Foreword
Current stagnation of nuclear power is an undeniable fact, but its causes temporary and not insurmountable. Meanwhile, there appears to be  no realistic alternative to developing nuclear for large-scale replacement of traditional fuels, and first of all hydrocarbons, to cope with the looming problems associated with progressive depletion of cheap fossil reserves and pollution of environment with combustion products. Population growth and economic advancement of developing countries, primarily in Asia, will in all probability aggravate these global problems before the middle of the coming century.

With this in view, the world community has to recognize the need for developing and demonstrating in the near future a nuclear technology which would allow large-scale deployment and, at the same time, would be publicly acceptable. To qualify for this role, such a technology should meet some essential requirements, viz.:

• provision of unlimited fuel supply through efficient use of natural uranium and, subsequently, thorium;

• exclusion of severe  accidents  with radiation releases requiring evacuation  of population under conditions of any equipment failures, human errors, and external impacts [With the obvious exception of extreme nuclear or other impacts resulting in total NPP destruction, such as a nuclear missile attack] achieved primarily due to the natural properties and behavior inherent in nuclear reactors and their components (natural safety);

• environmentally safe energy production and waste management in a closed fuel cycle involving in-pile burning of long-lived actinides and fission products and radiation-equivalent radwaste disposal without disturbing the natural radiation balance;

• barring of the nuclear weapons proliferation pathway associated with nuclear power by phasing out the technologies of plutonium separation from spent fuel and uranium enrichment and by physically protecting nuclear fuel against thefts;

• economic competitiveness due to low costs and fuel breeding, high efficiency of the thermodynamic cycle, and resolution of the NPP safety problems without adding to the complexity of plant design or imposing extreme requirements .upon equipment and personnel.

These requirements in all their variety can be met without going too far from the existing technology developed for military and peaceful applications if the principles of natural (inherent) safety are consistently translated into the reactor and process design.

The significance of such a technology for the entire world community and the immediate interests of the developing nations provide good reasons for initiating an international project to design and build a demonstration plant complete with pilot facilities of a closed fuel cycle. 

I. Nuclear Power at the First Stage
The work on peaceful applications of nuclear energy was started in the USA and the work on peaceful applications of nuclear energy was started in the USA and the USSR back in the·1940s when development of nuclear weapons was in full swing. The main motive behind this effort was the possibility of securing inexhaustible resources of cheap fuel, offered by nuclear breeding. Development of fast power reactors was initiated with this aim in view and the first units were brought into operation in the USSR, France and UK in the '70s'80s.

However, the first NPPs, commissioned in the "50s, were built around thermal reactors fueled with 235U, which had been tried out in production of weapons Pu and T and in submarines and laid the foundation for the first stage of nuclear power development. As of today, its capacity totaling ~350 GWe with an operational record of about 104 realtor-years, the sector accounts for ~17% of electricity generated in the world, while replacing ~6% of conventional fuels consumed in energy generation.

The potential resources of cheap uranium for these reactors are assessed at somewhat more than 107t, which in energy equivalent terms is less than the estimated resources of oil and gas, let alone coal. This means that 235U-fueled reactors are incapable of having a greater effect on the global consumption of conventional fuels. With its current share in electricity production, nuclear power based on traditional reactors, mostly LWRs, can go ahead for another ~40 years to supply the demands of fuel-deficient countries and regions. Nuclear power can be deployed on a much larger scale, using fast reactors - which only some 20 or 30 years ago was believed attainable within this century.

But then, the events of the '70s and '80s gave a sharp twist to the course of the entire energy sector development, and particularly of its nuclear branch. The oil crisis of the '70s was overcome, the fuel market was stabilized, and the energy conservation measures cut down the demands of industrialized countries for stepping up energy production. Large oil and gas fields were discovered and developed in the North Sea. In the '70s, the orders for construction of NPPs dwindled dramatically and in some countries vanished altogether. In the decades to come, the nuclear contribution to the global energy production will most likely decline.

The grave accidents at TMI and Chernobyl called for action to improve safety, with the ensuing increase in the cost of NPPs, and heightened the antinuclear sentiments of the public.

The '70s saw acute awareness of the environmental problems brought about by extensive growth of production, including energy generation. Large quantities of radwaste built up in the nuclear energy sector were increasingly causing grave concerns.
Contrary to expectations, the first generation of fast reactors proved to be much more expensive than LWRs. It is only to be regretted now that the root causes thereof were never brought to light - with the result of an ingrained prejudice against fast reactors stamped as inevitably costly machines. Another persistent fallacy is that fast reactors are akin to an atom bomb. The construction of these reactors ended with the First units.

