an environ-mentally acceptable manner.

In 2002, nuclear power supplied 20% of UnitedStates and 17% of world electricity consump-tion. Experts project worldwide electricity con-sumption will increase substantially in the com-ing decades, especially in the developing world,accompanying economic growth and socialprogress. However, official forecasts call for a

mere 5% increase in nuclear electricity generating capacity worldwide by2020 (and even this is questionable), while electricity use could grow by as

CHAPTER 1 — THE FUTURE OF NUCLEAR POWER —OVERVIEW AND CONCLUSIONS

In our view, it would be a mistake

at this time to exclude any of these

four options from an overall carbon

emissions management strategy.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM

2 M I T S T U D Y O N T H E F U T U R E O F N U C L E A R P O W E R

much as 75%. These projections entail little new nuclear plant constructionand reflect both economic considerations and growing anti-nuclear sentimentin key countries. The limited prospects for nuclear power today are attributa-ble, ultimately, to four unresolved problems:

Costs: nuclear power has higher overall lifetime costs compared to natural gaswith combined cycle turbine technology (CCGT) and coal, at least in theabsence of a carbon tax or an equivalent “cap and trade” mechanism forreducing carbon emissions;

Safety: nuclear power has perceived adverse safety, environmental, and healtheffects, heightened by the 1979 Three Mile Island and 1986 Chernobyl reac-tor accidents, but also by accidents at fuel cycle facilities in the UnitedStates, Russia, and Japan. There is also growing concern about the safe andsecure transportation of nuclear materials and the security of nuclear facil-ities from terrorist attack;

Proliferation: nuclear power entails potential security risks, notably the possi-ble misuse of commercial or associated nuclear facilities and operations toacquire technology or materials as a precursor to the acquisition of anuclear weapons capability. Fuel cycles that involve the chemical reprocess-ing of spent fuel to separate weapons-usable plutonium and uraniumenrichment technologies are of special concern, especially as nuclear powerspreads around the world;

Waste: nuclear power has unresolved challenges in long-term management ofradioactive wastes. The United States and other countries have yet to imple-ment final disposition of spent fuel or high level radioactive waste streamscreated at various stages of the nuclear fuel cycle. Since these radioactivewastes present some danger to present and future generations, the publicand its elected representatives, as well as prospective investors in nuclearpower plants, properly expect continuing and substantial progress towardssolution to the waste disposal problem. Successful operation of the planneddisposal facility at Yucca Mountain would ease, but not solve, the wasteissue for the U.S. and other countries if nuclear power expands substantially.

We believe the nuclear option should be

retained, precisely because it is an

important carbon-free source of power.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


Today, nuclear power is not an economically competitive choice. Moreover,unlike other energy technologies, nuclear power requires significant govern-ment involvement because of safety, proliferation, and waste concerns. If inthe future carbon dioxide emissions carry a significant “price,” however,nuclear energy could be an important — indeed vital — option for generat-ing electricity. We do not know whether this will occur. But we believe thenuclear option should be retained, precisely because it is an important carbon-free source of power that can potentially make a significant contribution tofuture electricity supply.

To preserve the nuclear option for the future requires overcoming the fourchallenges described above—costs, safety, proliferation, and wastes. Thesechallenges will escalate if a significant number of new nuclear generatingplants are built in a growing number of countries. The effort to overcomethese challenges, however, is justified only if nuclear power can potentiallycontribute significantly to reducing global warming, which entails majorexpansion of nuclear power. In effect, preserving the nuclear option for thefuture means planning for growth, as well as for a future in which nuclearenergy is a competitive, safer, and more secure source of power.

To explore these issues, our study postulates a global growth scenario that bymid-century would see 1000 to 1500 reactors of 1000 megawatt-electric(MWe) capacity each deployed worldwide, compared to a capacity equivalentto 366 such reactors now in service. Nuclear power expansion on this scalerequires U.S. leadership, continued commitment by Japan,Korea, and Taiwan, a renewal of European activity, andwider deployment of nuclear power around the world. Anillustrative deployment of 1000 reactors, each 1000 MWe insize, under this scenario is given in following table.

This scenario would displace a significant amount of car-bon-emitting fossil fuel generation. In 2002, carbon equiva-lent emission from human activity was about 6,500 milliontonnes per year; these emissions will probably more thandouble by 2050. The 1000 GWe of nuclear power postulatedhere would avoid annually about 800 million tonnes of car-bon equivalent if the electricity generation displaced wasgas-fired and 1,800 million tonnes if the generation wascoal-fired, assuming no capture and sequestration of carbondioxide from combustion sources.

2000PROJECTED 2050

GWe CAPACITY 2050

Total WorldDeveloped world U.S. Europe & Canada Developed East Asia FSUDeveloping world China, India, Pakistan Indonesia, Brazil, Mexico Other developing countries

1,000

625300210115

50325200

7550

17%23%

16%2%

19%29%

23%11%

NUCLEAR ELECTRICITY MARKET SHARE

REGION

Projected capacity comes from the global electricity demand scenario in Appendix 2, which entails growth in global electricity consumption from 13.6 to 38.7 trillion kWhrs from 2000 to 2050 (2.1% annual growth). The market share in 2050 is predicated on 85% capacity factor for nuclear power reactors. Note that China, India, and Pakistan are nuclear weapons capable states. Other developing countries includes as leading contributors Iran, South Africa, Egypt, Thailand, Philippines, and Vietnam.

Global Growth Scenario

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


FUEL CYCLE CHOICES

A critical factor for the future of an expanded nuclear power industry is thechoice of the fuel cycle — what type of fuel is used, what types of reactors“burn” the fuel, and the method of disposal of the spent fuel. This choiceaffects all four key problems that confront nuclear power — costs, safety, pro-liferation risk, and waste disposal. For this study, we examined three represen-tative nuclear fuel cycle deployments:

conventional thermal reactors operating in a “once-through” mode, in which discharged spent fuel is sent direct-ly to disposal;

thermal reactors with reprocessing in a “closed” fuel cycle,which means that waste products are separated from unused

fissionable material that is re-cycled as fuel into reactors. This includes thefuel cycle currently used in some countries in which plutonium is separatedfrom spent fuel, fabricated into a mixed plutonium and uranium oxide fuel,and recycled to reactors for one pass1;

fast reactors2 with reprocessing in a balanced “closed” fuel cycle, which meansthermal reactors operated world-wide in “once-through” mode and a bal-anced number of fast reactors that destroy the actinides separated from ther-mal reactor spent fuel. The fast reactors, reprocessing, and fuel fabricationfacilities would be co-located in secure nuclear energy “parks” in industrialcountries.

Closed fuel cycles extend fuel supplies. The viability of the once-throughalternative in a global growth scenario depends upon the amount of uraniumresource that is available at economically attractive prices. We believe that theworld-wide supply of uranium ore is sufficient to fuel the deployment of 1000reactors over the next half century and to maintain this level of deploymentover a 40 year lifetime of this fleet. This is an important foundation of ourstudy, based upon currently available information and the history of naturalresource supply.

The result of our detailed analysis of the relative merits of these representativefuel cycles with respect to key evaluation criteria can be summarized as fol-lows: The once through cycle has advantages in cost, proliferation, and fuel cyclesafety, and is disadvantageous only in respect to long-term waste disposal; the

We believe that the world-wide supply

of uranium ore is sufficient to fuel the

deployment of 1,000 reactors over the

next half century.

1. This fuel cycle is known asPlutonium Recycle MixedOxide, or PUREX/MOX.

2. A fast reactor more readilybreeds fissionable isotopes-potential fuel-because itutilizes higher energy neu-trons that in turn createmore neutrons whenabsorbed by fertile ele-ments, e.g. fissile Pu239 isbred from neutron absorp-tion of U238 followed bybeta (electron) emissionfrom the nucleus.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


two closed cycles have clear advan-tages only in long-term aspects ofwaste disposal, and disadvantages incost, short-term waste issues, prolifer-ation risk, and fuel cycle safety. (SeeTable.) Cost and waste criteria arelikely to be the most crucial for deter-mining nuclear power’s future.

We have not found, and based oncurrent knowledge do not believe it isrealistic to expect, that there are newreactor and fuel cycle technologiesthat simultaneously overcome theproblems of cost, safety, waste, andproliferation.

