Nm1 N.- 1 This report is not to be published nor may it be quoted as representing the Bank's views. INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT ECONOMICS OF NUCLEAR POWER June 4, 1956 Department of Technical Operations ECONOIC NUCLEAR POWER TODAY: WHERE AND UNDER WHAT CIRCUMSTANCES ? A Study by Corbin Allardice Adviser on Atomic Energy International Bank for Reconstruction and Development Washington, D. C. June 4, 1956. Author's Note This study has been prepared in an attempt to establish benchmarks against which can be assessed the economic feasibility of a 100 Mw or larger electric capacity nuclear reactor based on essentially today's technology. Such reference points are necessary in the determination of the Bank's role in nuclear power development and application, and in the evaluation of proposals for financing such an installation. The need for such a study was pointed out by Mr. S. Aldewereld, and his contributions to its inception and its execution are gratefully acknowledged. Mr. M. Rosen also provided invaluable guidance, as did Dr. W. Rembert, Ir. A. Wenzell, Mr. S. Lipkowitz and Mr. B. Walstedt. Messrs. F. Quackenboss and H. Hollister, of the United States Atomic Energy Commission also provided helpful assistance. TABLE OF CONTENTS Page Introduction ......................................... 1 - 3 Plan of Stuly ....................................... 3 - 4 Nuclear Energy Resources ............................ 4 - 6 Nuclear Fewer Facilities ............................ 7 - 13 Cost of Nuclear Power Facilities.................... 13 - 19 The Estimated Cost of Nuclear Power .................. 19 Operating and Naintenance Costs ..................... 19 - 21 Fuel Costs ............................................21 - 27 Amount and Pattern of Use ............................27 - 26 Depreciation .........................................28 - 31 Generating Costs for Nuclear Power, excluding Financial Charges ................................... .......31 - 32 Total Cost Including Return on Investment ............ 32 - 33 Comparative Costs of Nuclear and Conventional Thermal Power ............................................ 34 - 35 Conclusions ............................................3 - 36 Appendix I - Heat, Propulsion, and Smaller Power Nuclear Reactors ................................. 1 - S Charts (Nos.l - 7) INTRODUCTION 1. World energy demands are.increasing at such a rate as to require the full exploitation of both fossil and hydroelectric energy resources. Even nore importantly, world energy demand forecasts for the year 2000 indicate the absolute need for development of nonconventional resources. The rapid worldwide expansion of demand for electricity, which on the average has doubled every ten years, is particularly striking. The nonconventional resource most likely of early practical application, particularly to the generation of electricity, is nuclear energy. 2. While worldwide energy needs can be estimated over a relatively long term with considerable reliance, estimation of the short term energy needs of individual countries is a far more difficult task. Nevertheless, various studies that have been made indicate that except in those countries possessed of great untapped hydroelectric resources or plentiful resources of coal or oil, there is a general need for the development of unconventional energy resources to meet rising power needs in the next one or two decades.1/ This means that nuclear power, assuming it can be produced and sold at prices competitive with other energy sources, will play an important role in the economic development. 3. Besides the need to develop nuclear power to meet rising energy re- quirements, other factors operate in some countries to speed nuclear power development and early application. Industrial pressures to develop markets Numerous papers on world energy needs, the energy requirements of individual nations, and the possible role of nuclear and other non- conventional energy sources are contained in Vol.I, Proceedings of the International Conference on the Peaceful Uses of Atomic Energy. - 2 - for nuclear fuels and for nuclear power equipment act as stimuli in the more highly developed nations. Fear of "atomic colonialism" spurs other smaller nations to develop technical competence in the field. To the underdeveloped countries, nuclear power offers unique attraction because of its transporta- tion independence, flowing from its high energy content per unit weight. These and economic, social and political forces tend to place a high premium on nuclear development, even where economic need for a new source of energy is not urgent. On the other hand, there are well-recognized limitations on nuclear energy growth other than demand or ability to compete economically. In particular, many nations have definite limits of technical manpower and resources, and in these cases a decision to enter the nuclear power field may work to the disadvantage of more immediately needed economic development activities. 4. The Bank's interest in nuclear energy is primarily in facilities for the generation of electricity or heat from nuclear fuels, and the related production and processing facilities. Important other applications of nuclear energy to research and development, therapy, radioisotope work, food preservation and sterilization, and so forth, in general require small in- vestment with consequent small foreign exchange requirements. While the Bank may become involved in loans for such applications, the individual amounts would be relatively small. In the case of power, process or space heat or propulsion reactors, the capital requirements are large, with varying needs for foreign exchange. The remainder of this paper will deal only with nuclear power plants.- . Process or space heat, propulsion, and small power nuclear reactors are discussed. in Appendix I. - 3 - 5. No efficient means has been found to use the energy of fission to produce electricity directly. As presently conceived, nuclear reactors are simply machines in which some of the energy contained in the nuclei of atoms of fissile material (U235, U233, and Pu239) is released to become available in the form of heat which must be removed from the reactor by a coolant, such as water, gas, organic liquid or liquid metal, in order to make steam for heating or to drive a turbo-generator. (A gas coolant might be used to drive a gas turbo-generator or propulsion turbine.) The turbo-generator portion can be considered more or less standard. Thus, essentially, nuclear reactors are equivalent to the "fire-boxes" (including, in some cases, a por- tion of the "boilers") of modern thermal electric stations. 6. M1any different specific nuclear reactor systems have been proposed. Those appearing most promising or presently under advanced development or construction are briefly described in paragraph 13 et seq. As of today no one can say which of the presently conceived reactor systems will prove ultimately to be the "best" system. Indeed, analysis of statements of proponents of the various reactor systems and study of their technical details suggests that no one of them may be so significantly more attractive than the rest as to be the preferred reactor design, although some systems are today further along in development than others. As will be seen in later discussion of the reactors being built in various countries, the choice of reactor design is not purely a technical decision.L/ Plan of Study 7. Mr. Eugene R. Black, in a statement made in August 1955, pointed out that the development of commercial applications of atomic energy had important 1/ See Para.13 et seq. implications for economic development and for the International Bank. He said that at that time '".......no one can say where or under what circumstances these applications may become practicable......". The purpose of this study is to analyse the present status of nuclear development as concerns commercial nuclear power plantsJ to establish reasonably conservative costs for such facilities built on essentially today's technology, and to arrive at operating and maintaining costs and at costs of fuel for such a plant. It will then be necessary to con- sider how the nuclear plant should be depreciated, and at what average plant factor it is likely to be operated over its lifetime. The total cost of nuclear power can then be calculated at various percentage returns on investment. The costs thus established for electricity generated in a nuclear plant can then be compared with those for electricity generated in a conventional thermal plant, at various costs for fossil fuel. Analysis of these comparisons will provide the desired benchmarks to appraise "where and under what circumstances" nuclear power may become economically practicable. Nuclear Energy Resources 8. The primary raw material for nuclear power is uranium. Thorium is the basic raw material for U233 but can only be used in conjunction with some fission- able material providing neutrons to produce U233. A survey of available estimates of economically recoverable uranium and thorium indicates supplies equivalent to many times the energy content of the reserves of oil, gas and coal. Perhaps the most definitive statements are contained in a survey paper by Jesse Johnson, Director of the U. S. Atomic Energy Commission's Raw Materials Division, given at the Geneva Conference. Pertinent paragraphs are quoted below: "In 1948 the uranium supply of the Western Nations was almost entirely the product of two mines, one in the Belgian Congo and the other in Northern Canada. In the past, there had been little general interest in uranium and throughout most of the world there had been no serious search for it. Even now, vast areas promising from a geological standpoint are relatively unexplored. "Today there are major uranium operations in the Belgian Congo, Canada, South Africa, and the United States. Australia, France and Portugal also are producing uranium with favourable prospects for substantially increased production. "On the basis of present developments and geological evidence, resources of the producing nations of the West are estimated to be between one and two million tons of uranium. This uranium can be produced at moderate cost at an average of about 10 a pound for U 0 ...,furanium oxide7.... in a high grade concentrate........ ".....Reserves of commercial phosphate rock in the U.S. alone are estimated at 5 billion tons and the uranium content at 600,000 tons. The U.S. also has an estimated 85 billion tons of marine shale averaging slightly more than 1/10th of a pound of uranium per ton. This repre- sents a reserve of 5-6 million tons of uranium. "Known deposits of uraniferous phosphate rock and shale in other parts of the world equal or exceed those of the U.S. in grade and tonnage. The phosphate deposits of Horocco estimated at 20 billion tons are uranium-bearing. The Scandinavian Peninsula and other Baltic territories contain very large deposits of uraniferous shale. Uranium- bearing coal and lignite also have been found in a number of countries. "The cost of extracting uranium, as a primary product, from phosphate and shale may be between '30 and 0i50 per pound. If valuable by-products can be recovered the cost may be reduced. Between the commercial uranium deposits of today and the high cost uranium sources for the distant future there are deposits of good supply uranium at a cost of between 10 and i 30 a pound. The resources in this economic class are not well known but they must be large, perhaps several million tons of uranium...... Experience gained from the present uranium program has demonstrated that higher prices will bring in new sources of production and increase available reserves. "This general review of production and reserves indicates that uranium no longer can be considered a rare metal. There are exten- sive deposits throughout the world and there are processes for extract- ing the uranium economically. Uranium production already developed is sufficient for a major nuclear power program of worldwide extent. Additional production can be obtained when needed. When the vast low- grade resources are required, more efficient use of nuclear fuel through improved conversion or "breeding" may offset the higher uranium cost." (Emphasis added) 9. It is clear that the nuclear power industry will not be limited by lack of availability of fuel. It is important, however, to note that de- posits of uranium presently considered economically recoverable exist in only a few countries. Even for countries having such uranium supplies, a large capital investment is required to convert the ores to useable fuels. At present, only Belgium, Canada, France, U.K. and U.S . (apart from USSR) are producing high purity uranium metal on a comnercial scale suitable for nuclear power reactor use. Other countries, either not possessing uranium resources or having resources but lacking plants or necessary capital to convert the ores into useable fuels, will have to execute intergovernmental agreements with nations hav4ng nuclear fuels in useable form in order to support a nuclear pawer industry today or in the immediate future. In this regard, the U.S. has recently announced its willingness to make avail- able 20,000 kilograms of U235 for nuclear power reactor uses outside the U.S. All of the nations listed as presently producing uranium metal have indicated willingness to provide supplies to others for power uses. However, in the case of the U.S., and undoubtedly in all other cases, some sort of inter- governmental agreement will be required between the buyer and the seller nation. 10. This fact has obvious and important implications to the Bank since without fuel a nuclear plant is useless. In addition to all other factors entering into the evaluation of a project for a power plant, the Bank must assure itself that the necessary intergovernmental agreements for the supply of fuel, the reprocessing of used fuel elements, and the recovery of plutonium. uranium 235 and uranium 233 are made and that there is a reasonable likelihood that the agreements will continue in effect for the full period covered by the loan. -7- Nuclear Power Facilities 11. While the general concept of producing nuclear power is simple, involv- ing as it does the use of a reactor simply as heat source in lieu of a con- ventional firebox-boiler, the practical design of a specific reactor system involves enormous difficulty. Moreover, a choice has to be made between the number of design possibilities of comparable attractiveness open to the engineer. He must decide among three fuels, and must determine from nuclear considerations and from operating economics the amount and form of the chosen fuel. He must also determine what fertile material he will employ since, except in special purpose reactors, it is necessary to convert some non-fuel material (fertile material) into fuel material in order to achieve economic nuclear power. He must choose the neutron energy range in which the reactor will operate, since nuclear characteristics of the fuel, the fertile material, and all other materials in the reactor vary with neutron energy and thus effect importantly the performance and economics of the power plant. He must choose a suitable coolant to remove the heat. His system may be heterogeneous with pieces of fuel embedded in a moderator or otherwise fixed in discrete positions; the system may be homogeneous in which the fuel is a slurry or solution. If his reactor system operates in the slow or intermediate neutron range, he must select a suitable moderating material (moderator) to slow down the fast neutrons released in each fission event. Reflectors, control systems, operating temperature ranges, and a host of other design choices must be made. - 8 - Table 1- Choices to be Made in Reactor Design Fertile Neutron Fuel Material Energy Coolant Geometry Moderator U233 Thorium Fast Gas Heterogeneous Normal Water U235 Uranium Resonance Liquid Metal Homogeneous Heavy Water Pu239 Slow Normal Water Beryllium Hydro Carbons Beryllium Heavy Water Oxide etc. Carbon etc. 12. Table 1 suggests the wide range of possible design choices. All the combinations of choices listed (900 different reactors) do not make sense, and the listing is by no means a complete one. Nevertheless, perhaps as many as 100 of the combinations shown may be feasible. The reactor systems present- ly under advanced development or construction in Canada, France, the United Kingdom and the United States represent the more obvious design choices. Each system has been shown to be workable, but no system has as yet been operated as a full-scale commercial power producer. 13. The more promising systems now under advanced development or construction are briefly described below: (a) Aqueous Homogeneous Reactors -- in which a dilute solution of a salt of slightly enriched Uranium 235, Uranium 233, or Plutonium in heavy water is utilized as the fuel. In the 1/ From P/862 by A. M. Weinberg; Proceedings of the International Conference on Peaceful Uses of Atomic Energy, Vol.III, P.24. - 9 - "two region" design this solution is circulated through a tank surrounded by a larger vessel containing the blanket material, a solution or slurry of fertile material, thorium or uranium, which utilizes the neutrons escaping the core, to produce new fissionable material. A "single region" homogeneous reactor system, which would have no fertile blanket, is also proposed. (U.S.) (b) Boiling Water Reactors -- in which heat from the core is transferred by allowing the cooling water (either normal or heavy) to boil within the vessel containing the fuel elements. If normal uranium fuel is used the coolant-moderator is heavy water. (U.S.) (c) Fast Breeder Reactors -- in which an unmoderated core fuelled with plutonium or U235, and cooled, for example, with sodium is used. The core is surrounded by a uranium blanket to utilize the neutrons which escape the core. (U.K. and U.S.) (d) Gas Cooled Graphite Reactors -- in which normal uranium or uranium enriched with U235, U233, or Plutonium in the form of rods or slugs is contained in a graphite moderator. Carbon dioxide under pressure (or some