Fast reactors and the closed nuclear fuel cycle are also associated with the hazard of proliferation of nuclear weapons, wherefore such developments were stopped in the USA and have been recently halted in Europe as well. in other countries, the fast reactor effort has shrunk and is either setting its goals in remote future or is aiming at resolving more narrow problems, such as burning of Pu and long-lived radwaste from LWRs

For these reasons, the expected large-scale deployment of nuclear power was never brought into effect, but is deferred till some not entirely definite time in the future. Nuclear experts have focused their efforts on evolutionary improvement of tradition. reactors, looking forward to a renewed, if very limited, demand for them.

Separate teams in different countries continue looking for a nuclear technology for the future. Many of them went back to the reactor concepts, which had been previously given up for fast reactors: Th-U cycle, circulating molten-salt fuel, sub-critical reactors with accelerators or other neutron sources. Others go on working on fast reactors of the traditional types, counting on a cost reduction due to design optimization and initiation of serial construction.

There is hardly any hope for the whole variety of the options under consideration to move as far as engineering development and demonstration. It is also most unlikely that the uncoordinated studies, which do not even share perception of the goals, will succeed single-handed in selecting one option for serious engineering development.

In the past, various states proved their ability to concentrate the nuclear research effort on important military and other problems. In today's fairly smooth conditions of fuel and energy supply, the governments of advanced countries are showing but a lukewarm interest in the nuclear technology development, providing it with scanty funding and leaving it for the industry to shoulder on a commercial basis.

The nuclear community itself having no definite long-term concept, the current predictions cast nuclear energy for a very modest role in resolving the looming global problems.

II. Energy Technology For The Next Century
Meanwhile, the doubling of the global population expected by the mid-century, mostly due to the developing countries, and the ever-increasing number of nations taking the course of industrialization are bound to cause at least doubling of the world's demand for primary energy and trebling of the demand for electricity.

There is an enormous gap in the energy consumption levels and, hence, in the quality of life between industrialized and developing economies (see the Table).

Electricity consumption in industrialized and developing countries in 1996 (kWh per capita a year)

Ethiopia	Nigeria, Kenya, Cameroon	Pakistan, India	China, Egypt	Iran, Mexico	Republic of Korea, Spain, Italy	Russia,  UK	Germany, Switzerland, France, Japan	USA,  Finland	Sweden, Gander, Norway*
25	130-210	410-460	890-970	1450	4500-4900	5600-6200	6700-7920	13990-14350	16000-25900

*Very high energy consumption in some northern countries is determined not so much by climatic conditions as by
traditional development of energy-intensive industries relying on huge and cheap hydraulic resources.

The effort to bridge this chasm may shape up as a key trend and a prerequisite for stable advancement of the world in the century to come.

The growth of energy production will be in all probability accompanied by gradual depletion of cheap hydrocarbon reserves and by a rise of their prices. The world fuel market is likely to be increasingly affected by the endeavors of various countries to preserve their national hydrocarbon resources as a chief item of export, for many of them, and as fuel for transport and raw materials for chemical synthesis.

The hazard of possible international conflicts over oil and gas sources is another factor to be reckoned with.

Use of fossil fuels, including huge coal resources, can prove to be closer to its end than is currently expected, on account of the emissions of combustion products and global climatic changes. Besides the rise of fuel prices, measures taken to minimize harmful releases are bound to add to the capital costs of the energy sector.

For some countries, proliferation of nuclear weapons remains the most burning problem calling for immediate decision-making, which first of all refers to weapons-grade plutonium and uranium-235 production technologies spreading via the entirely legitimate channels of nuclear power.

It goes without saying that in addressing a certain course of events in a remote perspective we can only talk in terms of probability. But the vital importance of reliable energy supply and the extreme sluggishness of the fuel and energy complex, with its dimensions and capital requirements, call for taking action well in advance of any possible unfavorable turn of events. The main among such provisions would be the development of new energy technologies capable of large-scale and cost-effective replacement of fossil fuels. It has to be admitted, though, that the energetic work of the recent years has so far failed to yield an economically competitive technology of this kind.

For all the possible beneficial applications, the new renewable energy sources are still unable to compete economically with thermal machines in commercial power production on account of the extremely low and uneven energy flux densities. Exceptions to this are hydraulic energy and photosynthesis, with the solar energy fluxes concentrated and accumulated by natural forces. But their uses are affected by certain environmental and economic constraints.

Controlled thermonuclear fusion is yet to go through technical and economic demonstration.