Our analysis leads to a significant conclusion: The once-through fuel cycle bestmeets the criteria of low costs and proliferation resistance. Closed fuel cyclesmay have an advantage from the point of view of long-term waste disposaland, if it ever becomes relevant, resource extension. But closed fuel cycles willbe more expensive than once-through cycles, until ore resources become veryscarce. This is unlikely to happen, even with significant growth in nuclearpower, until at least the second half of this century, and probably considerablylater still. Thus our most important recommendation is:

For the next decades, government and industry in the U.S. and elsewhere

should give priority to the deployment of the once-through fuel cycle,

rather than the development of more expensive closed fuel cycle

technology involving reprocessing and new advanced thermal or fast

reactor technologies.

This recommendation implies a major re-ordering of priorities of the U.S.Department of Energy nuclear R&D programs.

Fuel Cycle Types and Ratings

ReactorECONOMICS Fuel Cycle

Once through

Closed thermal

Closed fast

× short term– long term

– short term+ long term

– short term+ long term

SAFETY

+ means relatively advantageous; × means relatively neutral; – means relatively disadvantageous

This table indicates broadly the relative advantage and disadvantage among the different type of nuclear fuel cycles. It does not indicate relative standing with respect to other electricity-generating technologies, where the criteria might be quite different (for example, the nonproliferation criterion applies only to nuclear).

WASTE PROLIFERATION

+

–

–

+

–

–

×

×

+ to –

+

–

–

MIT_ch01_1-16V1.qxd 7/21/2003 2:19 PM


PUBLIC ATTITUDES TOWARD NUCLEAR POWER

Expanded deployment of nuclear power requires public acceptance of thisenergy source. Our review of survey results shows that a majority ofAmericans and Europeans oppose building new nuclear power plants to meetfuture energy needs. To understand why, we surveyed 1350 adults in the USabout their attitudes toward energy in general and nuclear power in particu-lar. Three important and unexpected results emerged from that survey:

The U.S. public’s attitudes are informed almost entirely by their perceptionsof the technology, rather than by politics or by demographics such asincome, education, and gender.

The U.S. public’s views on nuclear waste, safety, and costs are critical to theirjudgments about the future deployment of this technology. Technologicalimprovements that lower costs and improve safety and waste problems canincrease public support substantially.

In the United States, people do not connect concern about global warmingwith carbon-free nuclear power. There is no difference in support for build-ing more nuclear power plants between those who are very concerned aboutglobal warming and those who are not. Public education may help improveunderstanding about the link between global warming, fossil fuel usage, andthe need for low-carbon energy sources.

There are two implications of these findings for our study: first, the U.S. pub-lic is unlikely to support nuclear power expansion without substantialimprovements in costs and technology. Second, the carbon-free character ofnuclear power, the major motivation for our study, does not appear to moti-vate the U.S. general public to prefer expansion of the nuclear option.

The U.S. public is unlikely to support

nuclear power expansion without

substantial improvements in costs and

technology.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


ECONOMICS

Nuclear power will succeed in the long run only if it has a lower cost thancompeting technologies. This is especially true as electricity markets becomeprogressively less subject to economic regulation in many parts of the world.We constructed a model to evaluate the real cost of electricity from nuclearpower versus pulverized coal plants and natural gas combined cycle plants (atvarious projected levels of real lifetime prices for natural gas), over their eco-nomic lives. These technologies are most widely used today and, absent a car-bon tax or its equivalent, are less expensive than manyrenewable technologies. Our “merchant” cost model usesassumptions that commercial investors would be expectedto use today, with parameters based on actual experiencerather than engineering estimates of what might be achievedunder ideal conditions; it compares the constant or “lev-elized” price of electricity over the life of a power plant thatwould be necessary to cover all operating expenses and taxesand provide an acceptable return to investors. The compara-tive figures given below assume 85% capacity factor and a40-year economic life for the nuclear plant, reflect economicconditions in the U.S, and consider a range of projectedimprovements in nuclear cost factors. (See Table.)

We judge the indicated cost improvements for nuclear power to be plausible,but not proven. The model results make clear why electricity produced fromnew nuclear power plants today is not competitive with electricity producedfrom coal or natural gas-fueled CCGT plants with low or moderate gas prices,unless all cost improvements for nuclear power are realized. The cost compar-ison becomes worse for nuclear if the capacity factor falls. It is also importantto emphasize that the nuclear cost structure is driven by high up-front capitalcosts, while the natural gas cost driver is the fuel cost; coal lies in betweennuclear and natural gas with respect to both fuel and capital costs.

Nuclear does become more competitive by comparison ifthe social cost of carbon emissions is internalized, for exam-ple through a carbon tax or an equivalent “cap and trade”system. Under the assumption that the costs of carbonemissions are imposed, the accompanying table illustratesthe impact on the competitive costs for different powersources, for emission costs in the range of $50 to $200/tonnecarbon. (See Table.) The ultimate cost will depend on bothsocietal choices (such as how much carbon dioxide emission

REAL LEVELIZED COSTCents/kWe-hr

Nuclear (LWR) + Reduce construction cost 25% + Reduce construction time 5 to 4 years + Further reduce O&M to 13 mills/kWe-hr + Reduce cost of capital to gas/coalPulverized CoalCCGTa (low gas prices, $3.77/MCF)CCGT (moderate gas prices, $4.42/MCF)CCGT (high gas prices, $6.72/MCF)

6.75.55.35.14.24.23.84.15.6

CASE (Year 2002 $)

a. Gas costs reflect real, levelized acquisition cost per thousand cubic feet (MCF) over the economic life of the project.

Comparative Power Costs

$50/tonne C

CoalGas (low)Gas (moderate)Gas (high)

5.44.34.76.1

CARBON TAX CASES LEVELIZED ELECTRICITY COST cents/kWe-hr $100/tonne C

6.64.85.26.7

$200/tonne C

9.05.96.27.7

Power Costs with Carbon Taxes

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


to permit) and technology developments, such as the cost and feasibility oflarge-scale carbon capture and long-term sequestration. Clearly, costs in therange of $100 to $200/tonne C would significantly affect the relative costcompetitiveness of coal, natural gas, and nuclear electricity generation.

The carbon-free nature of nuclear power argues for government action toencourage maintenance of the nuclear option, particularly in light of the reg-ulatory uncertainties facing the use of nuclear power and the unwillingness ofinvestors to bear the risk of introducing a new generation of nuclear facilitieswith their high capital costs.

We recommend three actions to improve the economic viability of nuclearpower:

The government should cost share for site banking for a number of plants,

certification of new plant designs by the Nuclear Regulatory Commission,

and combined construction and operating licenses for plants built immedi-

ately or in the future; we support U.S. Department of Energy initiatives on

these subjects.

The government should recognize nuclear as carbon-free and include new

nuclear plants as an eligible option in any federal or state mandatory

renewable energy portfolio (i.e., a “carbon-free” portfolio) standard.

The government should provide a modest subsidy for a small set of “first

mover” commercial nuclear plants to demonstrate cost and regulatory fea-

sibility in the form of a production tax credit.

We propose a production tax credit of up to $200 per kWe of the construc-tion cost of up to 10 “first mover” plants. This benefit might be paid out atabout 1.7 cents per kWe-hr, over a year and a half of full-power plant opera-tion. We prefer the production tax credit mechanism because it offers thegreatest incentive for projects to be completed and because it can be extendedto other carbon free electricity technologies, for example renewables, (windcurrently enjoys a 1.7 cents per kWe-hr tax credit for ten years) and coal withcarbon capture and sequestration. The credit of 1.7 cents per kWe- hr isequivalent to a credit of $70 per avoided metric ton of carbon if the electrici-ty were to have come from coal plants (or $160 from natural gas plants). Ofcourse, the carbon emission reduction would then continue without publicassistance for the plant life (perhaps 60 years for nuclear). If no new nuclearplant is built, the government will not pay a subsidy.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


These actions will be effective in stimulating additional investment in nuclear generatingcapacity if, and only if, the industry can live up to its own expectations of being able toreduce considerably capital costs for new plants.

Advanced fuel cycles add considerably to the cost of nuclear electricity. We consideredreprocessing and one-pass fuel recycle with current technology, and found the fuel cost,including waste storage and disposal charges, to be about 4.5 times the fuel cost of theonce-through cycle. Thus use of advanced fuel cycles imposes a significant economicpenalty on nuclear power.

SAFETY

We believe the safety standard for the global growth scenario should maintain today’sstandard of less than one serious release of radioactivity accident for 50 years from allfuel cycle activity. This standard implies a ten-fold reduction in the expected frequencyof serious reactor core accidents, from 10-4/reactor year to 10-5/reactor year. This reactorsafety standard should be possible to achieve in new light water reactor plants that makeuse of advanced safety designs. International adherence to such a standard is important,because an accident in any country will influence public attitudes everywhere. The extentto which nuclear facilities should be hardened to possible terrorist attack has yet to beresolved.