other suitable gas) is circulated through the reactor to remove the heat. (France and U.K.) (e) Liquid Metal Fuel Reactors -- in which a liquid fuel com- posed of a few hundred parts per million of Uranium 235 or Uranium 233 in molten bismuth circulates through a graphite moderator. The heat is removed from the liquid metal fuel solution and used to generate steam. - 10 - A liquid blanket of uranium - or thorium - bismuth slurry\surrounds the reactor, and utilizes neutrons from the core to make fissionable material. (U.S. and U.K.) (f) Organic Moderated Reactors -- in which slightly enriched fuel is used and the coolant-moderator is a suitable organic liquid such as terphenyl or diphenyl. The system does not have to be highly pressurized to prevent "boiling", and the organic liquid, which has good moderating properties, is also essentially non- corrosive. (U.S.) (g) Pressurized Water Reactors -- in which water (or heavy water) under high pressure to keep it from boiling is circulated through a vessel or tubes containing solid fuel elements of slightly enriched or, in the case of heavy water, natural uranium. The water is then passed through a boiler in which steam is produced to drive a turbo-generator. (Canada, U.S. and U.K.) (h) Sodium Graphite Reactors -- in which advantage is taken of the high temperatures and high efficiencies to be gained through the use of sodium as the coolant. The moderator is graphite and slightly enriched fuel is used. (U.S. and U.K.) (i) Water-Cooled Graphite Reactors -- in which natural or slightly enriched uranium is used as fuel. The moderator is graphite and the coolant is light water. The system is pressurized to avoid boiling. (U.S.) - 11 - Table 2 Plans for Power Producing Reactors in Selected Countries CANADA. Atomic Energy of Canada, Ltd. and Hydro-Electric Power Commission of Ontario to build 20 Mw (electricity) prototype heavy water, natural uranium power reactor; to be in operation 1958. Designing 100 Mw reactor of similar type. FRANCE. French Five-Year Program. G-1. Contains 100 tons of natural uranium, elements 26 mm dia, 100 mm long, sheathed, in Mg. Contains 1200 tons of graphite. Air cooled at atmospher- ic pressure. Under construction, to produce about 0 Mw heat and 5 Mw electricity in 1956. G-2 and G-3. Graphite-moderated; 100 tons of natural uranium, elements 26 mm dia, 300 mm long sheathed in Mg.C02-cooled in a pressurized closed circuit. Under construction, to produce 100 to 150 Mw heat, and 30 Mw electricity. UNITED KINGDOM. United Kingdomts 10-Year_Program. (,-' 2000 Mw electric by 1965) No. of Ele ctric Comoletion Type of Reactor Reactors Output, Mw Date Gas cooled, Calder Hall type 4 ) 1960-1 Gas cooled, improved 4 ) 4o to 800 1963 Higher power reactors 4 ) 1963-4 Liquid-metal-cooled reactors 4 ) 1000 1965 Production and power generation 6 Addition to above program UNITED STATES. Heat Electric Completion AEC Program Output,Hw Output,Mw Date Notes Pressurized Water Reactor(FWR) At 230 60-90 1957 - Experimental Boiling Water Reactor (EBWR) 20 5 1957 - Sodium Graphite Reactor 20 7.5 1956 - Homogeneous Reactor Experiment No.2 (HRE-2) 10 2 1956 - Experimental Breeder Reactor No.2 (EBR-2) 62.5 15 1958 - Industrial Reactor Program Yankee Atomic Electric Co. 500 134 1960 Pressurized light- water moderator and coolant. Nuclear Power Group 692 180 1960 Boiling-water type. Power Reactor Development Co. 300 100 1960 Fast breeder. Consumer Pub.Pwr.Dist.of 250 75 1959 Sodium-graphite Nebraska reactor. Consolidated Edison 500 236* 1960 Pressurized water, uranium-thorium converter. (* 140 1Iw from reactor, 96 Mw from oil-fired superheater) 1/ Based upon the compilation Presented in Nuclear Reactor Catalogue, prepared for the United Nations by H.S.Isbin; Proceedings of International Conference on Peaceful Uses of Atomic Energy, Vol.III, P.374, et seq. - 12 - 14. Table 2 lists presently well-advanced plans for experimental, demonstration or full-scale power reactors in Canada, France, the United Kingdom and the United States. It will be noted that the gas-cooled reactor system has taken priority in development in France and the United Kingdom. This is due to several factors, the most important being the scarcity of heavy water, the scarcity or unavailability of enriched uranium or plutonium, the greater relative experience with air or gas- cooled systems, and the inherent safety of such systems. In the United States, on the other hand, no serious limitation is placed by scarcity of any material, and the systems under development are therefore more diverse. In Canada, the emphasis is on natural uranium, of which Canada has a plentiful supply. Thus, it can be seen that purely technical consider- ations are not necessarily the influencing factors in choice of reactor systems for early development and construction in any country. Indeed, the U.S. program shows that each of the systems listed above (with the exception of gas-cooled, graphite moderated thermal reactors) has its strong adherents in the U.S. From the diversity of reactors under con- sideration it is evident that not only is there a difference of opinion as to the best technical system for producing econbmic electric pow6rg butkalso work done td datess6ems to indicate that e6ch of the various auproaches is feasible and none obviously impractical. 15. One point should be stressed regarding nuclear power facilities -- their design and manufacture requires a highly skilled technology available only in the more maturely industrialized nations. Thus, particularly in underdeveloped countries, the acquisition of nuclear power plants will involve not only foreign exchange, but also access to technical information, - 13 - technical assistance and probably the execution of intergovernmental agree- ments between the supplier nation and the purchaser nation. Indeed, present U.S. law allows the export of "utilization facilities" only under and pur- suant to an executed "Agreement for Cooperation". Because of security considerations and, perhaps more importantly, trade secret considerations particularly in the field of fuel element technology, it is probable that the export of nuclear power plants will, at least in the early years, in- volve some sort of intergovernmental agreement.2 16. In sum, several different technical systems for producing nuclear power are already in an advanced stage of development or construction, and each system has been shown to be feasible, although precise economic in- formation will not become available until full-scale power reactors have been operated. Cost of Nuclear Power Facilities 17. Before discussing estimated costs of nuclear facilities, it will be well to consider general aspects of the economics of nuclear power. The actual cost of building and operating an atomic power plant will, of course, determine whether electricity from that plant is competitive with electricity from other sources. It is clear that the capital cost of a nuclear power plant today is higher than the capital cost of an equivalent 1/ In response to a question as to specifications for fuel elements used in the U.K. gas-cooled reactors, Sir Christopher Hinton said they were of the finned type and enclosed in light metal cans. He added: "I would say quite frankly that I think it reasonable to maintain a measure of what I would call industrial secrecy about fuel element design; this is the sort of thing about which one feels one may perhaps not tell one's competitors in detail". Proceedings of the International Confer- ence on Peaceful Uses of Atomic Energy, Vol.III, P.369. - 14 - thermal power plant. The estimated costs of nuclear power plants (or the "prices" at which manufacturers have contracted to supply them) now being considered for commercial development are indicated to range between "230 and ''320 per kilowatt of capacity.1/ This compares with about $120 to ',160 per kilowatt for conventional thermal plants. On the other hand, it is expected that the fuel component of power cost for a nuclear power plant should be lower than that for fossil fuels in most places, and, unlike other fuels, that cost would be relatively the same for the same reactor system no matter where the plant is located. Assuming other operating and maintenance costs not associated with fuel are about the same, the savings on fuel over the life of the plant will, then, repre- sent the total amount available to cover the presently higher capital cost and faster depreciation of the atomic power plant. 1/ See para.23. - 15 - Table 3 Early Nuclear Power Plants Canada - UK - US Estimated Estimated Reactor Elec.Cap. Cap.Cost Completion Builder Type Mw. O/Kw. Date U.K.Atomic Energy Authority Gas Cooled > 70 ,,625 Fall 1956 U.S.Atomic Energy Pressurized Commission Water >60 /-7630 1957 Consumers Public Power Dist. Sodium Graphite 75 320 1959 Atomic Energy of Pressurized Canada Ltd. Heavy Water 20 600 1959 Yankee Atomic Elec. Pressurized Co. Water 134 230 1960 Power Reactor Dev.Co. Fast Breeder 100 54o 1960 Nuclear Power Group Boiling Water 180 250 1960 Consolidated Edison Pressurized Water 236 233 1960 U.K.Central Electricity 1/ Authority Gas Cooled > 150 .,*2 80- 1960-1961 1/ At the Geneva Conference, a cost of '350/Kw was given for a 150 Mw plant. Later estimates of larger plants place the cost at about ":280/Kw. Firm data will be available later this fall after ten- ders are received by the C.E.A. from manufacturers. 