With half a century of practical experience behind them, fission reactors might seem to be an eligible and realistic alternative to the traditional energy sources. But global deployment of thermal reactors fueled with 235U is limited by the availability of cheap uranium, fast reactors proved to be too expensive, and there are no as yet convincing solutions to the problems of NPP safety and radwaste management for the nuclear energy sector to operate on a truly large scale. Light-water reactors need enriched uranium and modern fast reactors call for a closed fuel cycle with separation of plutonium, e.g., from U-blankets. These technologies, which are also key processes in production of nuclear weapons, are being mastered by non-nuclear states which are developing nuclear power with the aim of attaining energy independence. On the other hand, new developments have got stuck at the stage of conceptual search, with small, if any, chances of moving further on their own. industrialized countries are taking a  very moderate interest in the latter, while the developing world, though standing in greatest need of such technologies, seems to be quite content to follow their lead and is not taking over the initiative.

Meanwhile, studies show that it is possible to create a nuclear technology, which will meet the safety and economy requirements of large-scale power production without going too far from the achievements of civil and military nuclear engineering. If, in the next few years, the states concerned recognize the vital need for resolving the problem, and for doing so in good time, and if they succeed in adopting a definite concept, the engineering development and demonstration of the latter can be carried out within a reasonable period of 10 to 15 years.

Thus the stage would be set for a nuclear power industry which in the next century could take on a major part - say) half - of the increase in the global demands for fuel and energy. This means that nuclear power would grow by  an order of magnitude from its present-day level of ~350 GWe by the mid-century and then would double or treble its capacity before the end of the century.

Building nuclear power of such dimensions, with considerable potentialities for tackling the problems of environment and fuel resources, might constitute a goal of the next stage - the Second Nuclear Era as it was called by A. Weinberg.

Nuclear technology development has long since acquired the traits of an international effort and in the century to come will be guided by the global energy requirements, with countries concerned joining forces to carry it on. In so far as, such a need is now greatest in the developing countries, it is their initiative in devising a nuclear technology for the next stage that can really give a practical turn to this work.

Such an initiative would undoubtedly find support with Russian nuclear experts whose vast experience and capabilities are currently in poor demand in the country, as well as with specialists from other countries seeking applications for their expertise.

Creation of a new nuclear technology would also answer the fundamental needs of industrialized nations and ought to be supported by their governments with the understanding that this technology would not add to the risk of a sprawl of nuclear weapons.

The criteria for adopting a nuclear technology for t4e next stage stem from the fairly general view of the future nuclear power discussed below.

III. Long-Term Scenario of Nuclear Power Development
A long-term scenario of nuclear power development, which is of course a most tentative job, is presented in Fig. 1. Curve 1 describes the continued advancement of nuclear power on traditional thermal reactors with 235U fuel - mostly LWRs.

Consuming ~200 t of natural uranium a year per 1 GWe and given ~107 t of cheap uranium, LWRs will have a total output of ~5104 GWeyr, with their operation in reactor-years making roughly the same figure, and will produce -104 t of fissile Pu (~200 kg/yr. per I GWe). Reuse of Pu, recovered at the facilities built in France, UK and Russia, in LWRs could increase the fuel resources of thermal reactors by 15%. But the unprofitability of closing the LWR fuel cycle for plutonium offers no incentive to expansion of these facilities, while the global spread of this technology requiring Pu separation would add to the hazard of proliferation of nuclear weapons.

Besides being far from efficient, plutonium burning in thermal reactors would in the long run bar the way to large-scale deployment of breeders at the next stage. That is why this scenario assumes that the thermal reactors of the First stage will continue operating mostly in the open fuel cycle. Furthermore, it is easier to produce weapon-grade plutonium in pressure-tube heavy-water reactors (HWR).

Many developing countries are showing interest in heavy-water reactors that allow using natural uranium and afford independence from the suppliers of enriched uranium. Increasing their share in the nuclear energy mix of the first stage (from today's ~5%) offers certain saving of natural uranium (a factor of 1.5 reduction of consumption per reactor) and leads to an increase in Pu production (roughly twofold per reactor). Fuel burnup, a factor of 4 to 6 smaller than in LWRs, results in greater buildup of spent fuel and added storage requirements.

Thermal reactors of different types are most likely to find application in a longer term as well, owing to their advantages in some energy production sectors: small and medium nuclear plants (tens and hundreds of MWth) are well suited to meet local heat and electricity needs in remote regions where construction of transmission lines and fuel delivery are difficult and costly or in regions where fuel itself is expensive while industries need, e.g., high-grade heat for technological purposes. Such plants are also convenient for large sea transport. To serve these needs, thermal reactors will subsequently have to shift to the Th-233U fuel cycle and a BR of ~0.8 to I (curve 2), with the 233U deficit covered by breeders.
 
 
 

Fig. 1. Tentative scenario of nuclear capacity growth
1- thermal reactors using 235U (mostly LWRs), 2- thermal reactors with Th-233 fuel cycle; 3 -  all nuclear capacities (thermal and fast reactors); 4-all generating capacities (nuclear and non-nuclear).