We do not believe there is a nuclear plant design that is totally risk free. In part, this isdue to technical possibilities; in part due to workforce issues. Safe operation requireseffective regulation, a management committed to safety, and a skilled work force.

The high temperature gas-cooled reactor is an interesting candidate for reactor researchand development because there is already some experience with this system, althoughnot all of it is favorable. This reactor design offers safety advantages because the highheat capacity of the core and fuel offers longer response times and precludes excessivetemperatures that might lead to release of fission products; it also has an advantage com-pared to light water reactors in terms of proliferation resistance.

These actions will be effective in

stimulating additional investment in

nuclear generating capacity if, and only

if, the industry can live up to its own

expectations of being able to reduce

considerably overnight capital costs

for new plants.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


Because of the accidents at Three Mile Island in 1979 and Chernobyl in 1986,a great deal of attention has focused on reactor safety. However, the safetyrecord of reprocessing plants is not good, and there has been little safetyanalysis of fuel cycle facilities using, for example, the probabilistic risk assess-ment method. More work is needed here.

Our principal recommendation on safety is:

The government should, as part of its near-term R&D program, develop

more fully the capabilities to analyze life-cycle health and safety impacts

of fuel cycle facilities and focus reactor development on options that can

achieve enhanced safety standards and are deployable within a couple of

decades.

WASTE MANAGEMENT

The management and disposal of high-level radioactive spent fuel from thenuclear fuel cycle is one of the most intractable problems facing the nuclearpower industry throughout the world. No country has yet successfully imple-mented a system for disposing of this waste. We concur with the many inde-pendent expert reviews that have concluded that geologic repositories will becapable of safely isolating the waste from the biosphere. However, implemen-tation of this method is a highly demanding task that will place great stresson operating, regulatory, and political institutions.

For fifteen years the U.S. high-level waste managementprogram has focused almost exclusively on the proposedrepository site at Yucca Mountain in Nevada. Althoughthe successful commissioning of the Yucca Mountainrepository would be a significant step towards the securedisposal of nuclear waste, we believe that a broader,strategically balanced nuclear waste program is needed toprepare the way for a possible major expansion of thenuclear power sector in the U.S. and overseas.

The global growth scenario, based on the once-through fuel cycle, wouldrequire multiple disposal facilities by the year 2050. To dispose of the spentfuel from a steady state deployment of one thousand 1 GWe reactors of thelight water type, new repository capacity equal to the nominal storage capaci-ty of Yucca Mountain would have to be created somewhere in the world everythree to four years. This requirement, along with the desire to reduce long-term risks from the waste, prompts interest in advanced, closed fuel cycles.

We do not believe a convincing case can

be made, on the basis of waste manage-

ment considerations alone, that the

benefits of advanced, closed fuel cycles

will outweigh the attendant safety,

environmental, and security risks and

economic costs.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


These schemes would separate or partition plutonium and other actinides —and possibly certain fission products — from the spent fuel and transmutethem into shorter-lived and more benign species. The goals would be toreduce the thermal load from radioactive decay of the waste on the reposito-ry, thereby increasing its storage capacity, and to shorten the time for whichthe waste must be isolated from the biosphere.

We have analyzed the waste management implications of both once-throughand closed fuel cycles, taking into account each stage of the fuel cycle and therisks of radiation exposure in both the short and long-term. We do not believethat a convincing case can be made on the basis of waste management considera-tions alone that the benefits of partitioning and transmutation will outweigh theattendant safety, environmental, and security risks and economic costs. Futuretechnology developments could change the balance of expected costs, risks,and benefits. For our fundamental conclusion to change, however, not onlywould the expected long term risks from geologic repositories have to be sig-nificantly higher than those indicated in current assessments, but the incre-mental costs and short-term safety and environmental risks would have to begreatly reduced relative to current expectations and experience.

We further conclude that waste management strategies in the once-throughfuel cycle are potentially available that could yield long-term risk reductionsat least as great as those claimed for waste partitioning and transmutation,with fewer short-term risks and lower development and deployment costs.These include both incremental improvements to the current mainstreammined repositories approach and more far-reaching innovations such as deepborehole disposal. Finally, replacing the current ad hoc approach to spent fuelstorage at reactor sites with an explicit strategy to store spent fuel for a periodof several decades will create additional flexibility in the waste managementsystem.

Our principal recommendations on waste management are:

The DOE should augment its current focus on Yucca Mountain with a

balanced long-term waste management R&D program.

A research program should be launched to determine the viability of

geologic disposal in deep boreholes within a decade.

A network of centralized facilities for storing spent fuel for several decades

should be established in the U.S. and internationally.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


NONPROLIFERATION

Nuclear power should not expand unless the risk of proliferation from opera-tion of the commercial nuclear fuel cycle is made acceptably small. We believethat nuclear power can expand as envisioned in our global growth scenariowith acceptable incremental proliferation risk, provided that reasonable safe-guards are adopted and that deployment of reprocessing and enrichment arerestricted. The international community must prevent the acquisition ofweapons-usable material, either by diversion (in the case of plutonium) or bymisuse of fuel cycle facilities (including related facilities, such as researchreactors or hot cells). Responsible governments must control, to the extentpossible, the know-how relevant to produce and process either highlyenriched uranium (enrichment technology) or plutonium.

Three issues are of particular concern: existing stocks of separated plutoniumaround the world that are directly usable for weapons; nuclear facilities, for

example in Russia, with inadequate controls; and transferof technology, especially enrichment and reprocessingtechnology, that brings nations closer to a nuclearweapons capability. The proliferation risk of the globalgrowth scenario is underlined by the likelihood that useof nuclear power would be introduced and expanded inmany countries in different security circumstances.

An international response is required to reduce the proliferation risk. Theresponse should:

re-appraise and strengthen the institutional underpinnings of the IAEA safe-guards regime in the near term, including sanctions;

guide nuclear fuel cycle development in ways that reinforce shared nonpro-liferation objectives.

Nuclear power should not expand unless

the risk of proliferation from operation of

the commercial nuclear fuel cycle is made

acceptably small.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


Accordingly, we recommend:

The International Atomic Energy Agency (IAEA) should focus overwhelmingly

on its safeguards function and should be given the authority to carry out

inspections beyond declared facilities to suspected illicit facilities;

Greater attention must be given to the proliferation risks at the front end

of the fuel cycle from enrichment technologies;

IAEA safeguards should move to an approach based on continuous materi-

als protection, control and accounting using surveillance and containment

systems, both in facilities and during transportation, and should implement

safeguards in a risk-based framework keyed to fuel cycle activity;

Fuel cycle analysis, research, development, and demonstration efforts

must include explicit analysis of proliferation risks and measures defined

to minimize proliferation risks;

International spent fuel storage has significant nonproliferation benefits

for the growth scenario and should be negotiated promptly and implemented

over the next decade.

ANALYSIS, RESEARCH, DEVELOPMENT, AND DEMONSTRATION PROGRAM

The U.S. Department of Energy (DOE) analysis, research, development, anddemonstration (ARD&D) program should support the technology path lead-ing to the global growth scenario and include diverse activities that balancerisk and time scales, in pursuit of the strategic objective of preserving thenuclear option. For technical, economic, safety, and public acceptance reasons,the highest priority in fuel cycle ARD&D, deserving first call on available funds,lies with efforts that enable robust deployment of the once-through fuel cycle.The current DOE program does not have this focus.

Every industry in the United States develops basic analytical models and toolssuch as spreadsheets that allow firms, investors, policy makers, and regulatorsto understand how changes in the parameters of a process will affect the per-formance and cost of that process. But we have been struck throughout ourstudy by the absence of such models and simulation tools that permit in-depth, quantitative analysis of trade-offs between different reactor and fuel

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


cycle choices, with respect to all key criteria. The analysis we have seen isbased on point designs and does not incorporate information about the costand performance of operating commercial nuclear facilities. Such modelingand analysis under a wide variety of scenarios, for both open and closed fuelcycles, will be useful to the industry and investors, as well as to internationaldiscussions about the desirability about different fuel cycle paths.

We call on the Department of Energy, perhaps in collaboration with other coun-tries, to establish a major project for the modeling, analysis, and simulation ofcommercial nuclear power systems — The Nuclear System Modeling Project.

This project should provide a foundation for the accu-mulation of information about how variations in theoperation of plants and other parts of the fuel cycleaffect costs, safety, waste, and proliferation resistancecharacteristics. The models and analysis should be basedon real engineering data and, wherever possible, practicalexperience. This project is technically demanding andwill require many years and considerable resources to becarried out successfully.