18. Table 3 lists nine nuclear power reactors now under construction or in advanced design stage. No cost information is available on gas-cooled French power reactors, but it can be assumed that their costs would be no less than those for the U.K. gas-cooled reactors. The U.K.AEA gas-cooled reactor will be the first central station power reactor to come into - 16 - operation. It is a dual-purpose plant, in that it is so designed as to permit the simultaneous production of power and of military grade plutonium having a low:percentagefof the Pu-2L0,iso epe. -Its successors, however, tthe gas-cooled U.K.CEA plants, will be designed with major emphasis on the production of power. The estimate of Q280/Kw for the large second generation gas-cooled reactors appears reasonable. 19. The first large U.S. power reactor is the Pressurized Water Reactor being built for the U.S. AEC by Westinghouse. This reactor, as are the Yankee Atomic Electric Co. and the Consolidated Edison reactors, is an outgrowth of U.S. experience in designing and building the propul- sion reactor for the USS Nautilus, and the system has been shown to be satisfactorily operable under the severe conditions encountered in forced submerged operation. The estimates given for the Yankee and ConEd re- actors are based upon extensive technical experience gained in building the Nautilus and PWR reactor and appear, therefore, to be reasonable, al- though they may contain subsidies from the manufacturers, since they were both contracted for after spirited competition among the leading electrical equipment and boiler manufacturers.1 The Consumers Public Power District sodium-cooled graphite reactor is based on extensive sodium handling technology derived from North American Aviation's dodium reactor experiment, the AEC's Experimental Breeder Reactor and its Submarine Intermediate Reactor, all of which use liquid sodium or sodium-potassium alloy as 1/ The ConEd nuclear reactor, for example, is budgeted at 5,ooo,coo. Babcock & Wilcox has bid to build the nuclear steam generating portion for a fixed sum which represents about one-third of the budget total. The remainder is for site improvement, turbines, generators and other standard equipment. - 17 - coolants. The cost estimate appears reasonable. The Nuclear Power Group boiling water reactor is also based on considerable U.S. AEC experience. In this case, the General Electric Company has bid to build the entire reactor and power generating equipment for $h4,000,000. It may be assumed that the General Electric Company believes it reasonable to expect that this or early successor boiling water reactors can be built profitably by a manu- facturer for "250/Kw. The other U.S. reactor system listed in Table 3 is the Fast Breeder Reactor, and represents the greatest extension of present technology, which fact is evident in the higher estimated cost per kilowatt. However, this reactor could have essentially zero net fuel costs, and possib- ly may have negative net fuell/costs. Its higher capital cost may therefore be justified economically. 20. The Canadian Nuclear Power Demonstration (NPD) pressurized heavy water reactor is more in the nature of an experimental than power reactor, although it will have a significant electrical output. It is estimated that "second generation" pressurized heavy water reactors of this design will cost about 1?250/Kw and will have capacities in excess of100 Mw electric. 21. On Chart 1 are plotted the ;,/Kw costs of these nine reactors accord- ing to the year each is estimated to come into operation. The reactors plotted in black solid dot (*) are those of an advanced type which involve a substantial extension of present technology and those that are the "first" or prototype reactors. Plotted as an open dot (o) are the reactors based 1/ Assuming there is a demand for the excess plutonium or U233 generated above and beyond the needs of the producing reactor. ESTIMATED COSTS OF POWER REACTORS ($/KW ELECTRIC CAPACITY) 700 7 700 600 600 94 500 500 400 400 .320 5 300 300 2306 200 200 I OO- ]- I- OO O1 0 1955 1956 1957 1958 1959 1960 1961 0 I. UKAEA FIRST PROTOTYPE COMMERCIAL 0 5. SGR SCALE-UP TO COMMERCIAL 2. USPWR FIRST PROTOTYPE COMMERCIAL 6. PWR, YANKEE ELEC.; WESTINGHOUSE 610 3. NPD FIRST CANADIAN HEAVY WATER PROTOTYPE COMMERCIAL 7. BWR, NUC. PWR. GRP.; GEN. ELEC. BID 4. FBR FIRST COMMERCIAL SCALE BREEDER (US) B. PWR, CON. ED.; BABCOCK E WILCOX BID 9. UKCEA GAS COOLED POWER STATION IBRO- Economic Staff H115 ESTIMATED COSTS OF POWER REACTORS 700 700 *2 600 * 600 )4 500 500 200 -400 0 C- 1 00 I4OO I-320 5 0 300 300 wL 7 230 U EY E 64 200 200 100 1100 0. - 0 50 100 150 200 250 300 MW ELECTRIC CAPACITY I.CANADIAN HEAVY WATER REACTOR 0 5. SGR SCALE-UP 2. USPWR PROTOTYPE 6. PWR, YANKEE ELEC. 3.UKAEA PROTOTYPE 7. UKCEA COMMERCIAL STATION 4.FBR COMMERCIAL SCALE BREEDER (US) 8, BWR, NUC. PWR. GRP. 9. PWR, CON. ED. IBRD - Economic Staff 16 III6 - 18 - on substantially developed technology or upon which firm bids have been obtained from manufacturers. For this latter class, which can be taken to represent the early commercial reactors, the 1;/Kw costs range between .230 and $'320. 22. On Chart 2 are plotted the capital costs of these reactors in relation to their electric capacity in megawatts. As the size of the plants approaches about 100 Mw, the cost per kilowatt tends to become asymptotic, in this case at about 230/Kw. 23. In sum, the cost of nuclear power plants of about 100 Mw electric capacity and upwards can be reasonably taken as between .230/Kw and 320/Kw. In the computations of cost of nuclear power which appear later in this paper, the capital cost will be set at ;250/Kw. It should be noted that the reactors upon which this range of estimated costs is based are being built in Canada, the United Kingdom and the United States. This estimate is conservative and actual capital costs may well be less.1/ Whether the same costs would obtain elsewhere would depend on local conditions. At least a portion of the added capital costs for export reactors (transportation, imported expert labor, etc.) might be partly balanced in some cases by lower general labor costs reflected in somewhat lower costs of standard construction such 1/ In-recent testimony before'the U.S.Joint Committee on Atomic Energy, Mr. Philip Sporn, President of American Gas and Electric, stated that the NPG boiling water reactor, which was estimated to produce 180,000 Kw at a capital cost of $250/Kw, would actually have a larger electric capacity and thus its capital cost would be'less than '250/Kw. Also, the PRDC breeder reactor is designed at 100,000 Kw electric capacity, but it is expected to have approximately 150,000 Kw electric capacity after operating experience is gained. Thus, its capital cost per Kw will be lower than indicated. - 19 - as buildings, excavation, and perhaps lower costs for turbine generators or other standard equipment obtainable locally. The Estimated Cost of Nuclear Power 24. Having determined on a conservative basis a reasonable capital cost for a 100 Mw nuclear power plant built on essentially today's technology, it is now necessary to establish costs for operating and maintenance, fuel, depreciation, and return on investment. Since the plant factor substantially affects the mills/Kwh charges attributable to depreciation, we must establish the average plant factor at which the nuclear plant might be expected to operate over its lifetime. Operating and Maintenance Costs 25. Since no commercial nuclear power reactors have been as yet operated, there is no firm experience upon which to base estimates of operating and maintenance costs per Kwh produced. Most studies on the economics of nuclear power reactors, however, indicate that such costs should be close to those for normal thermal stations. It is to be expected that for the first year or two of operation of early nuclear power plants, operating and maintenance costs may be higher, perhaps double those for standard plants; the basic simplicity of operation of a reactor, however, would seem to permit these costs in later years to be about the same as for conventional plants. Second or third generation plants should experience operating and maintenance costs about the same as for a conventional thermal station. In any event, this component of cost of electricity is not very large -- for large central station plants in the United States it adds about 0.5 to 1.0 mill/Kwh. 26. There is no reason to expect that the labor force required for a - 20 - nuclear power station will be substantially different in respect of cost than for a conventional station. Early plants can be expected to have somewhat higher than normal costs in first years of operation but later plants should have the same or possibly lower costs. 