How Nuclear Weapons Spread: Nuclear Weapon Proliferation in the 1990s (Operational Level of War)

HOW NUCLEAR WEAPONS SPREAD The production of the first nuclear fission chain reaction in 1942 ushered in a new era in history and science. With both detrimental and beneficial effects, the nuclear age had begun—an age which would have far-reaching effects on the lives of subsequent generations. In How Nuclear Weapons Spread, Frank Barnaby examines the implications of nuclear weapons, and considers the close relationship between peaceful and military nuclear programmes. The book looks in detail at the nuclear programmes of Third World countries which have or could soon have the bomb, such as Israel, Pakistan and India, as well as those which are thought to be developing their nuclear capabilities—such as Iran and North Korea. Even more alarming is the threat that terrorists might acquire nuclear weapons. Frank Barnaby assesses the reality of this risk, and considers methods of checking the spread of these weapons. The book also contains detailed descriptions of the components needed for nuclear fission and thermonuclear weapons, and discusses the need to test these weapons, as well as the difficulties of disarming and disposing of nuclear materials. How Nuclear Weapons Spread will be of great interest to students of International Politics, International Relations and Strategy Studies. Frank Barnaby is a former Guest Professor at the Free University in Amsterdam and a former Director of the Stockholm International Peace Research Institute. He is currently a defence analyst and is the author of many books, including Star Wars, 1987, The Automated Battlefield, 1987 and The Role and Control of Weapons, 1992.

THE OPERATIONAL LEVEL OF WAR Edited by Michael Krause, Deputy Chief of the US Army Center for Military History, and Andrew Wheatcroft

The Operational Level of War series provides for a theory of armed conflicts in the present and the immediate future. Unlike many theories, it is not rooted in abstractions but in the practice of war, both in history and the immediate past. The books in the series all contribute to the clearer understanding of the potentials and the dangers of war in the 1990s. The key contribution of the operational theory of war is to provide a link between strategy and tactics, a connection which is of unique importance in modern warfare. Titles already published in this series include: THE FRAMEWORK OF OPERATIONAL WARFARE Clayton R.Newell UNHOLY GRAIL Larry Cable MILITARY INTERVENTION IN THE 1990s Richard Connaughton THE SCIENCE OF WAR Brian Holden Reid THE ROLE AND CONTROL OF WEAPONS IN THE 1990s Frank Barnaby

HOW NUCLEAR WEAPONS SPREAD Nuclear-weapon proliferation in the 1990s

Frank Barnaby

London and New York

First published 1993 by Routledge 11 New Fetter Lane, London EC4P 4EE This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy copy of this or any of taylor & Francis or Routledge's collection of thousands of ebooks please go to www.eBookstore.tandf.co.uk.” Simultaneously published in the USA and Canada by Routledge 29 West 35th Street, New York, NY 10001 © 1993 Frank Barnaby All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data applied for ISBN 0-203-16832-1 Master e-book ISBN

ISBN 0-203-26354-5 (Adobe eReader Format) ISBN 0-415-076749 (Print Edition)