 

But centralized electricity production at large NPPs (of 1 GWe capacity), with power transmission to the grids, will in all probability remain the main sphere of nuclear energy application. Electricity is still the most universal and convenient form of energy, well suited for transmission and final uses; its generation grows at the quickest rate and will account for the predominant part of fuel consumption in the next century (its current share being roughly 1/3).

The experience with high-voltage transmission amassed in Russia among other countries and the anticipated advent of economical superconducting lines in the next century, open up possibilities for transmitting electricity from large NPPs over thousands of kilometers and for expanding its export.

The trend toward miniaturization observed in other industries is opposed here by the indivisibility of the technological process and by the increase in specific costs of construction and operation with lower capacities, which is especially true for NPPs where such costs are the largest contributors to the total expenses.

Electricity being a standardized and universal product, changes in the market demand do not entail reorganization of the production process, which together with the ease of long-distance transport is another point in favor of large power plants and reactor units. This does not preclude the use of nuclear energy for heat supply, while utilization of waste heat from NPPs remains an important problem yet to be solved.

On these grounds, large-scale nuclear power deployment, represented by curve 3 with a conventional start in 2020, is assumed to involve construction of mostly large NPPs. Such a scale can be provided only by breeders with BR³1.

A crucial objective of this stage is the cost-effective and safe utilization of Pu produced both by the reactors of the first stage (104 t) and as a result of nuclear arms reduction (possibly over 200 t). With the use of Pu, BR³1 is attainable only with fast reactors, which predetermines their principal role at this stage.

C. Rubbia and other researchers are considering an option of starting up fast reactors on Th-Pu fuel and then easing them into the Th,-233U cycle with a view to alleviating the radwaste and nonproliferation problems. It should be borne in mind, however, that the Th-U cycle is not nearly as good with fast reactors as is the U-Pu cycle in terms of not only the neutron budget but also of the associated thermohydraulic and other characteristics which are pivotal for economic efficiency and safety. From the viewpoint of radwaste and nonproliferation, the Th-U cycle has small, if any, merits.

There is one fundamental advantage of the Th-U cycle, which lies in a good neutron and fuel balance in thermal reactors to the extent of their reaching BR~I. This advantage would be a decisive factor, if the high costs proved to be characteristic of not only the first generation but also of all fast reactors. However, there is no serious evidence to justify this view. Quite on the contrary, by the physical and engineering principles of their design and control, large fast reactors with a liquid metal coolant are inherently simpler than LWRs and other thermal .reactors; besides, they have a higher fuel and energy efficiency and, hence, can be cheaper too, provided better solutions are found for their design. There are apparently no reasons for the high cost of the first fast reactors, other than the use of the highly chemically active sodium. Prevention of its contact with water and air under normal operation and during accidents requires a 3- circuit cooling configuration, a safeguard vessel, sophisticated systems for monitoring and protection of steam generators, and for refueling, and affects the auxiliary equipment and structures of the NPP. The possibility of sodium taking fire or coming to the boil during accidents, with consideration for the positive - at least locally – void effect of reactivity,  prevents fast reactors from fully realizing their inherent safety potentialities.

One of the main reasons for using light-weight and heat-conducting sodium as a coolant in the First fast reactors, was its capability to remove high heat fluxes from the fuel, with a resultant decrease in the fuel inventory and in the Pu doubling time, T2. In the post-war decades, the annual rate of energy production growth reached 6 to7% (up to 12% in the USSR) and, with Pu being scarce, short doubling time, T2, was regarded as an important criterion in fast reactor development. Along with high power density, another associated requirement was a high breeding gain (BR-I), for which purpose a uranium blanket was provided. For a longer term, consideration was given to high-density  and  heat-conducting  fuels,  such  as  metal  alloys,  monocarbides  and mononitrides, which afforded a simultaneous rise in the BR and power density as compared to the well-tried oxide fuel.

The situation is dramatically different now. The growth rates have dropped (a threefold' increase of electricity production over slightly more than 50 years corresponds to an average rate of ~2% a year), Pu is building up in large quantities, so short T2 is no longer needed. The scenario depicted in Fig. I may well be fulfilled by fast reactors with BR~I and moderate power density. Given 104 t of Pu and ~1.5104 t of  235U in the spent fuel of the first-stage reactors, it is possible to bring in fast reactors ~4000 GWe in capacity, using Pu mixed with slightly enriched (1to 496) uranium (additionally enriched regenerated fuel of thermal reactors). As nuclear power settles on an even keel, these reactors will move into the U-Pu cycle. With an optimum CBR of l.05 (minimum reactivity variations), nuclear generating capacities can reach ~8000 GWe due to Pu breeding in the early 22nd century. Therefore, their development may focus entirely on the safety and economy objectives. These goals can be met by replacing sodium with a chemically passive high-boiling coolant, eliminating the uranium blanket with assured in-core breeding CBR=BR~1, and by using high-density, heat-conducting fuel instead of oxide fuel (for the purpose of attaining CBR-I and reducing reactivity margins, rather than for increasing power density). It will be shown below that these and other measures result in an economically efficient high-power fast reactor with an essentially higher level of safety.