27. Kost studies on nuclear power economics assume little or no varia- tion from standard maintenance costs. This point of view is reasonable if one examines the components of a nuclear power plant. The turbo-generator side is standard, and the only increase in maintenance would be due to low level radiation from the primary steam or from contamination of the secondary coolant. The coolant and steam systems will be required to be built to very close specifications, which is reflected in the higher capital cost, but should also be reflected in low maintenance charges. In the re- actor itself, there are problems of radiation damage, and of corrosion, both of which are given primary consideration in the design and should be expected therefore not to cause large maintenance problems. 28. The average of the estimated costs reported in 16 studies of the economics of nuclear power at Geneva for the combined operating and mainten- ance cost for a large nuclear power station is 2.02 mills/Kwhr. This value, which is between 2 and 4 tines the cost derived from U.S. experience with large conventional thermal stations, is certainly reasonable, and probably higher than will be actually experienced. Its use in this paper is there- fore considered as conservative. 29. Insurance costs for a nuclear plant will undoubtedly be higher than for a conventional thermal plant. However, the cost of insurance for a conventional plant is such a small percentage of the total operating costs - 21 - that even a substantial increase in insurance cost would not materially affect the total cost of power generated. It may be well, however, to indicate the nature of the insurance problem presented by a nuclear power station. The problem falls into two areas: insurance against loss of the plant itself and insurance against third party liability in the event of a major nuclear incident, admittedly an event of extremely remote probability, but which might cause grave dislocation of normal life to many persons as a result of radioactivity that may be released. These risks can be minimized by completely enclosing the reactor in a containment vessel, by locating the plant in a less densely populated area, and by design of additional safety features in the plant itself. If the plant is government-owned, it is "self insured". If the plant is privately owned, insurance will undoubted- ly have to be obtained. In the United States insurance companies have indicated willingness to underwrite "60,000,000 in coverage for each re- actor. There is a strong possibility that a larger-amountc6f.coveragewwill become available as the insurance companies investigate the problem further. However, it appears that insurance against a major catastrophe, involving claims of perhaps a hundred million dollars or more, may require some form of government acceptance of liability in the United States. Solution of the insurance problem will undoubtedly vary in different countries, and should be examined in each particular case. It is expected that the cost of insurance will nevertheless still represent a small proportion of the cost of power. The 2 mills/Kwh operation and maintenance cost used in this paper is considered sufficiently liberal to cover the cost of insurance. Fuel Costs 30. A key factor in achieving economically competitive nuclear power is - 22 - the cost of the nuclear fuel, including fabrication, value of new fission- able material made, reprocessing, and inventory charges. This cost is dependent upon the type of reactor, the amount of fissionable material - both originally charged and newly made - that is effectively utilized for power production, the particular fuel reprocessing cycle, the uses to which by-product fissionable materials can be put, and the total inventory of fuel (i.e. fuel awaiting loading, fuel in the reactor, and fuel in repro- cessing plants). The economic feasibility of nuclear power is directly related to the success with which efforts to minimize the fuel component of cost is met. Thus, while it is possible today to make reasonable and probably conservative estimates as to the cost properly chargeable to fuel, the economic feasibility and relative attractiveness of different reactor systems will be conclusively demonstrated only as actual operating experi- ence is gained, not only with the reactor but also with ancillary metallurg- ical and chemical facili'ies. In this regard, intergovernmental agreements as to fuel prices, fabrication and reprocessing of fuels and inventory charges can be expected to have a major bearing on the feasibility of competitive nuclear power. 31. At present all supplies of nuclear fuels are government-controlled and costs of production have not been released. Prices at which the United States will sell or lease certain materials for research reactors were released at Geneva -- $P,000 per kilogram of uranium at a U235 enrichment of 20%, MO per kilogram for natural uranium in metal form, and )61.50 per kilogram for heavy water. Prices for quantities of these materials for use in commercial power stations have not yet been established, or at least have not yet been announced. Most studies estimate values for - 23 - U23$ at between &15 and t30 per gram, and for natural uranium at about $6040 per kilogram. In this paper a value for U23$ of t2S per gram and a value for natural uranium metal of ,!0 per kilogram has been used as a basis for calculation. 32. The fuel costs of any reactor will depend upon a large number of variables and must, of course, be calculated individually for each particular reactor system. However, it has been found possible to develop general expressions for nuclear fuel and inventory costs that are at least indicative of the range of costs to be expected for certain reactor systems. Such expressions and sample calculations for various reactor systems are con- tained, for example, in a report entitled "Nuclear Fuel and Inventory Costs for Power Reactors" (LRL-138) by D. Kallman, R. A. Pierce, and W. S. Scheib, Jr., of the California Research and Development Company, dated June 1954 and prepared under U. S. Atomic Energy Commission Contract No.AT(11-1)-74. 33. Basically, in a reactor some atoms of a fissionable material are fissioned or otherwise destroyed, some are left in the fuel elements, and some new atoms of fissionable material are formed. There are involved, in computing the fuel component of power cost, certain metallurgical, fabrica- tion, reprocessing and waste disposal costs in addition to the values of the fissionable material destroyed or made. Since fissionable material in the reactor fuel cycle represents a large amount of immobilized capital, 1/ In connection with the announcement of Feb.23, 1956 that the U.S. would make available 20,000 kilograms of U23$ for use in power reactors in other countries, Chairman Lewis L. Strauss of the U.S. AEC said the material had a value of Q$50,000,000, which is consistent with the 125.00 per gram value proposed for use in this paper. inventory charges are made against the fissionable material held in the fuel cycle.-/ The following expressions can be written to reflect these charges: (a) Nuclear Fuel Cost: For Heterogeneous Thermal Reactors: Fuel Cost = f.m. destroyed + f.m. undestroyed - Pu recovered + Fabrication + Reprocessing + Waste Disposal. (b) Inventory Cost for Thermal Reactors: Inv.Cost = f.m. feed in reactor + f.m. Feed in Separations Cycle. 34. The assumptions used in the calculations contained in LRL-138, re- ferred to above, are not unreasonable, and it is not considered necessary here to describe them fully. Let us now consider the costs for fuel and inventory as calculated in LRL-138 for a thermal regenerative reactor using normal uranium as fuel: Table 4 Quantities in grams per Unit Costs Fuel 1000 Kwh. per gram Mills/Kwh. Fissionable material destroyed 0.20 $ 6.20 1.24 Fissionable material undestroyed 0.20 6.20 1.24 Plutonium recovered 0.16 25.00 (4.00) Fabrication 56.0o 0.005 0.28 Reprocessing 56.00 0.015 0.84 Waste Disposal 0.24 0.50 0.12 Inventory Charges 0.65 Total Fuel and Inventory Costs 0.37 1/ See para.38. - 25 - It will be noted that the plutonium produced in the reactor is valued at $25.00 per gram. This figure is about twice as high as that generally assumed in the Geneva papers. If a value of '15.00 is substituted, the resulting overall fuel and inventory cost would be 1.97 mills/Kwh. The fissionable material, destroyed and undestroyed, is valued at $6.20 per gram which represents normal uranium at 2LL per kilogram. Since our value for normal uranium is 'tO per kilogram, the fissionable material destroyed and undestroyed should be valued at $,.62 per gram. The cost of fission- able material destroyed then becomes 1.13 mills/Kwhr. The other principal item in these calculations subject to serious error is the amount of fissionable material undestroyed. The amount of fissionable material undestroyed is dependent upon the length of time the uranium fuel element can be left in the reactor without being seriously damaged, or without poisoning the chain reaction through buildup of fission products. The cost attributable to fissionable material