CONTENTS Introduction 1

THE LINK BETWEEN PEACEFUL AND MILITARY NUCLEAR PROGRAMMES

viii 1

2

NUCLEAR POWER IN ELECTRICAL ENERGY SUPPLY

12

3

THE PLUTONIUM ECONOMY AND HIGHLY-ENRICHED URANIUM

17

THE COMPONENTS OF NUCLEAR AND THERMONUCLEAR WEAPONS

25

5

NUCLEAR-WEAPON TESTING

38

6

DISMANTLING NUCLEAR WEAPONS

45

7

DISPOSING OF PLUTONIUM

50

8

THE PROSPECTS FOR THE NUCLEAR ARSENALS

56

9

INDIA’S NUCLEAR PROGRAMME

59

10

PAKISTAN’S NUCLEAR PROGRAMME

65

11

ISRAEL’S NUCLEAR PROGRAMME

69

12

IRAQ’S NUCLEAR-WEAPON PROGRAMME

75

13

NORTH KOREA’S NUCLEAR PROGRAMME

62

14

ARGENTINA’S NUCLEAR PROGRAMME

87

4

15

BRAZIL’S NUCLEAR PROGRAMME

92

16

SOUTH AFRICA’S NUCLEAR PROGRAMME

96

17

IRAN’S NUCLEAR PROGRAMME

99

18

NUCLEAR PROLIFERATION TO SUB-NATIONAL GROUPS

103

19

PREVENTING THE SPREAD OF NUCLEAR WEAPONS

108

20

THE PROLIFERATION OF NUCLEAR-WEAPON DELIVERY SYSTEMS

113

Appendix

120

References

125

INTRODUCTION On 2 December 1942, a team of scientists at the University of Chicago, under the leadership of Enrico Fermi, achieved the first nuclear fission chain reaction in the world’s first nuclear reactor. This remarkable event began a new era in history and science. It has had far-reaching effects on all our lives, some good, others bad. The beneficial effects triggered off by Fermi’s team include considerable advances in medicine, agriculture and industry as well as invaluable contributions to many branches of pure and applied scientific research. The bad effects have mainly arisen from the application of the nuclear chain reaction in nuclear weapons. Two of these weapons annihilated Nagasaki and Hiroshima in a flash, killing a quarter of a million people. Since those tragic days in August 1945, two generations have lived under the shadow of the mushroom cloud. This shadow has been largely lifted by the end of the Cold War. But the risk of local nuclear wars is still with us. The magnitude of this risk, a major threat to global security, is proportional to the number of nuclear-weapons states in the world. The spread of nuclear weapons to countries that do not yet have them is, therefore, a crucial issue. Nuclear fission occurs when a neutron enters the nucleus of an atom of, for example, uranium-235, one of the isotopes of the element uranium. The nucleus becomes very unstable and quickly splits into two parts, a process called fission. As the nucleus fissions, it shoots out two or three neutrons. These neutrons can be used to produce new fissions in uranium-235 nuclei. If enough uranium is present, at least one neutron from each fission event will produce another fission and a self-sustaining chain reaction will be achieved. The two parts into which a fissioned uranium nucleus splits are called fission products, and are normally radioactive isotopes. In addition to the fission products and neutrons, each fission event produces some energy. In a nuclear weapon, a very large number of fissions produces a large amount of energy, resulting in a large explosion. In a nuclear reactor, the amount of fission energy produced is carefully controlled. In a nuclear-power reactor, this energy is used to produce electricity. Nuclear reactors are also used to produce a large variety of radioactive isotopes, usually called radioisotopes. Many of these are used in medicine. Over the past fifty years the medical use of radioisotopes has become routine, diagnosing and treating disease in huge numbers of patients. Some of the most spectacular advances in medicine in recent years have been in imaging and scanning the organs of the body, such as the heart and brain. These techniques, which enable doctors to make precise and accurate diagnoses, often without the need for exploratory surgery, depend on the use of small amounts of radioisotopes, such as technetium-99.

Radioisotopes are also used in the treatment of disease. Some cancers are treated by, for example, inserting a rod or thin wire of radioactive material into the patient’s body, directly into the cancer site. As the radioisotope decays, the radiation given off attacks the tumour and kills the cancer cells. After a period of days or weeks, the wire is removed. This technique is particularly useful in breast, mouth, and uterine cancers. In another method of treatment, a radioisotope may be injected or swallowed by the patient. It then enters the bloodstream and circulates through the body. Certain radioisotopes are taken up by specific organs. Radioactive iodine, for example, is taken up by the thyroid gland. Once the radioisotope has been absorbed by the organ from the bloodstream, it will remain there for some time allowing the radiation it emits to work on the cancer. Radioactive iodine has been used with great success to cure thyroid cancer. It is also used to treat an overactive thyroid gland in patients who do not have cancer. A third method of treatment involves the placing of a radioisotope in a cavity of the body, which is then left to irradiate the area affected by the cancer. Cancer of the cervix, for example, is treated by the insertion of a radioisotope in the vagina. Cobalt-60 is produced for this type of treatment. Industry routinely uses radioisotopes in a variety of ways. For example, the radiation from a radioisotope like caesium-137 is used in a gamma-ray camera to photograph thick sections of steel to make sure that there are no cracks or other faults in them. Radioisotopes are also used to measure wear inside machinery, a technique which has led to advances in car engine design and improvements in fuel efficiency. The use of radioisotopes has considerably improved industrial processes in the coal, mineral, paper and steel industries. The use of radioisotopes in agriculture has greatly benefited developing countries. Their application to pest control is particularly beneficial. Insects that damage crops, like fruit flies, and those that carry disease, like tsetse flies, are controlled by the sterile-male technique. This involves using the radiation from a radioisotope such as cobalt-60 to sterilize a large number of male insects. The males may mate but will produce no offspring. When released into the wild in infested areas they will compete for females with non-sterilized males. The population of insects will be greatly reduced. The use of this technique to eliminate the screw worm fly is a major success story. The screw worm fly damages livestock by laying its eggs under the cattle’s skin so that the larvae hatch and kill the host. Nuclear fission has brought us all undoubted benefits through medicine, agriculture, and industry. But its application to nuclear power is much more contentious. Fifty years after the first reactor went into operation, nuclear-power reactors are producing about 17 per cent of the world’s electricity. The figure for the European Community is about 30 per cent. Nuclear enthusiasts argue that, in doing so, nuclear power is helping to reduce the emissions of carbon dioxide, the most serious environmental problem arising from the production of energy. Those opposed to nuclear power do not deny these benefits but argue that they are outweighed by the dangers of nuclear-power reactors. These include the health problems arising from the exposure of the public and workers in the nuclear industry to radiation from radioisotopes routinely emitted from nuclear-power stations and emitted during