We believe that development of advanced nuclear technologies — either fastreactors or advanced fuel cycles employing reprocessing – should await theresults of the Nuclear System Modeling Project we have proposed above. Ouranalysis makes clear that there is ample time for the project to compile thenecessary engineering and economic analyses and data before undertakingexpensive development programs, even if the project should take a decade tocomplete. Expensive programs that plan for the development or deploymentof commercial reprocessing based on any existing advanced fuel cycle tech-nologies are simply not justified on the basis of cost, or the unproven safety,proliferation risk, and waste properties of a closed cycle compared to theonce-through cycle. Reactor concept evaluation should be part of the NuclearSystem Modeling Project.

On the other hand, we support a modest laboratory scale research and analy-sis program on new separation methods and associated fuel forms, with theobjective of learning about approaches that emphasize lower cost and moreproliferation resistance. These data can be important inputs to advanced fuelcycle analysis and simulation and thus help prioritize future developmentprograms.

The modeling project’s research and analysis effort should only encompasstechnology pathways that do not produce weapons-usable material duringnormal operation (for example, by leaving some uranium, fission products,

For technical, economic, safety, and

public acceptance reasons, the highest

priority in fuel cycle R&D, deserving

first call on available funds, lies with

efforts that enable robust deployment

of the once-through fuel cycle.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


and/or minor actinides with the recycled plutonium). The closed fuel cycle cur-rently practiced in Western Europe and Japan, known as PUREX/MOX, does notmeet this criterion. There are advanced closed fuel cycle concepts involvingcombinations of reactor, fuel form, and separations technology that satisfythese conditions and, with appropriate institutional arrangements, can havesignificantly better proliferation resistance than the PUREX/MOX fuel cycle,and perhaps approach that of the open fuel cycle. Accordingly, the govern-ments of nuclear supplier countries should discourage other nations fromdeveloping and deploying the PUREX/MOX fuel cycle.

Government R&D support for advanced design LWRs and for the HighTemperature Gas Reactor (HTGR) is justified because these are the two reactortypes that are most likely to play a role in any nuclear expansion. R&D supportfor advanced design LWRs should focus on measures that reduce constructionand operating cost. Because the High Temperature Gas Reactor (HTGR) haspotential advantages with respect to safety, proliferation resistance, modularityand efficiency, government research and limited development support toresolve key uncertainties, for example, the performance of HTGR fuel forms inreactors and gas power conversion cycle components, is warranted.

Waste management also calls for a significant, and redirected, ARD&D pro-gram. The DOE waste program, understandably, has been singularly focusedfor the past several years on the Yucca Mountain project. We believe DOEmust broaden its waste R&D effort or run the risk of being unable to rigor-ously defend its choices for waste disposal sites. More attention needs to begiven to the characterization of waste forms and engi-neered barriers, followed by development and testing ofengineered barrier systems. We believe deep boreholes, asan alternative to mined repositories, should be aggres-sively pursued. These issues are inherently of interna-tional interest in the growth scenario and should be pur-sued in such a context.

There is opportunity for international cooperation in this ARD&D programon safety, waste, and the Nuclear System Modeling Project. A particularly per-tinent effort is the development, deployment, and operation of a word widematerials protection, control, and accounting tracking system. There is nocurrently suitable international organization for this development task. A pos-sible approach lies with the G-8 as a guiding body.

Our global growth scenario envisions an open fuel cycle architecture at leastuntil mid-century or so, with the advanced closed fuel cycles possiblydeployed later, but only if significant improvements are realized through

The closed fuel cycle currently practiced

in Western Europe and Japan, known

as PUREX/MOX, does not meet this

nonproliferation criterion.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM


research. The principal driver of this conclusion is our judgment that natural uraniumore is available at reasonable prices to support the open cycle at least to late in the centu-ry in a scenario of substantial expansion. This gives the open cycle clear economicadvantage with proliferation resistance an important additional feature. The DOE shouldundertake a global uranium resource evaluation program to determine with greater con-fidence the uranium resource base around the world.

Accordingly, we recommend:

The U.S. Department of Energy should focus its R&D program on the once-through fuel

cycle;

The U.S. Department of Energy should establish a Nuclear System Modeling project to

carryout the analysis, research, simulation, and collection of engineering data needed

to evaluate all fuel cycles from the viewpoint of cost, safety, waste management, and

proliferation resistance;

The U.S. Department of Energy should undertake an international uranium resource

evaluation program;

The U.S. Department of Energy should broaden its waste management R&D program;

The U.S. Department of Energy should support R&D that reduces Light Water Reactor

(LWR) costs and for development of the HTGR for electricity application.

We believe that the ARD&D program proposed here is aligned with the strategic objec-tive of enabling a credible growth scenario over the next several decades. Such a ARD&Dprogram requires incremental budgets of almost $400 million per year over the next 5years, and at least $460 million per year for the 5-10 year period.

MIT_ch01_1-16.qxd 7/16/2003 1:09 PM

17C h a p t e r 2 — B a c k g r o u n d a n d P u r p o s e o f T h i s S t u d y

In 2000 nuclear power produced about 17% ofthe world’s electricity from 442 commercialreactors in 31 countries. The United States hasthe largest deployment, with 104 operatingreactors producing 20% of the country’s elec-tricity, followed by France, Japan, Germany,Russia, and South Korea. The reliability of theseplants has improved considerably in recentyears (for example, capacity factors of U.S.nuclear reactors have achieved 90%), and manywill have their originally expected operatinglives extended significantly. Nuclear power isclearly an important source of electricity in theUnited States and the world.

If current policies continue, however, nuclearpower is likely to decline gradually and conceiv-ably disappear in this century from the world’selectricity supply portfolio. We believe remov-ing nuclear power as a supply option would bea mistake at this time. The primary reason isthat nuclear power is an important source ofelectricity that does not rely on fossil fuel andhence does not produce greenhouse gas emis-sions. This is the primary motivation for ourexamination for an inter-connected set of issuesthat will challenge nations individually and col-lectively over the next century. The issues are:

reducing atmospheric pollution and emis-sions of greenhouse gases;

meeting dramatically increased energy, andespecially electricity, demand throughout theindustrialized and developing world; and

assuring security and minimizing conflictassociated with energy supply.

Our study undertakes to:

describe the characteristics of a nuclearpower infrastructure that would make a sig-

nificant contribution to reducing CO2 emis-sions;

identify the issues that must be addressed ifnuclear power is to make a contribution onthis scale; and

outline the needed program of analysis,research, development, and demonstration.

GLOBAL WARMING

Most developed countries are in the early stagesof implementing policies to stabilize and ulti-mately reduce greenhouse gas emissions andthe attendant global warming. The scientificconsensus about the risks of further significantincreases in atmospheric greenhouse gas con-centrations grows steadily stronger and morewidely endorsed. This consensus underlies astrong impetus for governmental actions thatprepare the ground for meeting possibly strin-gent CO2 emission constraints in the decadesahead, specifically global emission levels com-parable to or below those of today, despite aconsiderable increase in energy production anduse. Developing countries will need to limit thegrowth of greenhouse gas emissions while theirenergy consumption increases dramatically. Forexample, if atmospheric concentration of CO2

is not allowed to exceed twice its pre-industrialvalue, then CO2 emissions in the 21st centurywill need to be held to half the cumulative totalexpected under a “business as usual” trajectory,1

and the annual emission rate would eventuallyneed to fall well below the 2000 value. Whileour focus is on global warming because of itsoverwhelming international implications, werecognize that reduction in other emissionsfrom fossil fuel combustion would have impor-tant regional and local benefits for clean air.

Chapter 2 — Background and Purpose of This Study

MIT_ch02_17-24.qxd 7/16/2003 1:21 PM


strategy. Each of the options presents technical,economic, environmental, political, and humanbehavioral issues that make their ultimate mar-ket penetration uncertain.

Nuclear power is a special case, however. If cur-rent trends continue, nuclear power will gradu-ally decrease and perhaps even disappear aspart of the global energy portfolio, thus failingto make any long-term contribution to reduc-ing greenhouse gas emissions. Few nuclearpower plants are under construction world-wide, and of those, most are being built in asmall number of developing countries or devel-oped countries in East Asia.2 In most developedcountries, the use of nuclear power is notexpected to expand and, in many of these coun-tries, including the United States, nuclear powerhas been explicitly excluded from policies tostabilize and reduce carbon emissions (e.g.,direct and tax subsidies for renewable energyand energy conservation, high mandated pur-chase prices for renewable energy, renewableenergy portfolio standards). In Britain, nuclearpower plants pay a “carbon tax,” even thoughthey have essentially no CO2 emissions. Webelieve that a more objective approach will havea better chance at meeting the global warmingchallenge. Indeed, it is likely that our energyfuture will exploit all of the four options to onedegree or another. This study addresses theissues associated with maintaining the nuclearpower option.