unburned assumed in the above calculation requires irradiation of about 4,500 megawatt days per ton, which is higher than has been achieved on any large scale to date. If irradiation of about 3,00 MWAD/ton, which has been achieved in Canadian and other natural uranium reactors, is assumed, the direct cost of fissionable material unburned at $5.62 per gram would be about 1.70 mills/Kwh. Because more material would have to be fabricated and reprocessed, these costs and the inventory charge would also rise, making the total fuel and inventory cost about 2.90 mills/Kwh. 35. Fuel and inventory costs for sixteen reactors (not all thermal, natural uranium systems) were estimated in various Geneva papers. The average of these costs is about 2.85 mills per kilowatt hour. The spread of such fuel costs, depending upon value assigned to plutonium either as a - 26 - by-product or as replacement fuel, the irradiation level to which the uranium fuel elements may be subjected, and the fuel cycling chosen, depends upon the reactor system employed. For breeder reactors the fuel costs could become zero, and even may show as a credit. For thermal reactors in which irradiations of 8-10,000 ITWD/ton can be achieved, the fuel and inventory cost can be perhaps as low as 0.5 mill/Kwhr. These latter systems are, however, not achievable with present technology; they may be expected in third or later generation plants. 36. It should be noted that our calculation of 2.90 mills/Kwhr as the fuel component of power costs assumes no substantial extension of present technology and no increase in efficiency of use of nuclear fuel over the whole life of the power plant. It is more reasonable to expect some in- crease in efficiency, or decrease in unit costs, as experience further is gained.l/ If a modest allowance of 10 to 15 percent of the calculated cost is taken, the fuel cost component would be about 2.50 mills/Kwhr. 37. In summary, a reasonable estimate of fuel and inventory costs for early commercial nuclear power reactors such as presently under design or construction appears to be about 2.5 mills per kilowattthour over the life of the plant. It must be emphasized that the reasonable value suggested above is a generalized value. For any specific reactor system, a specific fuel and inventory cost must be calculated using the specific parameters of the specific system. 1/ In the foregoing calculations, a thermal efficiency of 20 percent has been used. Mr. Philip Sporn, President of American Gas and Electric, has reported that the NPG Boiling Water Power Reactor is expected to achieve a thermal efficiency of 28 percent. - 27 - 38. The foregoing estimate of 2.50 mills/Kwh includes approximately 0.9 mills/Kwh to cover inventory charges on the immobilized capital represent, ed by the fissionable and fertile material committed to the nuclear power re- actor, including not only all the nuclear material in the reactor, but also that being fabricated into fuel elements, being stored for cooling after irradiation, and being chemically processed. The amount of material so committed will depend upon the specific reactor. According to Dr. W. K. Davis, irector of the U.S. Atomic Energy Commissionts Division of Reactor Development, the total value of the inventory of nuclear fuel may be as high as $50 per electrical kilowattifor some reactors, and in typical heterogeneous reactors the value will be in the range of $20 to ch0 per kilowatt. In homogeneous reactors the value may be appreciably less.1/ Accepting the higher value, i.e. !'0 per kilowatt, a 100 Mw reactor might require something like 15,000,000 worth of nuclear fuel in inventory. The allowance of 0.9 mills/Kwh, which has been included in the fuel and inventory charge of 2.50 mills/Kwh, would, in a 100 Mw plant operated at a 50 percent plant factor, provide income of about !395,000 per year, an amount which should be ample to cover financial charges on carrying an even greater fuel inventory. Amount and Pattern of Use 39. Since the capital costs for a nuclear power station are higher than for a comparable thermal station the amounts charged in the selling price to cover return on that capital investment and depreciation are larger than would be required in the case of a thermal plant. In order to bring those 1/ P/477 - "Capital Investment Required for Nuclear Energy" by W.K.Davis; Proceedings International Conference on Peaceful Uses of Atomic Energy. - 28 - charges within reasonable limits, it is necessary to spread them over as large a number of units of production -- kilowatt hours -- as possible. This indicates the need to operate nuclear power plants at as high a plant factor as possible, or, in other words, to use them as base load stations. LO. Since we have also seen that the capital investment would probably be sharply higher for plants of less than about 100 Mw capacity, we must be sure that the system into which the nuclear plant is to be integrated is capable of accepting a 100 Mw plant as a base load installation with a high plant factor. 4l. Whether the nuclear plant can be operated with a high plant factor is essentially a function of system demand and the operating and fuel costs of the nuclear plant in relation to conventional plants in the system. (We assume that operational shutdowns will be no greater in a nuclear plant than in a thermal plant.) Therefore, in assessing the feasibility of a 100 Mw nuclear plant from an economical point of view, it is paramount that we make sure the electrical system, of which it is to be a part, is such as to allow the nuclear plant to be operated at a high plant factor throughout the year. Depreciation 42. For a conventional thermal plant, the plant life will be taken as 33 1/3 years, and the depreciation rate will be 3% of the capital cost per annum on a straight line basis. For the nuclear plant, about.30% or 40% of the capital cost is represented by conventional equipment which would have the same life as in a conventional plant, i.e., 33 1/3 years. The remaining - 29 - 60% to 70% of' the capital cost comprises non-standard, specialized items for which a plant life of 20 years will be assumed. This would appear to be a conservative estimate3 the early large size nuclear reactors at Hanford, Washington which began operation about twelve years ago, are still perforiing satisfactorily as is the smaller air cooled reactor at Oak Ridge, which began operation somewhat earlier. The combined over- all plant life of the nuclear plant will therefore be taken as 25 years, and the plant will be depreciated at 4% of its capital cost per annum on a straight line basis. 43. The amount of money allocated each year to cover depreciation (four percent of the capital cost) would, if placed in a sinking fund-at, say, 4.75 percent interest, actually permit the complete writing off of the plant in 17 years, or about 8 years before the technically determined life of the plant had expired. The use of a 4 percent, straight line de- preciation schedule is considered acceptably prudent from a banking point of view. 44. The question of obsolescence of a nuclear power plant should be mentioned at this point. In a conventional thermal plant the primary reason for obsolescence is that new plants display a consistently higher thermal efficiency and consistently lower operating costs, and since the cost of operations and fuel are together the most significant portion of selling price of electric power at the bus bar in conventional thermal plants, the new plants are able to make power cheaper than old ones. It is to be expected that later nuclear plants will undoubtedly be able to produce electricity at lower fuel costs than the earlier plants. However, the older plant will also benefit from advancing technology and in some - 30 - cases will be able to show comparable decreases in fuel costs in later loadings, as new alloys or new methods of fabrication enable a larger burnup of the fissionable material in the core. As was pointed out earlier, the fuel component of cost of electricity generated in a nuclear power station should be significantly less than the fuel component of cost of electricity generated in a conventional thermal station. Thus, from the operating and fuel cost point of view, the problem of obsolescence of a nuclear power plant is more comparable to that of a hydroelectric station thanito W'conventional-thermal station. 