Excess neutrons in a fast reactor without a U blanket in the U-Pu cycle and a high flux of fast neutrons endow fast reactors with the advantage of transmutation of long-lived radionuclides, which allows resolving the problem of radwaste without creating special burners. The equilibrium fuel composition (CBR-I) opens the way for a reprocessing technology which consists basically in moderate removal of fission products and rules out Pu extraction in this process. Use of such a technology in "non-nuclear" countries would afford a certain degree of their independence from nuclear nations without violating the international nonproliferation regime.

This discussion leads us to the conclusion that the choiceof fast reactors in the U-Pu cycle as a basis for large-scale nuclear power, made by its founders back in the 1940s-50s, remains valid in the new conditions as well. [Back in the 1940s, E. Fermi in the USA and A. I. Leipunsky in Russia came to appreciate the unique neutron surplus fast reactors with U-Pu fuel, which was theoretically assessed at over 2 as against ~I in thermal reactors running on uranium-235 and plutonium-239 and against ~1.3 in the case ofuranium-233. Aside from energy, neutron surplus is the main asset of the reactors and its availability in excess of I provides physical grounds for overcoming not only breeding problems but, given adequate engineering solutions, problems of safety, waste, nonproliferation and, in the final analysis, of economy.] But these conditions and the experience amassed call for new approaches to the creation of such reactors.

It has been already mentioned above that thermal reactors have a scope for long-term development in some energy sectors, with their switchover to the Th-U cycle in the future. Considering their relatively small proportion in the future nuclear energy mix, it can be shown that the 233U deficit in these reactors may be covered without too much trouble by providing fast reactors with a small thorium blanket to utilize part of the leakage neutrons.

With breeding well established and the problems of radwaste settled - mainly through transmutation of long-lived actinides - there seem to be no constraints on the duration of nuclear power operation from the viewpoint of cheap fuel availability and radwaste accumulation. But a complete concept of nuclear development should incorporate, among other things, the final stage with phaseout of NPPs and elimination of large quantities of radioactive material from the reactor inventories. This suggests the need for effective burners without nuclear fuel reproduction, which makes the ongoing quest and studies in this area meaningful. However, even if our hopes for the advent of economical and safe breeders come true before too long, the engineering development of such burners will still be a task of some more remote future.

IV. Requirements to Reactor and Technology, Choice of Reactor Type

1. URANIUM CONSUMPTION, FUEL CYCLE

• Efficient utilization of stockpiled Pa, redaction of specific uranium consumption by no less than an order of magnitude without aiming at short doubling times.

• High-power fast reactor in the U-Pu cycle, moderate power density, CB=CBR~1, no uranium blanket.

• Minimum capacity of a reactor with CBR-I assessed at -300 MWe. CBR~1 calls for high-density-fuel. For many reasons, UN-PuN fuel appears to be an optimum choice. The environmentally hazardous 14C produced in the 14N (n,p) 14C reaction, makes a small contribution (~1 %) to the total radiotoxicity of radioactive waste. If insignificant quantities of volatile carbon compounds are generated and released in the course of fuel recycling and if this element is incorporated into stable radwaste compositions sent to disposal, there will be no need for nitrogen enrichment with the 15N isotope. In the future, however, it may prove useful for certain neutron budget improvement.

2. NPP SAFETY
Exclusion of severe accidents which may result in fuel failure and large radioactive releases (prompt criticality excursion, loss of coolant, fire, steam and hydrogen explosions).

If the operation period of nuclear power in the next stage of its development exceeds 106 reactor-years, the probability of the above accidents should be kept well below 10-6 per reactor-year. Probabilities of such a level,: calculated by PSA methods, have neither operational experience (the existing nuclear power has operated for about 104 reactor- years) nor convincing theoretical data to support them. PSA techniques are useful for planning safety improvements at NPPs and allow quite dependable predictions related to the near-term nuclear power development, but they are unsuitable for preparing a really strong safety case for large-scale nuclear power.

Therefore, at reactors of the next stage such accidents should be deterministically excluded under conditions of any human errors, failures of or damage to equipment and safety barriers, owing to the intrinsic physical and chemical properties and behavior of the fuel, coolant and other reactor components (natural safety).

Needless to say, there is no way to avoid radioactive releases in case of total reactor and plant destruction as a result of a nuclear attack or a fall of a large asteroid, and these events should be mentioned in the design documentation as exceptions. [Normally, it is possible to estimate the probability of extreme natural impacts. If found to be much lower than 10d per reactor-year, as is the case with a large asteroid, the probability may be regarded as negligible and the accident in question excluded.]