nuclear accidents, such as the one at Chernobyl. There is also the vexed problem of the disposal of the huge amounts of radioactive wastes produced by nuclear-power reactors. A satisfactory sol ution to this problem has yet to be found. The demonstration of the first nuclear chain reaction was rapidly followed by the construction of military reactors for the production of plutonium for nuclear weapons. Since 1945, reactors in China, France, India, Israel, the UK, the USA and the former Soviet Union have produced about 200 tonnes of plutonium for the nuclear arsenals of these countries. In addition, civilian nuclear-power reactors used for the generation of electricity have produced worldwide about 800 tonnes of plutonium. Now that the Americans and Russians are beginning to dismantle many of their nuclear weapons under disarmament agreements, an increasing amount of military plutonium is being removed from nuclear weapons and stored under civilian control. And an increasing amount of plutonium from civilian reactors is being chemically separated from spent reactor fuel elements and kept in plutonium stores. Many believe that the production of plutonium is the worst legacy of the nuclear age. The problem with plutonium is its nature. Plutonium is man-made in nuclear reactors; only traces occur in nature. It is an exceedingly toxic material. The inhalation of just a minute particle can cause lung cancer. The ingestion of plutonium can cause liver and bone cancer. The half-life of plutonium is about 24,000 years, so that for all intents and purposes it remains permanently in the environment. Contaminated areas would be uninhabitable until decontaminated. But the main problem with plutonium is that it is an extremely efficient explosive. A few kilogrammes is enough to manufacture a nuclear explosion as powerful as the one that destroyed Nagasaki. There are no economically viable peaceful uses of plutonium. The bulk of it will, therefore, have to be permanently disposed of. This would be best achieved by incorporating the plutonium into glass and permanently disposing of it in geological depositories. But this process will inevitably take time. In the meantime, large amounts of plutonium will have to be stored. With so much plutonium in the world, the danger that some will fall into the wrong hands, including the hands of terrorists, is obvious. It is crucial that plutonium stores are very secure. The important question is: who should own, operate and safeguard plutonium stores, while preparations are being made for the permanent disposal of the plutonium? Because the theft of plutonium could have serious global consequences, the international community needs to be confident that plutonium is being stored securely. It will be confident only if the stores are under strict international management rather than the current system of national ownership and storage. The following chapters deal with various issues related to the spread of nuclear weapons. The intimate link between peaceful and military nuclear programmes is described, including the elements of the nuclear fuel cycle needed to produce the fissile material for nuclear weapons. The role of nuclear power in energy supplies, the evolution of the plutonium economy, and the civilian use of highly-enriched uranium are then discussed.

A description of the components needed for nuclear-fission and thermonuclear weapons follows and the need to test these weapons is discussed. The problems of dismantling the nuclear weapons to be destroyed under disarmament agreements and the methods of disposing of plutonium are then described. The nuclear programmes of India, Pakistan, Israel, Iraq, North Korea, Argentina, Brazil, South Africa and Iran are described. India, Pakistan and Israel either have nuclear weapons or could produce and deploy them very quickly. Iraq had a nuclear-weapon programme until the 1991 Gulf War. North Korea is thought by many to be developing nuclear weapons. Argentina, Brazil and South Africa were, until recently, probably developing them. Iran is suspected of having ambitions to acquire nuclear weapons. The risk that terrorists will acquire nuclear explosives is discussed as are ways of preventing the spread of nuclear weapons. Finally, the proliferation in Third World countries of nuclear-weapon delivery systems—combat aircraft and ballistic missiles—is described.