45. As was mentioned earlier, the number of hours per year that the nuclear plant will generate power (i.e. the plant factor at which it will operate) has an important bearing on the cost of power generated. A study conducted by the General Electric Company of modern conventional thermal power stations in the United States revealed that over the life of these plants a plant factor of h3% was achieved. The G.E. study suggests that nuclear stations, because of lower fuel costs, should achieve a lifetime plant factor of 50%. On the other hand, the U.K. Central Electricity Authority believes early nuclear plants will operate with a lifetime plant factor of 75 percent. Because the fuel and operating component of cost of the nuclear plant will be lower than such costs in a thermal plant, it is not unreasonable to expect that the nuclear plant, in competition with thermal plants, will be operated in later years of its life at a much higher rate than would a thermal plant of equivalent age (see comments on obsolescence, Para.h). The writer tends to favor the view of the UKCEA and believes a nuclear plant "competing" with conventional thermal plants will achieve an overall lifetime plant factor of well over 60', assuming the system demand is not a limiting - 31 - factor. Even in competition with newer nuclear plants, the older nuclear plant may be expected to be operated at a plant factor of about 5O/0. Generating Costs for Nuclear Power, excluding Financial Charges 46. Table 5 contains calculations of the cost of generating nuclear power (including depreciation) in a 100 Mw nuclear plant costing .250/Kw of electric capacity operated at various plant factors. It represents generation costs in a nuclear plant that might be constructed on essentially todayts technology. The fuel and inventory cost (titled "Nuclear Fuel" in Table 5) is that previously computed.as conservative, and may in practice decrease over the life of the plant to about 1.0 mills/ Kwh. It is maintained as a constant in Table 5 regardless of plant factor, since the only effect of increasing the plant factor is a slight decrease in the inventory component of the charge. The operation and maintenance cost used in Table 5 is two to four times that experienced in conventional thermal stations in the United States. Over the life of the nuclear plant, it is expected that this cost might decrease to perhaps 1.0 mills/Kwh. Also, this cost will in fact decrease with an increase in plant factor. However, the 2.0 mills/Kwh cost is maintained as a constant in these calculations in order to introduce again a conservative bias. The depreciation costs shown in Table 5 are calculated on a straight line basis at 4 percent per annum, based upon a plant life of 25 years. However, as was pointed out in paragraph 43, if the depreciation allocation each year is invested at, say, 4.75 percent interest, the plant will in fact be completely written off in 17 years. Once again, this gives a conservative bias to the calculations. - 32 - Table 5 Cost of Generating Nuclear Power at Various Plant Factors (Excluding Return on Investment) Expressed in U.S.Mills per Kilowatt Hour Plant Factor 5o% 60% 70% 80% 90% Nuclear Fuel 2.5 2.6 2.5 2.5 2.5 Operation and Maintenance 2.0 2.0 2.0 2.0 2.0 Depreciation 2.3 1.9 1.6 1.h 1.3 Generating Cost (excluding Return on Investment) 6.8 6.4 6.1 $.9 5.8 Total Cost Including Return on Investment 47. In view of the larger capital investment required for a nuclear power plant as compared with a conventional thermal station the burden of financial charges is of major importance in evaluating "where and under what circumstances" nuclear power may be economically attractive. As was pointed out earlier, the average lifetime plant factor at which the power plant will be operated also substantially affects the cost of electricity generated. Table 6 shows estimated cost of electricity generated in a nuclear plantrat various financial charges and at various plant factors. 1/ The term "return on investment" includes interest on debt, dividends on equity if any, and income and ad valorem taxes if any. Table 6 Total Cost of Generating Nuclear Power at Various Plant Factors and at Various Returns on Investment (Expressed in U.S.Mills per Kilowatt Hour) Plant Factor Return on Investment 3% 1f 5% 6% 7% 8% 9% 10% 11% 12% 13% 16% 15% Generating Cost 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 Financial Cost 1.7 2.3 2.9 3.4 4.0 4.6 5.1 5.7 6.3 6.8 7.4 8.0 8.6 Total 8.5 9.1 9.7 10.2 10.8 11.4 11.9 12.5 13.1 13.6 14.2 14.8 15.4 60o Generating Cost 6.4 6.4 6.4 6.4 6.4 6.4 6.4 6.4 6.4 6.4 6.h 6.4 6.4 Financial Cost 1.4 1.9 2.4 2.9 3.3 3.8 4.3 4.8 5.2 5.7 6.2 6.7 7.1 Total 7.8 8.3 8.8 9.3 9.7 10.2 10.7 11.2 11.6 12.1 12.6 13.1 13.5 Generating Cost 6.1 6.1 6.1 6.1 6.1 6.1 6.1 6.1 6.1 6.1 6.1 6.1 6.1 70%o Financial Cost 1.2 1.6 2.0 2.4 2.9 3.3 3.7 4.1 4.5 4.9 5.3 5.7 6.1 Total 7.3 7.7 8.1 8.5 9.0 9.4 9.8 10.2 10.6 11.0 11.4 11.8 12.2 Generating Cost 5.9 5.9 5.9 5.9 5.9 5.9 5.9 5.9 5.9 5.9 5.9 5.9 5.9 Financial Cost 1.1 1.4 1.8 2.1 2.5 2.9 3.2 3.6 3.9 4.3 4.6 5.0 5.4 Total 7.0 7.3 7.7 8.0 8.4 8.8 9.1 9.5 9.8 10.2 10.5 10.9 11.3 Generating Cost 5.8 5.8 5-8 5.8 5.8 5.8 5.8 5.8 5.8 5.8 5.8 5.8 5.8 90% Financial Cost 1.0 1.3 1.6 1.9 2.2 2.5 2.9 3.2 3.5 3.8 4.1 4.4 4.8 Total 6.8 7.1 7.4 7.7 8.0 8.3 8.7 9.0 9.3 9.6 9.9 10.2 10.6 - 34 - Comparative Costs of Nuclear and Conventional Thermal Power 48. We are now in a position to compare the cost of electricity generat- ed in a nuclear plant with the cost of electricity generated in a. convention- al thermal station. The conventional thermal station has lower operating and maintenance costs -- about 0.8 mills/Kwh on the average in the U.K. and U.S. -- in contrast with the 2.0 mills/Kwh cost used in this paper for the nuclear plant. Since the capital investment for a conventional thermal station has been taken as j120/Kwl as compared with the !20/Kw used in this paper for the nuclear plant, and since depreciation has been set at a more rapid rate for the nuclear plant (4 percent per annum, straight line, as contrasted with 3 percent per annum, straight line, for the conventional thermal station), depreciation and financial charges will also be higher for the nuclear plant than for the conventional thermal station. On the other hand, the fuel component of cost of electricity generated in a nuclear power plant will be considerably less than the cost of fuel used in a conventional thermal station, except in low-cost fossil fuel areas. 49. The relative attractiveness of the nuclear plant vis-a-vis the conventional thermal station is largely dependent upon the cost of fuel for the conventional station and upon the rate of financial charges appropriate in the specific location. Charts 3 to 7 compare the cost of electricity generated in a nuclear power plant of 100 Mw electric capacity built on essentially today's technology, with the cost of electricity generated in a conventional thermal station of similar capacity, at various fuel costs for 1/ Conventional thermal stations of 100 Mw are estimated to range from V 120/Kw to '160/Kw. The $120/Kw figure used in these calculations favors the conventional thermal stations. COMPARATIVE COST OF ELECTRICITY NUCLEAR AND CONVENTIONAL THERMAL AT VARYING COSTS OF FOSSIL FUEL (50% PLANT FACTOR) 20 I NUCLEAR (I ...7.5 FUEL CONVENTIONAL THERMAL AT VARIOUS FUEL 0 COSTS i 0 5 MO - 50 10 I % RETURN ON INVESTMENT 0 IBROD- Economic Staff L111 COMPARATIVE COST OF ELECTRICITY NUCLEAR AND CONVENTIONAL THERMAL AT VARYING COSTS OF FOSSIL FUEL (60% PLANT FACTOR) 20, O ul -j I -j 7.5 FUEL CONVENTIONAL THERMAL AT 5.5 FUEL VARIOUS FUEL 0d1 t COSTS LLI 4.5 FUEL - O 0 0 5 UD, mO 0 5 10 15 % RETURN ON INVESTMENT IBRD - Economic Staff 1119 COMPARATIVE COST OF ELECTRICITY NUCLEAR AND CONVENTIONAL THERMAL AT VARYING COSTS OF FOSSIL FUEL (70% PLANT FACTOR) 20M -J 15 NUGLEAR 7.5 FUEL CONVENTIONAL 6.5 FUEL THERMAL AT IL 101... .. - *" VARIOUS FU EL j COSTS 0 Z) 0 5 10 15 % RETURN ON INVESTMENT IBRD - Economic Staff 1120 - COMPARATIVE COST OF ELECTRICITY NUCLEAR AND CONVENTIONAL THERMAL AT VARYING COSTS OF FOSSIL FUEL (80% PLANT FACTOR) 20 I -J 15 _ NUCLEAR -c- - 7.5 FUEL 1 0. CONVENTIONAL S 10 6.5U THERMAL AT 5.5 FUEL VARIOUS FUEL J COSTS 5 4 10 15 % RETURN ON INVESTMENT C IBRD-Economic Staff 2 I121 COMPARATIVE COST OF ELECTRICITY NUCLEAR AND CONVENTIONAL THERMAL AT VARYING COSTS OF FOSSIL FUEL (90% PLANT FACTOR) 20 1 W 0 _j NUCLEAR --7.5 FUEL CONVENTIONAL j -.THERMAL AT uj 5 F - - - VARIOUS FUEL -L -COSTS O- 0: 5 5 10 15 % RETURN ON INVESTMENT IBRD - Economic Staff 1122 - - 35 - a conventional plant. Chart 3 (at a 50o plant factor) shows that wherever fossil fuel cost is 6.5 mills per Kwh or less (equivalent to $10.15 per metric ton for 10,000 BTU per pound coal burned in a plant having a 35% efficiency), a conventional thermal station is more economical than a nuclear station, irrespective of what rate of return is attributed to the investment in the plant. On the other hand, if the fossil fuel cost is 7.5 mills per Kwh (0$17.00 per metric ton for 10,000 BTU per pound coal burned in a plant having a 35% thermal efficiency) a nuclear plant could afford to pay up to 71% return((after depreciation) on the investment and still produce electricity more economically than a conventional thermal plant. 