ELECTRICITY DEMAND

The U.S. National Academy of Engineeringnamed electrification as the premier engineer-ing achievement of the twentieth century3. Thisis a remarkable statement for the century oflasers, computers, airplanes, and other ubiqui-tous and important technologies and is indica-tive of the extraordinary impact of electricity inimproving the quality of people’s lives.Accordingly, it should not be surprising thatglobal electricity use is expected to increasedramatically in the years ahead, even takinginto account improvements in end use efficien-cy. Growth in electricity use is expected espe-cially in developing countries, as they strive tomeet basic needs and to modernize and indus-trialize their economies.

We believe that the United States will eventual-ly join with other developed countries in theeffort to reduce greenhouse gas emissions, evenif the mechanisms for doing so are uncertainfor the moment. Developing countries – cer-tainly the large ones, such as China, India,Pakistan, Brazil, and Indonesia – must ulti-mately be party to this effort if it is to succeed.Achieving the reductions in greenhouse gasemissions likely to be required will be a majortechnical and economic challenge to bothdeveloped and developing countries that willpersist for many decades into the future.

The power sector contributes about a third ofgreenhouse gas emissions worldwide. TheEnergy Information Administration (EIA) ofthe U.S. Department of Energy projects that, inthe absence of CO2-control policies and tech-nologies, electricity’s share of global emissionsof greenhouse gases (CO2 and others) willclimb to over 40% by 2020. In the United States,almost 90% of the carbon emissions from elec-tricity generation come from coal-fired genera-tion, even though this accounts for only 52% ofthe electricity. (About 29% of United Stateselectricity comes from carbon-free nuclear andrenewables-based generation; about 19%comes from natural-gas-fired and oil-fired gen-eration, but both of these fuels release less car-bon per kilowatt-hour than coal-fired genera-tion does.)

There are few realistic options to reduce signif-icantly carbon emissions from electricity gener-ation (besides lowering standards of living):

increased efficiency in electricity end-useand generation;

increased use of renewable energy technolo-gies (e.g., wind, solar, biomass, and geother-mal);

introduction of carbon capture and seques-tration at fossil-fueled (especially coal)power plants on a massive scale; and

increased use of nuclear fission power reac-tors (and possibly fusion at a later date).

As we have argued in Chapter 1, our view is thatit would be a mistake to exclude at this time anyof these four basic options as a possibly importantpart of an overall carbon emissions management

MIT_ch02_17-24.qxd 7/16/2003 1:21 PM


The U.S. Department of Energy’s EIA projects a75% increase in global electricity use in twodecades, from 2000 to 2020. By mid-century, athreefold increase or more is credible and,indeed, expected. Table 2.1 gives the growth ratefor electricity use in different regions of theworld as anticipated in the EIA “business-as-usual” projections to the year 2020.4

There is a strong correlation between electricityconsumption per capita and the United Nations“human development index” (HDI), whichcombines indicators of health, education, andeconomic prosperity.5 Industrialized countrieshave an HDI above 0.9 (on a scale of 0 to 1) andper capita energy consumption above 4000kWe-hrs.

Large developing countries, such as China,India, Pakistan, and Indonesia, are well belowthe industrialized country HDI and aspire toadvance by rapid economic growth. Overall,energy consumption per capita in the develop-ing world is currently less than a fifth of that inthe developed world. Unless provided withassistance or incentives, these developingnations are likely to seek the lowest cost supplyalternatives that can meet their growing indus-trial and consumer demand for electricity. Thisprospect clearly raises the specter of substantial-ly increased greenhouse gas emissions, sincecoal is likely to be an economic choice for manydeveloping countries, e.g. China and India. Howthese developing countries meet their electricitydemand is of central interest to the discussion ofglobal warming, since over time their choices willinfluence global emissions levels more than meas-ures taken by the developed world. Greater elec-tricity consumption is desirable because itaccompanies social and economic advance, butwe want the electricity production to take placein an economic and environmentally acceptablemanner.

The attractiveness of nuclear power as anoption will be determined by many country-specific factors. To understand how muchnuclear power would be needed to make a sig-nificant contribution to reducing CO2 emis-sions by 2050, and where it might be deployed,we present, in Appendix 2, a simple scenario forelectricity growth over the next fifty years. Thescenario is not based on economic forecasting,

but on a model of what electricity growth couldbe as countries attempt to raise individual livingstandards to acceptable levels within crediblegrowth constraints. The model assumes a mod-est 1%/year annual growth in per capita elec-tricity consumption for developed countriesand a growth rate for developing countries thattakes them to 4000 kWe-hrs/person/year in2050 (i.e., we determine the growth rate as anoutcome). Population projections are thosecurrently provided by the United Nations. Theone additional constraint in the scenario is thatthe annual growth rate in total electricity pro-duction for any country is capped at 4.7%; thisis one half percent above EIA’s projected elec-tricity growth rate for the developing worldoverall up to 2020. Sustaining a 4.7%/yeargrowth rate for fifty years yields a factor of tenincrease; although within the realm of possibil-ity with appropriate policies and sufficientresource investment, this cap on total growthrepresents a very ambitious target for any indi-vidual developing country. Within this scenario,global electricity production is slightly belowthe EIA reference in 2020 and about a factor ofthree greater in 2050 than it is today. The impli-cations of this scenario for four categories ofnations are described below.

Developed countries. Among the major devel-oped countries, the United States is unique inhaving a projected large increase in populationand a concomitant large increase in total elec-tricity demand. If the global deployment ofnuclear power is to grow substantially by mid-century, the United States almost certainly mustbe a major participant. Nuclear power growth isunlikely to be very large in other key developedcountries, such as Japan (with an anticipatedpopulation decline) or France (with a stablepopulation and a power sector already domi-nated by nuclear power).

Table 2.1 Anticipated Growth of Electricity (billion kWe-h)4

1999 2020 GROWTH RATE %

Industrialized(US)FSUDeveloping

7,5003,2001,5003,900

10,9004,8002,1009,200

1.81.91.84.2

Total World 12,800 22,200 2.7

REGION(billion kWe-h)

MIT_ch02_17-24.qxd 7/16/2003 1:21 PM


More advanced developing countries.Countries such as China, Brazil, Mexico, andIran can reach the 4000 kWe-hrs/person/yearbenchmark with annual growth rates of elec-tricity consumption in the 2%-3% range.Although improved business, regulatory, finan-cial, political, and other conditions may beneeded, these countries would likely be veryimportant for an expanded nuclear power sce-nario. By 2050, they will have large urban pop-ulations (above 85%), an important factorfavoring the introduction of large base loadplants. This model is, of course, subject to coun-try-specific caveats; for example, Iran has abun-dant natural gas supplies, so its pursuit ofnuclear power logically raises proliferation con-cerns. Collectively, countries in this group haverelatively little nuclear power today but couldturn to nuclear power to meet a fraction of theirfuture electricity supply needs, as South Koreahas done.

Less advanced developing countries. Countriessuch as India, Pakistan, Indonesia, Philippines,and Vietnam (with a combined projected pop-ulation of 2.5 billion in 2050) may, with consid-erable progress in their political, legal, financial,and regulatory regimes and an associatedincrease in domestic and foreign investment intheir energy sectors, reach 2000-3000 kWe-hrs/person/year by mid-century. This will be atall order. Nuclear power may account for partof the dramatic increase in electricity supplycalled for in these countries (India is an excep-tion in that it already has fourteen units), butpursuing such a capital- and management-intensive technology will prove challenging. Inmany cases, proliferation concern – the concernthat the commercial nuclear fuel cycle will beused as a source of materials and/or technologythat will lead to proliferation of nuclearweapons – will accompany development of sub-stantial nuclear technology infrastructures.

Least advanced developing countries. Manylarge developing countries, with a particularconcentration in Africa, cannot come close tothe per capita benchmark within economicallycredible scenarios. These countries are not goodcandidates for nuclear power, barring anunforeseen breakthrough in technology andcapital requirements.