1 THE LINK BETWEEN PEACEFUL AND MILITARY NUCLEAR PROGRAMMES Now that the Cold War, and consequently the East-West nuclear arms race, is over, preventing the spread of nuclear weapons to countries that do not now have them has become a (if not the) top foreign-policy priority of the major, and many other, powers. Most political leaders would agree with American Secretary of State James Baker that nuclear-weapon proliferation ‘is perhaps the greatest security threat of the 1990s’. There is a direct link between the proliferation of nuclear weapons to countries that do not have them and the spread of nuclear technology for peaceful purposes. Hannes Alven, the Swedish Nobel Prize-winning nuclear physicist, has described the peaceful and military atoms as ‘Siamese twins’. A nuclear threat that should not be underestimated is nuclear terrorism. There is a considerable risk that sub-national groups, including terrorists and even small groups of criminals, will in the future acquire fissile material—particularly plutonium—and construct nuclear explosives. Nuclear terrorism has considerable ramifications for national, regional and even world security. The future of ex-Soviet nuclear weapons is of particular concern. Some of them, no longer under firm central political and military control, may be stolen, sold or transferred to other countries, sub-national groups or even criminals. Some observers believe that, for the next few years, the ex-Soviet nuclear arsenal is the most likely source for the illegal acquisition of nuclear explosives by governments or sub-national groups. But, as plutonium becomes more available, it is increasingly likely that some will be stolen and used to fabricate nuclear explosive devices. Other factors which increase the risk of nuclear terrorism include: the relatively small amount of plutonium needed to make a nuclear explosive; the availability in the open literature of much of the technical information needed to design and put together a nuclear device; and the small number of people needed to fabricate a primitive nuclear device. Only minute amounts of plutonium occur in nature. Virtually all the plutonium that exists is man-made, in nuclear reactors, particularly nuclear-power reactors. Nuclearpower reactors are the key element in the nuclear fuel cycle.

THE NUCLEAR FUEL CYCLE Nuclear energy has many applications in medicine, agriculture, industry and so on. It is

The link between peaceful and military nuclear programmes

2

used to propel ships and to provide power supplies for, for example, spacecraft. But, so far as controlling nuclear-weapon proliferation is concerned, the most controversial peaceful application of nuclear energy is the widespread use of nuclear-power reactors to generate electricity. A country with a nuclear-power programme will inevitably acquire the technical knowledge and expertise, and will accumulate the fissile material necessary to produce nuclear weapons. The extent of the spread of nuclear-power reactors, and the industries and technologies associated with the production of fuel for these reactors, must, therefore, be appreciated if the nuclear-weapon proliferation problem is to be fully understood. The nuclear industry can be divided into two main parts—one designing, developing and installing nuclear-power reactors and the other providing the fuel for them. The fabrication of nuclear-reactor fuel involves a number of industrial plants. These, and the nuclear-power reactors themselves, form the nuclear fuel cycle (fig. 1). A cycle occurs because nuclear reactors produce quantities of a new nuclear fuel as they consume nuclear fuel. The new fuel can, in turn, be used as a source of nuclear energy, creating still more nuclear fuel, and so on. In the vast majority of the nuclear-power reactors operating today, the original fuel is uranium. As uranium fissions in the reactor to produce energy, plutonium is produced. Plutonium can be used as fuel in breeder reactors which can be made to produce more plutonium than they consume. This excess fuel can be used to fuel new breeder reactors, and so on. But the starting-point of the nuclear fuel cycle is natural uranium. The processes involved in the nuclear fuel cycle include: mining the uranium; milling the ore; the conversion of U3O8 to UO2; the production of uranium hexafluoride (UF6); the enrichment of the uranium, to increase the amount of the fissile isotope uranium-235 in it; the fabrication of reactor fuel elements; the use of the fuel elements in a nuclearpower reactor to generate electricity; the reprocessing of spent reactor

How nuclear weapons spread

3

Figure 1 The nuclear fuel cycle fuel elements, to separate the unused uranium, plutonium, and fission products; and the storage and final disposal of radioactive wastes from the reprocessing plant.