Fast reactors with high-density fuel, operating in the U-Pu cycle, can be designed to have optimum CBR~1, no Xe and Sm poisoning, small power effects of reactivity due to high heat conductivity of the fuel, and a small effect of delayed Np decay. This allows keeping the total reactivity margin at a level of DKtot£beff and hence excluding a fast runaway under any erroneous actions or failures in the reactivity control system.

Without a uranium blanket, a fast reactor has a deeply negative integral void effect. Passive  control  and  cooling  features,  feedbacks governed by ta large: negative temperature coefficient dK/dT, a high-level of natural circulation of the coolant,  prevent dangerous temperature growth under of normal conditions.

Sodium interaction with air and water, which may result in burning and explosions, in hydrogen generation and in a loss of coolant, together with the possibility of a local positive void effect during boiling of this coolant, suggest that it should be traded for another, chemically inert, coolant which boils at a much higher temperature. With the high power density in the core and the short doubling time T2 no longer required; it becomes possible to use, for instance, the heavy coolant which has been successfully employed in Russian naval reactors, namely PbBi eutectic - or Pb which is close to the former in all physical and chemical characteristics,' except for the melting point. Use of Pb settles the problems of the high cost and small resources of Bi, and of volatile 210Po with its high alpha activity, produced from Bi. The problems caused by the high melting temperature of Pb (3270C) can be resolved through the use of proper temperature conditions and reactor cooling so as to stay within acceptable steel temperatures and to exclude blockage of lead circulation paths even under off-normal conditions.

All potential accidents in a naturally safe reactor caused both by internal and external impacts, except for those mentioned above, arc treated as design-basis events.

The deterministic safety requirement implies that the ultimate design-basis accident (UDBA) - i.e. the accident which covers any event resulting from human errors or multiple failures of equipment) including loss of forced cooling) failure of the scram function, insertion of full reactivity margin, damage to outer harriers such as containment and reactor vessel - should not cause fuel failure and radioactive releases such that would require evacuation of people from, the territory around the plant.

Analysis of hypothetical (non-credible) accidents including large and rapid reactivity addition, fuel failure and collapse with resultant secondary criticality, is an optional job performed to obtain ultimate estimates.

Extreme external impacts leading to destruction of the plant, reactor and its vault, will be mentioned in the design documentation, but their analyses are also optional.

3. RADWASTE
Any predictions concerning safe disposal of large amounts of radwaste for tens or hundreds of thousands of years give rise to doubts about the validity of geological and especially "historical" forecasts for such remote future.
These doubts can be removed if the radiation hazard from buried radwaste is brought into balance with that of uranium extracted from the earth (radiation-equivalent radwaste disposal), and this is adopted as a requirement for nuclear technology.
The requirement can be satisfied in the following way:

• long-lived products of U decay (Th, Ra) can be co-extracted with uranium in the mining process and then handled together with actinides. This step will also facilitate rehabilitation of U mining areas on completion of the work there;

• U, Pu, and other actinides produced during reactor operation, first of all Am, can be returned to reactor to be transmuted by fast neutrons into fission products;

• radwaste can be subjected to treatment with a view to removing actinides so that it contain only ~10-3 of Pu;

• after cooling, radwaste can be brought into a mineral-like state or some other physical and chemical form not prone to migration in soil;

• radwaste can be buried in naturally radioactive geological formations remaining after U mining, in such amounts that they will be equivalent to extracted U in terms of their radiation and biological hazard. [The present-day processes of uranium mining without co-extraction of thorium and radium, which disturb geological structures and cause flooding, may sometimes hinder complete rehabilitation of mining sites and radiation-equivalent disposal of waste. But this goal should be approached step by step, bearing it in mind that breeding will cut uranium input in the energy cost to such small proportions that any improvements in the uranium mining process the sake of environmental protection, will be well justified.]

It should be pointed out that the long-lived Np produced in reactors has a low activity and can be dumped untransmuted without affecting the radiation equivalence. Moreover, if returned to reactors, Np adds to the fuel activity by producing highly active 238Pu and 236pU which decays to 232U. This increase, however, is not too great to prohibit returning Np into the reactor together with Pu and Am.

Cm, whose main isotopes have a relatively short half-life and a high activity, especially neutron-induced, would significantly increase the fuel activity if returned to the reactor for transmutation, thus impeding fuel refabrication. Therefore, it would be better if Cm were separated from fuel during reprocessing, cooled for some 70-100 years and then returned to the reactor in the form of decay products, i.e. Pu isotopes.