In sum, electricity utilization is likely to increasesignificantly worldwide over the next half-centu-ry, requiring a major investment in bothreplacement and expansion of generatingcapacity. Much of the expansion will take placein the developing world. Selected developedcountries will be central to a major increase innuclear power, but large parts of the developingworld are unlikely participants. If developingnations do adopt nuclear power, all nations ofthe world will have an interest in how thesecountries regulate their nuclear enterprise withrespect to reactor and fuel cycle safety, trans-portation of nuclear materials, waste disposal,and especially proliferation safeguards.

SECURITY

Yet another reason for thinking about thenuclear option — national security — is notnew. The dependence of the developed worldon oil from the Middle East, an unstable regionof the world, has long presented a risk to theeconomies of the United States and other coun-tries that depend on imported oil, such asJapan, Germany, and France. The United States’dependence is linked principally to fuel for thetransportation sector, but many other countriesrely on oil for significant power generation.Nuclear power offers one option for reducingthis dependence.

Within the time horizon addressed in this study,however, the national security implications ofexpanded nuclear power may be even more sig-nificant with respect to natural gas, which dis-plays the same lack of geographic correlationbetween supply and demand that has definedthe geopolitical landscape for oil. It is likely thatmany nations, including the United States, mayimport large quantities of LNG or liquids fromgas, produced from stranded gas in diverseregions of the world.

There is another national security dimension tonuclear power. Combating nuclear proliferationis one of our most important foreign policyobjectives. There is no doubt about the greatrisk to the security of the United States and therest of the world that the spread of nuclearweapons to other states and perhaps non-stateactors would bring. So there is a major security

MIT_ch02_17-24.qxd 7/16/2003 1:21 PM


interest in how all aspects of nuclear commercedevelop around the world. For example, theextensive U.S. “Cooperative Threat Reductionprogram,”6 provides assistance to Russia for thepurpose of improving their efforts to protecttheir nuclear weapons and nuclear explosivematerials against theft.7 On the other hand,there is considerable tension between theUnited States and Russia created by Russianassistance to Iran on commercial nuclear power,especially since Iran is awash in natural gas.

Indeed, it is worth recalling that the unresolvednuclear fuel cycle “schism” of the 1970s betweenthe United States and its European and Japaneseallies stemmed from nonproliferation concerns.In the Ford and Carter administrations, theUnited States stopped the recycling of plutoni-um in commercial reactors because of prolifer-ation risks associated with a “plutonium econo-my.” The hope that others would emulate thispolicy was not realized, as energy resource-poorcountries, such as France and Japan, evaluatedthe balance of risks differently. As countrieslook to shape today’s nuclear fuel cycle policyand R&D decisions in the context of the worldenvironmental, economic development, andsecurity needs of the next fifty years, finding acommon path among the G-8 and others canitself contribute significantly to managing pro-liferation concerns. The expansion of nuclearpower, should it occur, will raise proliferationconcerns that call for ongoing Americanengagement in nuclear fuel cycle issues inde-pendent of nuclear power’s level of contribu-tion to domestic electricity generation.

THE CHALLENGES OF NUCLEAR POWEREXPANSION

Despite the strong rationale for reducing green-house gas emissions that contribute to globalwarming, for meeting increasing demand forelectricity, and for improving the national secu-rity aspects of energy supply, the EIA’s “busi-ness-as-usual” projection for nuclear powerindicates a mere 5% increase in 2020, even asworld electricity use increases by 75%. After2020, if significant investments are not made,nuclear power supply would decline as existingreactors are retired. EIA projects significantincreases in nuclear generated electricity in

China, Japan, and South Korea, largely offset-ting decreases in the United States and WesternEurope. In the United States, the last nuclearplant order was in 1979. There is considerableanti-nuclear sentiment in Europe: Belgium,Germany, the Netherlands, and Sweden are offi-cially committed to phasing out nuclear powergradually; and there is public opposition tonuclear power in Japan and Taiwan. To be sure,several countries are still on a path to constructnew operating units — South Korea, Finland,India, and Russia are examples — and Chinamay yet commit to substantial new nuclearplant construction.

There are several reasons why nuclear powerhas not met the expectations for capacitygrowth projected several decades ago. One fac-tor is that the public perception of nuclear ener-gy is unfavorable, in part due to concern abouteffects of radiation that the public associateswith nuclear energy. More importantly, theadverse impression derives from real andunique problems presented by this technology.These problems are:

Unfavorable economics. Most operatingnuclear plants are economical to operate whencosts going forward are considered, i.e. whensunk capital and construction costs are ignored.However, new plants appear to be more expen-sive than alternate sources of base load genera-tion, notably coal and natural gas fired electric-ity generation, when both capital and operatingcosts are taken into account.

Coal plants have capital costs intermediatebetween those of gas and nuclear. Even with SO2

and NOx controls that meet U.S. new sourceperformance standards, new coal plants arewidely perceived to be less costly than nuclearplants. However, if CO2 emissions were in thefuture to become subject to control and a signif-icant “price” placed on emissions, the relativeeconomics could become much more favorableto nuclear power.

Perceived adverse safety, environmental, andhealth effects. After the 1979 accident at ThreeMile Island in Harrisburg, Pennsylvania and the1986 accident at Chernobyl in the Soviet Union,public concern about reactor safety increasedsubstantially. The 1999 accident at the Tokai-

MIT_ch02_17-24.qxd 7/16/2003 1:21 PM


Mura plant underscored safety concerns aboutthe nuclear fuel cycle outside of the reactor.There is also concern about transportation ofnuclear materials, and waste management. TheSeptember 11, 2001 terrorist attack on theWorld Trade Center and the Pentagon haveheightened concerns about the vulnerability ofnuclear power stations and other facilities, espe-cially spent fuel storage pools, to terroristattack. There is concern about radiation expo-sure of citizens and workers from activities ofthe industry despite good regulation and healthrecords. There are significant environmentalimpacts, ranging from long-term waste dispos-al to the handling and disposal of toxic chemi-cal wastes associated with the nuclear fuel cycle.

Proliferation. The possibility exists that nationswishing to acquire or enhance a nuclearweapons capability will use commercial nuclearpower as a source of technological know-howor nuclear weapons usable material, notablyplutonium. Although this has not proved to bethe preferred pathway to nuclear weapons capa-bility, the possession of a complete nuclear fuelcycle, including enrichment, fuel fabrication,reactor operation, and reprocessing, certainlymoves any nation closer to obtaining such acapability. The key step for achieving nuclearweapons capability is acquisition of sufficientweapons-usable fissionable material, eitherhigh-enriched uranium or plutonium.Unfortunately, reprocessing of spent fuel for thefuel cycle operation in Europe, Russia, andJapan has led to the accumulation of about 200tonnes of separated plutonium. The associatedrisks have been viewed with increased alarmsince the 9/11 events that demonstrated thereach of international terrorism. Radiationexposure from spent fuel that is not reprocessedis a strong, but not certain, barrier to theft andmisuse.

Difficulty of waste management. There aremany radioactive waste streams created in vari-ous parts of the nuclear fuel cycle. Whatdeservedly receives the most attention is thehigh level waste containing the fission productsand/or transuranic (TRU) elements createdduring energy generation. The spent fuel fromnuclear reactors contains radioactive materialthat presents health and environmental risksthat persist for tens of thousands of years. At

present, no nation has successfully demonstrat-ed a disposal system for these nuclear wastes.On the other hand, Finland has decided on apath to manage spent fuel, and the United Stateshas decided to proceed with licensing of YuccaMountain as a geological repository. At thesame time, many of the discussions surround-ing alternative reactors and fuel cycles are moti-vated by a desire to reduce high-level wastemanagement challenges.

The potential impact on the public from safetyor waste management failure and the link tonuclear explosives technology are unique tonuclear energy among energy supply options.These characteristics and the fact that nuclear ismore costly, make it impossible today to make acredible case for the immediate expanded use ofnuclear power.

Inevitably, there will be a high degree of govern-ment involvement in nuclear power, even inmarket economies, to regulate safety, waste, andproliferation risk. This is, in itself, another chal-lenge for nuclear power. There is considerablevariation in how different countries approachthe issues of safety, proliferation, and wastemanagement. This often complicates the role ofgovernments in setting international rules –especially for preventing proliferation, but alsofor safety and waste management – that servecommon interests. Poor safeguarding of nuclearmaterials or facilities in any nation could resultin acquisition of nuclear explosives by a roguestate or terrorist group for use in anothernation. The Chernobyl accident demonstratedthe potential for radioactivity to spread acrossborders and thus the importance of uniformlyhigh safety standards and advanced safety tech-nologies (such as western reactor containmentdesigns).