URANIUM PRODUCTION The element uranium is very widely distributed, mainly dispersed through the rocks of the Earth’s crust. But only a small fraction occurs in concentrated ores. Economically mineable deposits occur in, for example, quartz-pebble conglomerates, phosphates, sandstones, shales and veins. There is no major difficulty in mining uranium ores— ordinary underground and open-pit methods and place-leaching are used.

The link between peaceful and military nuclear programmes

4

After mining, the ore is processed in mills to extract the uranium in the oxide form— U3O8. Because of its colour, this material is called ‘yellow cake’. After production, yellow cake is refined and purified. According to the London-based Uranium Institute, excluding the former Soviet Union and China, a total of 28,360 tonnes of uranium were produced commercially in 1990, considerably less than the 43,800 tonnes produced in 1980 (the average annual production in the 1980s was 38,200 tonnes). Canada accounted for 30 per cent of the total uranium produced in 1990 (in seven uranium mines); the USA for 14 per cent (in eleven mines); Australia for 12 per cent (in two mines); Namibia for 11 per cent (in one mine); France and Niger for 10 per cent each (in thirty-five mines); South Africa for 9 per cent (in seven mines); and Gabon for 2.4 per cent (in one mine) (Uranium Institute 1991). There has been a considerable reduction in the number of commercial uranium mines. The number operating in the above eight main producers decreased from 408 in 1980 to 73 in 1990 and 49 in 1991. The biggest reduction was in the USA—from 343 in 1980 to 11 in 1991; the reduction in France was from 38 in 1980 to 22 in 1991. Mines have been closed and production cut back because during the 1970s uranium production was considerably in excess of consumption in nuclear reactors and because orders for new reactors were cancelled. Consequently, large stockpiles grew and uranium prices in real terms fell significantly. At the same time, production costs rose and profitability dramatically decreased. Uranium stockpiles are likely to remain high for some time. The Commonwealth of Independent States (CIS) alone has a uranium inventory of roughly 170,000 tonnes; Western European countries and North America together have an inventory of about 70,000 tonnes. Annual uranium production is running at about 26,000 tonnes, excluding the Commonwealth of Independent States and China. The Commonwealth plans to export annually about 5,000 tonnes and China plans to sell abroad about 1,000 tonnes. The world’s power reactors in 1991 required about 55,000 tonnes of uranium and it is forecast that they will require about 64,000 tonnes by the year 2000 (Uranium Institute 1991). Since large numbers of nuclear weapons are being dismantled as the nuclear arsenals of the former Soviet Union and the USA are reduced, much highly-enriched uranium will become available. This may well be diluted with natural or depleted uranium to produce fuel for nuclear-power reactors. The amount of military highly-enriched uranium in the Commonwealth of Independent States is estimated to be equivalent to about 400,000 tonnes of uranium and there is a similar amount in the American nuclear arsenal. The total highly-enriched uranium in the two arsenals could be a source of one-fifth of the western world’s reactor fuel ‘for about the next 20 years’ (Nuclear Fuel 1992). The spot price of uranium has considerably decreased since the end of the 1970s. In 1992, uranium cost about US$18 per kilogramme (US$8 per pound) of U3O8; in 1979, the spot price was over US$88 per kilogramme (US$40 per pound) of U3O8. Given the size of existing inventories and the slow growth of nuclear power (and hence of the demand for uranium) over the next decade, a sustained recovery in uranium prices is unlikely for several years. Moreover, if the Commonwealth of Independent States dumps its massive stockpiles of uranium, to raise hard currency, prices would be further significantly depressed.

How nuclear weapons spread

5

In summary, there will be no shortage of uranium at favourable prices to fuel the world’s nuclear-power reactors for many years to come.