50. At a 9% plant factor (Chart 7), again a nuclear station could not compete with 46. mills/Kwh fossil fuel. It could compete with 5.5 mills/ Kwh (12.3 per metric ton for 10,000 BTU per pound coal burned at a thernal efficiency of 35%), fuel at a 61% return or less. If cost of fuel were 6.6 mills/Kwh ('14.77 per metric ton for 10,000 BTU per pound coal burned in a plant having a 35% thermal efficiency) a nuclear plant could afford up to 122L on investment and still be more economical. CONCLUSIONS 51. It is concluded that a nuclear power station having an electrical capacity of 100 Mw or larger could be designed and built on essentially present technology. In certain locations such a plant would have a high degree of probability of producing electricity at costs competitive with those of electricity produced from fossil fuels. It appears possible to establish circumstances that would have to be met in order for it to do so today- (a) The generation and distribution system into which the - 36 - nuclear plant is to be integrated must be large, capable of accepting a 100 Mw plant at a high plant factor. (b) The nuclear plant would have to be located in a country with relatively high fossil fuel costs, and with sufficient availability of capital so that the return on investment in the plant could be moderately low. (c) The country must have executed whatever intergovernmental agreements that are necessary to assure a continuing supply of fuel at prices consistent with thosd used in this paper, reprocessing, and, if necessary, the import of components. (d) The country must have a degree of economic stability so that if the nuclear plant should cost more than expected or should not perform as anticipated, the excess cost could be absorbed without a significant adverse effect. (e) Until further operational experience has been obtained, it would not be prudent to establish the nuclear plant in a system where it would represent a considerable proportion of the total system generating capacity. APPENDIX I HEAT, PROPULSION, AND SMALLER POWER NUCLEAR REACTORS Heat Reactors 1. The considerations that have been discussed relative to power reactors apply to the process or space heat reactors, and to propulsion reactors with some modification. The capital cost of a process heat re- actor wouldprobably be less -- perhaps as much as 20 or 30 percent less -- than for an electric power reactor, since the turbo-generator side of the plant would be essentially eliminated. On the other hand, the problem of finding a suitable system or plant to utilize the large amount of heat pro- duced is limiting. 2. The effect of size on process heat reactor costs would be similar to that noted in the case of the nuclear power reactors, and there will be a size below which the process heat reactor will steeply rise in cost per unit output. Just what that range of size will be is yet to be determined. In sum, as concerns process or space heat reactors, no calculations or analyses of their economics have been published; thus, while that application is of interest to the Bank, it appears premature to attempt to arrive at judgements as to its economic feasibility. It should be noted that Sweden is planning an experimental space heating reactor of about 90 Mw thermal capacity to be completed around 1960. This reactor is planned to provide space heating to portions of the City of Vasteras (population about 65,000). Norway is also considering an experimental industrial heat reactor for use in conjunction with a wood processing plant at Halden. This reactor would have a thermal capacity of 10-12 Mw and would begin operation in about three years. It is expected to provide about 20-25% of the plantfs hourly steam requirements. Detailed information on these reactors, and the estimated economics of their -2- operation, should begin to be available for analysis soon. Propulsion Reactors 3. As for propulsion reactors, again the field looks interesting from an economic point of view; the information on it published today is not, however, sufficient to form the base of any judgements on its practicability. The first application of a reactor to propel a vehicle is in the atomic submarine "Nautilus". The Nautilus is powered by a pressurized water reactor and because the reactor needs no oxygen to support "combustion", she has in effect an unlimited range at very high speed submerged at great depth. Conventionally powered submarines, on the other hand, are severely limited in such a situation, being capable of only about an hourts operation at high speed when deeply submerged. In comparison, the U. S. Navy has announced that the Nautilus cruised over 1600 miles at an average speed of 16 knots submerged at depth. The relatively unlimited underwater range of the Nautilus has been likened in significance to the development of ironclad naval vessels in that it will demand a revolutionary change in naval tactics both defensive and offensive. 4. The application of nuclear power to the propulsion of naval vessels is only in its infancy, but already a half dozen nuclear submarines are being built and the Navy and AEC are beginning work on a land-based prototype of a large surface ship reactor. In these naval applications, however, the cost of propulsion is secondary to performance and minimum displacement. A higher cost per mile or per hour can be tolerated because of the unique performance of the nuclear propelled ship. 5. In the case of commercial ship propulsion, however, costs must be con- sidered. The higher nuclear costs tend to make commercial ship propulsion - 3 - less attractive, at least insofar as the U.S. is concerned. Work is going on both in the U.S., the U.K., and Norway, however, to develop a practical and economic reactor system for merchant ship propulsion. Such applications, however, are much farther from realization than the propulsion of military ships where cost is a minor consideration. Small Power Reactors 6. The discussion that has preceded has concerned the feasibility of building 100 Mw or larger nuclear power reactors which might produce electricity at costs competitive with conventionally fuelled thermal power stations in some situations. The development of small nuclear power stations, suitable for use in remote locations such as the Arctic or in underdeveloped countries where the demand for electricity occurs in relative- ly small units, has not progressed as far as has the development of larger central station units. Work is going on, particularly in the United States and Canada, to develop reactor systems for such smaller, specialized uses. As a rough estimate, if nuclear power can be produced for 20 to 30 mills/Kwh in plants having capacities of 3 to 5 Mw electric, there would be a consider- able demand for such reactors in the remote areas of Canada, for example, in some areas of Africa, and undoubtedly in Asia and South America. To make 20-30 mills/Kwh nuclear power, the capital cost for the reactor would probably have to be no more than about I600/Kw. This implies a production rather than a custom scale manufacture. As of today, however, it is not possible to evaluate the economic feasibility of building a reactor in the 3-5 Mw range. 7. For reactors in the 10 to perhaps 30 Mw range, some estimates have been made. There, the capital costs might be about '600/Kw, and the fuel and inventory costs perhaps 9 mills/Kwh. The operating and maintenance costs might be about 2 mills/Kwh. If a 10% return on investment is assumed, and a plant factor of 80%, such a plant might produce power which could be sold at the bus bar for under 25 mills/Kwh. It is to be expected that the fuel cost andthe capital cost will lower as the technology develops. It is not unreasonable to expect later plants of this size to produce power which could be sold at the bus bar at about 15 mills/Kwhr at a plant factor of 80% and perhaps lower depending upon financial charges. The three reactors being considered by American and Foreign Power for installation in South America fall in this category. It will be possible to arrive at a somewhat more definitive appraisal of the economic feasibility of reactors in this size range after the bids for the three AFP reactors have been submitted, and the designs made available for analysis later this year. The 10 Mw reactor which Westinghouse is to build for the Brussels exposition is also in this size category. 8. In the medium size reactor range (30 to 80 Mw), the problems of development are much like those for the larger central power plants, and their introduction will follow, rather than precede, the commercial entry of the larger plants which are already under development. It is to be expected that the capital cost of such reactors will be higher per kilowatt of capacity than for the larger stations, since the cost of turbo-generator equipment and the cost of the reactor are affected more exponentially than linearly by a decrease in electric capacity. However, the fuel costs of such reactors should not be significantly different than for the larger central station nuclear plants. Because the cost of electricity from conventional stations in this size range is usually higher than electricity for larger thermal stations, the nuclear plant will undoubtedly be able to compete in this size range also. However, as of today, it is not possible to develop a detailed analysis of what cost to expect since little developmental interest has so far been directed toward such a reactor. The United States is plann- ing to design and construct several reactors in this size range, but no estimates of cost or of performance are as yet available.