Nuclear power’s value as a carbon-free electric-ity supply technology has also generally notbeen recognized in government policies.Government policies have focused on targetingrenewable energy resources and end-use effi-ciency improvements through a combination ofdirect subsidies, tax subsidies, renewable energyportfolio standards, appliance efficiency stan-dards, and other “second best” mechanisms topromote carbon-free supply technologies andto reduce electricity demand. Nuclear power

MIT_ch02_17-24.qxd 7/16/2003 1:21 PM


has generally been excluded from these pro-grams. While the European Union will intro-duce a carbon dioxide emissions trading systemin a few years, countries have not yet turned tobroad policies to internalize the social costs ofcarbon emissions that would provide incentivesfor investment in all carbon free electricity sup-ply or energy efficiency technologies, includingnuclear power. Thus nuclear power does notcompete on a level playing field and, from thisperspective, is presently being discriminatedagainst in policies designed to respond to thechallenge of reducing carbon dioxide emissions.

Given these difficulties, it is fair to ask whethernuclear energy can ever recapture its attractive-ness as a major energy supply option. However,this is not the question we seek to address. Theanswer to such a question necessarily dependson how societies and technology evolve (eco-nomic growth, electricity demand, fuel prices,environmental constraints, premium attachedto energy security, the cost of alternatives suchas renewables, new technologies such as fusion).

The difficulties facing nuclear power shouldnot, at this time, rule it out as one of a smallnumber of options that may be attractive toexercise in the future, as countries developresponses to the energy and environmentalchallenges of this century. We believe that it isimportant for governments to adopt policies thatenable the full range of significant options avail-able. Nuclear is one of those options. Whether itis an option that will eventually be exercised willdepend on many unknown contingencies.

Given the difficulties that confront nuclearpower, the effort required to overcome them isjustified only if nuclear power potentially canmake a significant impact on the major chal-lenges of global warming, electric supply, andsecurity. That is, for nuclear power to meritstrategic focus and sustaining actions on thepart of government , there must also be a com-mitment to significant expansion of nuclearpower that will sustain and perhaps modestlyincrease its share of global electricity genera-tion, even as use of electricity multiplies.

NOTES

1. T.M.L. Wigley,“Stabilization of greenhouse gas concentra-tions,” in U.S. policy on climate change: What next,?Aspen Institute, Washington D.C., 2002.

2. In 2003, five new units are expected to come into opera-tion: one in the Czech Republic, two in China, and two inSouth Korea. An additional 18 plants are under con-struction worldwide, primarily in China, Taiwan, India,Japan, and South Korea.

3. National Academy of Engineering websitehttp://www.greatachievments.org/

4. U.S. Department of Energy, Energy InformationAdministration (EIS) International Energy Outlook 2002.

5. S. G. Benka,“The Energy Challenges,” Physics Today (April2002) p. 38.

6. This is the Nunn-Lugar-Domenici program. See “ DOE’snon-proliferation programs with Russia,” Co- chairsHoward Baker and Lloyd Cutler, January 10, 2001, TheSecretary of Energy Advisory Board, U.S. DOE.

7. For a recent report card, see M. Bunn, A. Wier, and J.P.Holdren,“Controlling nuclear warheads and materials,”Nuclear Threat Initiative, Washington, D.C., March 2003.

MIT_ch02_17-24.qxd 7/16/2003 1:21 PM

MIT_ch02_17-24.qxd 7/16/2003 1:21 PM

25C h a p t e r 3 — O u t l i n e o f t h e S t u d y

Our study makes two assumptions: First, as dis-cussed in Chapter 1, that nuclear energy is animportant energy supply option for the future,but that exercising the option for significantdeployment requires that the four significantchallenges — cost, safety, waste, and prolifera-tion — must be addressed and overcome. Second,as discussed in Chapter 2, that the public andprivate sectors can justify devoting the resourcesnecessary to overcome these four challenges only ifthere is some reasonable possibility for major ben-efit to society from having this option availablein the future.

Therefore, we must consider large-scale deploy-ment of nuclear power as a possible outcomeand understand fully the ramifications of turn-ing to nuclear power to provide a significantsource of non-carbon electricity supply. From apublic policy perspective, the scenarios thatmerit analysis are either a large-scale deploy-ment or a phase-out of nuclear power over thenext half-century. We stress that our approachis to evaluate expansion of nuclear energy as anoption possibly needed in the future to meet asignificant fraction of world electricity demandwhile addressing global environmental chal-lenges. We are not declaring a specific goal for aparticular time. Our evaluation criteria are:

favorable economics;

effective waste disposal;

high proliferation resistance; and

safe operation of all aspects of the fuel cycle.

To undertake this evaluation we need to estab-lish a point of reference for nuclear deploymentthat might be realized 50 years from now. To set

this point of reference, we stipulate as the basisof a scenario that nuclear energy will retain orincrease its current share of electricity generationat mid-century.1 The projected growth rate ofelectricity over a half century period is uncer-tain. The average rate of growth will dependimportantly on several variables, notably therate of economic growth and the price of elec-tricity. A range of possibilities2 is presented inTable 3.1.

We adopt 1000 to 1500 GWe as the mid-centu-ry reference point range for our study. This islarge enough to reveal the challenges that needto be faced to enable the large-scale deploymentof nuclear energy. Our analysis and conclusionsconcerning what we refer to as the globalgrowth scenario, as described in Chapter 1,would not change significantly if this numberof deployed reactors were somewhat higher, norif the time period to reach full operationaldeployment were extended. We have examinedthe rate of deployment that would need tooccur for a deployment in the range of 1000 to1500 GWe and note that it is unlikely to proceedin a linear manner; for the next ten to fifteenyears, deployment is likely to be slow, and there-fore the rate would necessarily accelerate dur-

Chapter 3 — Outline of the Study

Table 3.1 Alternative Reference Points for Nuclear Deployment in 2050 in GWe for Different Assumptions about Electricity Growth Rates and Nuclear Market Sharea

ALTERNATIVE AVERAGE ELECTRICITY GROWTH RATES 2000–2050 %

2.0 2.5

172025

838970

1,235

1,0601,2351,545

NUCLEAR GENERATION MARKET SHARE % 1.5

650770880

a. We assume the global average capacity factor increases from 75% to 85%.

MIT_ch03_25-28.qxd 7/16/2003 1:24 PM


assumptions made. The 1000 GWe of nuclearpower assumed in the global growth scenariowould avoid about 800 million tonnes of car-bon equivalent if the electricity generation dis-placed was gas-fired and 1,800 million tonnesof carbon equivalent, if the generation wascoal-fired, (assuming no capture and sequestra-tion of CO2 combustion product). Thus, the1000 GWe nuclear program has the potential ofdisplacing 15 - 25% of the anticipated growth inanthropogenic carbon emissions. In 2050,deployment of 1000 Gwe of nuclear powerwould generate about 20% of worldwide elec-tricity production, if electricity productiongrows at 2% per year. Evidently, the globalgrowth scenario would have nuclear power gen-erating significant amounts of electricity thatwould otherwise likely be generated by fossilfuels.

FUTURE STRUCTURE OF THE NUCLEARINDUSTRY

Significant expansion of nuclear power hasimplications for the structure of its supportingnuclear industry infrastructure. In an unregu-lated economy comprised of private businessfirms competing in the marketplace, marketforces determine the organization and structureof the firms that design, construct, and operatenuclear power plants and supporting fuel cyclefacilities. However, because nuclear technologyinvolves significant public issues of safety, wastemanagement, and proliferation, the govern-ment has a responsibility to ensure that whatev-er industry structure develops will facilitate,rather than impede, attention to these issues.The intersection of these public issues and freemarket operations cannot be handled throughminor government regulation, as is possible insome other industries. An additional layer ofgovernment involvement stems from the tradi-tional structure of electric utilities as verticallyintegrated monopolies. Government interven-tion has been necessary to ensure that the oper-ations of the electric utility industry are effi-cient and that other public objectives for elec-tricity supply are achieved. We do not today

ing the expansion period. The implied con-struction rate near the mid-century endpoint ofthe global growth scenario would be challeng-ing and exceed any rate previously achieved.

The pattern of deployment of nuclear poweraround the world is also important, especiallyfrom the viewpoint of assessing the risks of prolif-eration. Table 3.2 indicates how 1000 1000MWe(or equivalent smaller reactors) might be dis-tributed around the world in the time period2030 to 2050. Although this illustrative deploy-ment is highly speculative, it provides a con-crete instance of how the global growth sce-nario might be realized.