URANIUM ENRICHMENT Ordinary uranium consists of a mixture of three types of atom. As is commonly known, the nucleus of an atom contains protons and neutrons. All the nuclei of a given element contain the same number of protons, the atomic number. The atomic number of uranium is 92. The number of neutrons in the nuclei of the atoms of a given element may differ. The element is then said to have a number of isotopes. Ordinary uranium, for example, consists of a mixture of two isotopes. One contains 143 neutrons and the other 146 neutrons. These isotopes are referred to as U-235 and U-238, the numbers being the total number of protons and neutrons in the nuclei of each isotope. A crucial difference between U-235 and U-238 is that a nucleus of U-235 will undergo fission when any neutron, even one moving very slowly, collides with it. But a neutron can cause a U-238 nucleus to fission only if its velocity exceeds a certain value. Consequently, U-235 is of the greatest practical importance. Isotopes which undergo fission after capturing either a slow or a fast neutron are called fissile materials. They are of fundamental value both as fuel for civilian nuclear-power reactors and as material for the fabrication of nuclear weapons. Some nuclear-power reactors use natural uranium as fuel. But most use uranium in which the concentration of U-235 has been increased from its natural value of 0.72 per cent to typically about 3 or 4 per cent. The process of increasing the concentration of U235 is called uranium enrichment. Because U-235 and U-238 are chemically identical, it is necessary to use a physical method to separate and enrich them. Three methods are available for enriching uranium—the gaseous diffusion, the gas centrifuge and the jet nozzle techniques. The gaseous diffusion method relies on the fact that, in a gaseous mixture of two isotopic molecules, the molecules of the lighter isotope will diffuse more rapidly through a porous barrier than those of the heavier one. The gas used is UF6, which is solid at room temperature but easily vaporized. The problem is that UF6 is very corrosive and reactive and so special materials have to be used for the construction of the pipes, pumps and so on used in the process. Also, because the proportion of U-235 is raised by only a small fraction in each diffusion stage, many stages must be used to obtain a significant enrichment. A gaseous diffusion plant uses large amounts of electrical power, usually requiring the construction of a separate power station to run it. Historically, most uranium has been enriched by the gas diffusion method. Gas diffusion plants have been operated in all the declared nuclear-weapon powers (China, France, the UK, the USA and Russia) initially to enrich uranium for military use. No diffusion plant has been built outside the declared nuclear-weapon powers. Given the materials now available, the gas centrifuge method has replaced gas diffusion as the preferred method for enriching uranium.

The link between peaceful and military nuclear programmes

6

The centrifuge method of separating isotopes in gaseous form depends on the principle that the gravitational force on a particle is proportional to its mass. A centrifuge produces a field of force analogous to gravity but much more powerful. A gas centrifuge for uranium enrichment consists of a vacuum tank containing a long, rotating drum with a nozzle at one end and an orifice at the other. Uranium hexafluoride gas is pumped in via the nozzle and, as the gas moves up inside the rotating drum, molecules of the uranium hexafluoride gas will tend to be flung outwards by the centrifugal force. Molecules of uranium hexafluoride gas in which the uranium is U-238 are slightly heavier than molecules of U-235 hexafluoride. There will, therefore, be a difference in the centrifugal force acting on the molecules of different masses when the gas is rotated at very high speed. Molecules of the lighter U-235 isotope will diffuse towards the centre. The inner portion thus becomes enriched in U-235 and this is collected at the exit orifice. A plant containing many gas centrifuges in a cascade is needed to enrich a useful quantity of uranium. The slightly enriched flow of uranium gas from the first centrifuge is fed into the nozzle of the next centrifuge in the cascade and so on. The uranium is circulated around the cascade until the desired degree of enrichment is obtained. Gas centrifuges for uranium enrichment came into their own as new materials of high tensile strength became available for the outer casing of the drum, which rotates at very high revolutions per second, and the rotor bearings. Carbon fibre and special maraging steels are particularly suitable for use in gas centrifuges for uranium enrichment.

How nuclear weapons spread

7

Figure 2 The chain reaction, showing the fission of uranium-235 and the subsequent capture of neutrons by uranium-238 leading to the production of plutonium All the declared nuclear-weapon countries operate large gas centrifuge plants. In addition, Japan, Pakistan and South Africa operate them. Argentina, Brazil, India and possibly Israel are developing them, as was Iraq before the 1991 Gulf War. The Urenco plant at Almelo, the Netherlands, is a British-Dutch-German enterprise.