Attainment of radiation equivalence between mined uranium and radwaste at the time of its disposal may be facilitated by separation of Sr and Cs so that only 1-5% of them will remain in waste. The extracted Sr and Cs may then be utilized as radiation or heat sources. Long-lived I and Tc (with 1-10% of them going to waste), if extracted, can be returned to reactors for transmutation. The remaining radwaste (with the activity of about 104 Ci/l) can be stored in casks cooled by dry air under natural circulation. The activity of the remaining radwaste stored in this way would decrease by three to four orders of magnitude in 200 years, which simplifies the technology of final disposal and enhances its safety (Fig.2). Analysis shows that such storage facilities can be designed to be fairly simple and moderately expensive.
 

Fig. 2. Radiation balance, without (S-1) and with allowance for migration (S=IO) S=ARE /Au as a function of the time of long-term monitored cooling of LHW. with contribution to waste, %:

 

Curve	Sr	Cs	U	Pu	Np+Am+Cm
1	10	15	0.05	0.1	0.1
2	0.1	1	0.05	0.1	0.1
3	0.1	1	0.01	0.01	0.01

There would be no problems or risk associated with long-distance transport of radwaste and fuel if fuel cycle and radwaste storage facilities were set up on NPP sites.
 

4. NONPROLIFERATION
A fast reactor with CBR-I and having no uranium blanket uses fuel of equilibrium composition which does not necessitate Pu separation or addition during refabrication. To adjust the fuel composition, it is sufficient to add 238U to compensate for its burnup.

This fact allows introducing the following requirement the reprocessing technology should rule out all possibilities of using it for Pu .separation. In this case, reprocessing will boil down essentially to removing, fission products from the fuel.  From the viewpoint of their influence on reactivity, it is acceptable to remove fission products so that ~1-10% of them remains in the fuel, which would simple the technology and facilitate  its  choice,  though  increasing  the  fuel  activity,  in  particular  during refabrication. This is not a major complication, however, since the process is remote anyway. Besides, a high activity of fuel is another warranty against its theft.

The main point here is that such a technology will not add to the risk of proliferation and hence may find worldwide application.
Needless to say that no technological improvements are able to rule out illicit application of the existing techniques of Pu separation - e.g., from LWR fuel – or uranium enrichment for the purpose of obtaining weapon-grade materials. This problem can be successfully dealt with only by political steps meant to enhance the nonproliferation regime and improve the safeguards. Moreover, it has to be resolved irrespective of the further route of nuclear power and its technology.

Promotion of the breeding technology appears to be the most cost-effective way of utilizing Pu accumulated in spent fuel from modem reactors. In this case, spent fuel should be moved from cooling ponds into reactors and fuel cycle facilities which afford much better protection and reduce the risk of illicit plutonium separation and utilization.

If, in addition to the above merits, this technology becomes sufficiently attractive in terms of economics, safety and waste management, the countries concerned may progressively introduce it to replace both traditional breeders and LWRs running on enriched uranium, thus barring the legitimate path for proliferation of nuclear weapons.

Fast reactors without a uranium blanket, with CBR-I and a moderate power density, have many traits and capabilities essential for attaining this goal:

• They have no U blanket to produce weapon-grade Pu and, owing to the small reactivity margin, exclude the possibility of U assemblies placed in the core for Pu accumulation;

• Small reactivity variations during refueling, moderate power density and on-site arrangement of fuel cycle facilities allow quasicontinuous on-load refueling. Spent fuel can be cooled during 31012 months in an in-vessel storage facility and then sent directly to reprocessing and refabrication. Hence, there will be no need for out-of-pile storage of spent and fresh fuel. Such fuel handling can greatly simplify supervision and practically excludes fuel thefts and long-distance shipments.

Initial reprocessing of spent fuel from thermal reactors and fabrication of first cores for fast reactors will have to be done at facilities available in the nuclear countries, but this dependence will not be as strong as the dependence on regular supply of enriched uranium. Consideration may be given to setting up nuclear technology centers around these facilities under international jurisdiction.

The aqueous technology widely used now and other options being studied at the moment, are tailored to the existing reactors which require Pu separation. To meet the above requirement, it is necessary to alter the existing reprocessing technologies or to develop a new process, and this is one of the Or challenges along with the development of new reactors' In addition to chemical processes, consideration should be also given to physical methods for removal of fission products from fuel, which rely on a factor of two difference in atomic weights, diffusion, thermal treatment, etc.

No concept of a closed fuel cycle satisfying the requirements of large-scale nuclear power, has been suggested yet. But it will not fail to appear once the objective is properly defined, and then the required technology can be developed and demonstrated within the period of the reactor development.

5. ECONOMICS
With cheap fuel, the cost of nuclear electricity production is largely determined by the costs of NPP construction, operation and maintenance, which have grown considerably on account of the engineering measures taken to enhance the plant safety. NPPs of the next stage should be cheaper than modern LWRs, to be economically competitive in many countries and regions.