Nuclear power expansion on this scale is notlikely to happen without United States leader-ship. It also requires continued European com-mitment and the initiation or expansion ofnuclear power programs in many developingcountries around the world. If nuclear deploy-ment on the scale of the global growth scenariowere to occur, however, it would avoid a signif-icant amount of carbon dioxide emissions,largely by displacing carbon emitting fossil fuelgeneration. Today, carbon equivalent emissionfrom human activity totals about 6,500 millionmetric tonnes per year. This value will probablymore than double by 2050, depending on the

Table 3.2 Global Growth Scenario

2000PROJECTED 2050

GWe CAPACITY 2050

Total WorldDeveloped world U.S. Europe and Canada Developed East AsiaFSUDeveloping world China, India, Pakistan Indonesia, Brazil, Mexico Other developing countries

1,000

625300210115

50325200

75 50

17%23%

16%2%

19%29%

23%11%

NUCLEAR ELECTRICITY MARKET SHARE

REGION

Projected capacity comes from the global electricity demand scenario in Appendix 2, which entails growth in global electricity consumption from 13.6 to 38.7 trillion kWe-hrs from 2000 to 2050 (2.1% annual growth). The market share in 2050 is predicated on 85% capacity factor for nuclear power reactors. Note that China, India, and Pakistan are nuclear weapons capable states. Other developing countries includes as leading contributors Iran, South Africa, Egypt, Thailand, Philippines, and Vietnam.

MIT_ch03_25-28V1.qxd 7/21/2003 2:23 PM

27C h a p t e r 3 — O u t l i n e o f t h e S t u d y

know how the nuclear industry will evolve butwe mention issues that we believe are an impor-tant determinant of the future success ofnuclear power.

The tension between public responsibility andprivate market operation has been present sincethe beginning of commercial nuclear power. Inthe U.S., the assumption was that any privateutility was in principle capable of owning andoperating a nuclear power plant and should beallowed to do so under appropriate governmentsupervision with regard to safety. Several othercountries have followed this route, notablyJapan and Germany. In other countries, such asRussia and China, nuclear power has beenentirely the responsibility of the central govern-ment. Elsewhere the pattern has been mixed. InFrance, all nuclear plants have been operated bya single state-owned utility, Electricite deFrance. Similar arrangements have applied inSouth Korea and Taiwan. In Spain and Swedena small number of investor-owned utilities havebuilt and operated nuclear power plants.

No arrangement has proved free of tension. Inmany countries with state-owned electricpower monopolies, there has been a movetowards privatization and increased competi-tion, while in the U.S. it is widely recognizedthat in the current environment, small investor-owned utilities operating a single nuclear powerplant are more likely to encounter operationalproblems and to experience higher generatingcosts.

We do not believe that a single organizationalmodel for nuclear power will be applicablethroughout the world. We do believe that indus-trial organization is an important considerationfor the future expansion of nuclear power. Tooversimplify, too much government involve-ment is likely to make nuclear power expensiveand uncompetitive, and too little governmentinvolvement risks safety, waste, and prolifera-tion problems. International cooperation is alsocritical for the effective management of thesepublic issues, especially proliferation. Thus, theindustrial structure in each country must be

compatible with whatever international normsare adopted

The structure of the nuclear industry is alsoimportant because of its influence on innova-tion, productivity, and performance. A neces-sary condition for the expanded nucleardeployment postulated in the global growthscenario is that nuclear power plants and othernuclear facilities be designed, built, and operat-ed to expectation. This performance, in turn,depends upon sound technological choices,high quality design and construction, and theavailability of competent construction projectmanagement teams, craft labor, and operatingand maintenance personnel. Moreover, thegrowth of capability in all these categories mustoccur in the context of a deployment schedulethat will be highly uncertain. These are all mat-ters of industrial organization that are critical tothe prospects for the expansion of nuclearpower but do not happen automatically. Nor isit clear that governments are sufficiently agile orwise to adopt policies that will encourage theproper sequencing of industrial capabilities andneeds.

OUTLINE OF THE STUDY

In conducting this study, our first step was todefine the character of the global growth sce-nario, i.e., the nature and size of the fuel cyclenecessary for it to function. The results are dis-cussed in Chapter 4.

Our second step was to answer the question: “Issuch a mid-century scenario technically, eco-nomically and politically credible?” We do thisby evaluating how well the global growth sce-nario can meet the four challenges of cost, safe-ty, waste disposal, and proliferation risk. This isundertaken in Chapters 5 through 8.

Our third step was to consider public attitudesto an expanded nuclear future. Chapter 9reports on the result of an Internet-based pollthat we conducted and its implications.

MIT_ch03_25-28.qxd 7/16/2003 1:24 PM


Our fourth step was to make recommendationsthat would retain the nuclear option. These rec-ommendations, presented in Part 2 of thereport, addresses both domestic and interna-tional issues and includes both technical andinstitutional measures. We identify organiza-tional changes that we believe would increasethe chance of success of the effort and decreasethe cost. The technical measures involve a sus-tained and disciplined program of analysis,research, development, and demonstration ofvarious aspects of the nuclear enterprise. We donot seek to establish rigid goals or a fixedtimetable for the technical program. The paceof the program should be determined by itstechnical success in the context of the worldenergy and environmental outlook. We antici-pate that the cost of the technical programwould be borne by other countries, as well as bythe United States.

This study approach is conditioned by the beliefthat the nuclear power option makes sense onlyif possible deployment is quite large, since nosmall deployment can make a significant con-tribution to dealing with the greenhouse gasproblem. Support for keeping the nuclearpower option open will therefore depend onconvincing the public and their elected repre-sentatives that large-scale deployment can over-come the four challenges. We believe that estab-lishing a vision for a possible large- scale deploy-ment of nuclear energy that is both technicallyand politically credible is a necessary condition forgaining public support. Indeed it is misleading tofocus on small increases in nuclear capacity jus-tified by significant CO2 reduction. Further-more, small deployments ignore or do not facesquarely the challenges that must be overcomefor nuclear energy to become a significant con-tributor to controlling CO2 emissions.

It will take sustained effort to accomplish thenecessary technical and institutional stepsneeded to make nuclear an attractive energyoption. Given the expected evolution of world-

wide energy supply and demand, however, webelieve there is time to undertake this work . Wedo not believe that nuclear energy will go for-ward without such a comprehensive approach.The construction of a few reactors in the shortterm and a technology driven R&D program isnot sufficient. Although R&D is a vital ingredi-ent, a comprehensive program should addressall four of the key criteria in order to create aclear and sound vision of the energy future. Asimilarly broad approach should be applied toall energy supply and end-use efficiency tech-nologies under consideration. A policy directedto a single solution is inadequate. We also recog-nize that the deployed nuclear fuel cycle will notsimply “jump” to a new reality. But, we believethe evolution will be guided by a clear picture ofwhere we are headed and how we will get there.

NOTES

1. Some advocate hydrogen production as an objective fornuclear power.To be economical hydrogen produced byelectrolysis of water depends on low cost nuclear power.Hydrogen can also be produced by high temperature ther-mal cracking with heat provided by a nuclear reactor.Thisapproach is presently highly speculative. Our belief is that ifnuclear proves to be an economical choice for electricityproduction, it may prove to be interesting for hydrogen pro-duction, whether the production is through electrolysis orhigh temperature thermal splitting of water. However, ifnuclear is not an economical choice for electricity produc-tion, it is most unlikely to be used for large-scale productionof hydrogen.

2. For example, the EIA projects worldwide electricity growthrate of 2.7% for the period 2000-2020. If we project that thisgrowth rate continues [See Table 3.1.] through mid-centuryand recognize that about 350 GWe nuclear capacity is cur-rently deployed worldwide (in over 400 units), then the mid-century point of reference for nuclear maintaining its mar-ket share is 1325 GWe of deployment in 2050.This deploy-ment might correspond to 1325 reactors, each with capaci-ty of 1000 MWe or more units of smaller rated capacity.There are higher and lower projections of world electricityuse.The MIT Emissions Prediction and Policy Analysis (EPPA)project projects an average growth rate between 1995 and2004 of 1.8% that gives a nuclear deployment of 950 Gwe in2050, assuming nuclear power retains its market share. If weassume the EIA 2.7% growth rate to 2020 and the lower MITEPPA 1.8% growth rate between 2020 and 2050, the calcu-lated number is 1164 GWe of nuclear power in 2050.

MIT_ch03_25-28V1.qxd 7/21/2003 2:23 PM

chapter 1 — the future of nuclear power — overview … · chapter 1 — the future of nuclear...

Documents