As is the case with most of the sophisticated technological systems, the NPP cost is a product of many factors. No separate improvement in one area (for instance, use of natural water circulation instead of pumping it in BWRs) can reduce this cost by more than a few percent. NPPs with fast reactors should be made at least twice cheaper, which calls for some overall approach covering the main equipment, systems and structures, whose high costs stem from safety requirements. The answer in this case comes from the philosophy of natural safety.

A truly high safety level of new plants, achieved mostly by elimination of potentially dangerous design solutions and due to the use of the laws of nature, will make it possible to simplify plant design, to ease the requirements for basic and auxiliary systems, structures and for personnel, and will obviate the need for additional safety systems.  These potentialities can be translated into plant design with consistent application of the natural safety principles.

V. An Example of a Naturally Safe Fast Reactor
The conceptual design of a fast reactor with UN-PuN fuel and lead coolant (BREST) developed not so long ago, proves that it is possible to meet all the above requirements using a proven technology.

Design and analytical studies were performed and then optimized for reactors with capacity in the range from 300 MWe to 1200 MWe. Experiments were carried out at U-Pu-Pb critical assemblies to verify reactor physics and to refine nuclear data. Steels were subjected to long-term corrosion testing in Pb circulation loops. Experiments were performed to study Pb interaction with air and water of high parameters, interaction of nitride fuel with Pb and steel claddings, etc.

Computational studies on the ultimate design-basis accident, as defined above, have shown that this reactor can survive it without fuel failure and with moderate radioactive releases. Investigations into hypothetical accidents indicate that reactivity addition of up to several beff at a rate of up to 50 b/s does net-cause lead boiling or large release of mechanical energy. According to these studies, lead density, which is close to that of fuel, and its convective flows prevent fuel collapse which might otherwise result in the formation of a secondary critical mass.

The lead-cooled fast reactor has a simpler design as compared to an LMFR with a sodium coolant:

• single vessel or pool-type configuration without a metal vessel (the reactor is placed directly in a concrete vault with thermal insulation between concrete and lead);

• two circuits for and emergency cooling; decay heat removed by natural circulation of air in tubes located in the lead coolant of the primary circuit;

• refueling without washing the coolant off fuel assemblies;

• reactivity control provided mostly by lead in tubes located in the side blanket, with lead levels in the tubes regulated by gas pressure;

• passive control and protection features, including those with threshold response; high level of natural circulation of coolant; less stringent requirements to the speed of operation with simplification of control and protection systems;

• simpler design of steam generators, with no need for fast-acting leak detection systems and quick-response valves;

• less  sophisticated  fire  protection,  ventilation  and  other  support  systems  and components; simpler rooms of the cooling circuits and other plant constructions.

The technical and economic estimates made for an NPP with two 1200 MWe BREST reactors confirm that it is possible to reduce capital costs of such NPPs and the cost of their electricity, as compared to those of VVER plants.

Operating experience of reactors with a ~ heavy coolant, extensive in-pile testing of nitride fuel, calculations and experiments performed in the course of the conceptual design, made the basic aspects of the concept clear enough to embark on engineering development. The engineering design of a demonstration BREST-300 plant with a fuel cycle has already gone through its first stage and is to be completed in 2002 together with the essential computational and experimental verifications. The strategy of nuclear power development in Russia envisages construction of one unit at the Beloyarsk NPP site before the year 2010. The costs of R&D and construction of BREST-300 with the pilot fuel cycle facilities are estimated at about I billion dollars for the case of this being an entirely Russian effort. Based on the BREST-300 experience, it is planned to design and build a first NPP of this type before the year 2030, which will open a new stage of nuclear power development on a large scale.

For the design and construction of the BREST-300 NPP, Russia has at its disposal competent experts, institutional and experimental facilities in all the essential areas, machine-building and construction capabilities. On the other hand, while recognizing that an international effort would take more time and resources, Minatom sees some fundamental and pragmatic points in favor of inviting the countries concerned to scientific, technical and financial cooperation in this project at certain stages and under conditions to be agreed upon.

Consideration should be also given to the possibility of a full-scale international project which may begin with joint definition of objectives and adoption of a concept for engineering development. An example of such a project is outlined below.

VI. International Project
If the Governments concerned succeed in reaching an agreement in respect of their intent to undertake a joint demonstration project of a proliferation-resistant naturally safe nuclear power plant, the associated work could be carried out within a reasonable time, proceeding as follows:

Phase 1

• organizational provisions for the Project;

• development of requirements for the reactor and the fuel cycle;

• adoption of reactor concept;

• preliminary design of the 150-300 MWe demonstration reactor;

• R&D Program elaboration;

• full-fledged agreement on the Project.

Phase 2

• engineering design of the reactor and the associated R&D;