NUCLEAR-POWER REACTORS When a U-235 nucleus captures a neutron, a nucleus of the isotope U-236 is formed. A

The link between peaceful and military nuclear programmes

8

U-236 nucleus is very unstable, and usually rapidly splits (fissions) into two fragments, nuclei of elements of medium atomic numbers called fission products. During this fission process, the fission products are normally accompanied by two or three neutrons (fig. 2). The total sum of the masses of the fission products plus the masses of the fission neutrons is invariably less than the mass of the U-236 nucleus. This mass difference appears as energy accompanying the fission process. The amount of energy (E) produced is given by Einstein’s famous formula E=mc2, where m is the mass difference and c is the velocity of light. Although the mass difference m is very small, the velocity of light squared is large and, therefore, a relatively large amount of energy is released during fission. The complete fissioning of one gram of U-235 would produce about 23,000 kilowatt-hours of heat, equivalent to burning about 3 tonnes (3 million grammes) of coal. The neutrons emitted during the fission process are able to initiate further fissions in neighbouring U-235 nuclei. Provided one of the two or three neutrons emitted during a fission event can be made to fission another nucleus, a self-sustaining process, called a chain reaction, can be produced. Energy is then generated for as long as the chain reaction goes on. A nuclear-power reactor is basically a furnace where the heat produced by a controlled chain reaction is used to generate electricity. Typically, the heat is used to turn water into steam and the steam is used to drive a turbine which generates electricity. A typical modern power reactor generates about 1,000 million watts (1,000 megawatts or MW) of electricity, enough to supply the domestic and industrial electricity for, for example, a British city with a population of about 1 million people. In natural uranium only about one atom in 140 is U-235; the other 139 are atoms of U238. Because there is so much U-238 in it, natural uranium on its own cannot be used to produce a chain reaction. To overcome this problem the uranium is mixed with a substance known as a moderator. The nuclei of a suitable moderator are small in size so that if a fast neutron collides with one of them it will lose a large fraction of its velocity—just as a billiard ball will lose velocity when it collides with another billiard ball. After one, or a few, collisions the velocity of a fast neutron will be reduced (i.e. moderated) to a low velocity at which it can be effectively captured by a U-235 nucleus, producing fission, but at which it will probably avoid being captured by a U-238 nucleus. (The lower the velocity of the neutron, the lower the probability that it will be captured by a U-238 nucleus.) As seen above, a chain reaction will be sustained if, for each U-235 nucleus undergoing fission, at least one fission neutron causes fission in another nucleus. Some fission neutrons will not be available for producing further fissions. Some, for example, will be captured by U-238 and, therefore, lost to the fission process. Others will be captured by nuclei of the moderator. Some will escape through the exterior surface of the mass of the uranium and moderator. Provided the total number of neutrons produced by fission exceeds the total number lost by the three processes just described, a chain reaction will be possible. The size of the system in which the number of neutrons lost just balances the number produced by fission is called the critical size.

How nuclear weapons spread

9

In a typical nuclear-power reactor, the fuel is in the form of a number of cylindrical fuel elements. Heat is removed from the fuel elements by a coolant, such as water, which flows over them. The coolant then flows through a heat exchanger where it turns water in a secondary circuit into steam to drive the turbine. The fuel, moderator and coolant form the core of the reactor, which is usually surrounded by a layer of material, such as graphite or natural uranium, which reflects back into the core many neutrons which would otherwise escape, contributing to the efficiency of the reactor. A nuclear-power reactor must be provided with a control system to start and stop the reactor and to shut it down rapidly in an emergency. A reactor control system usually depends on a system of rods, made from a substance, such as cadmium or boron, which absorbs neutrons efficiently. The rods are moved automatically in and out of the reactor core to control the rate at which fission takes place. Normally, the reactor will be initially fuelled with more than the minimum amount of fuel to sustain a chain reaction. The control rods are inserted far enough into the core to mop up the excess neutrons and are withdrawn as the fuel is burnt away at a rate which sustains a chain reaction. Various types of nuclear-power reactors are possible, based on different combinations of fuel, moderator material, and coolant. But only three types have significant commercial importance. In them the material used as the moderator is graphite, ordinary (light) water or heavy water (in which the hydrogen is replaced by deuterium); and the coolant is gas (such as carbon dioxide), light water or heavy water. Graphite-moderated, gas-cooled reactors are used, for example, in France and the UK; heavy-water moderated reactors, cooled by heavy water or gas are popular in, for example, Canada; but the most frequently used reactors are light-water cooled and moderated.

REPROCESSING Normally, the fuel elements are removed from a nuclear-power reactor after three or four years. Even after this time, only about 1 or 3 per cent of the uranium will be used up. The elements must, however, be removed because the fission products absorb neutrons. As the amount of fission products increases, more and more neutrons will be absorbed and lost to the fission process. Eventually, the chain reaction will cease. Some of the U-238 nuclei in the fuel element will capture fission neutrons. When a U238 nucleus captures a fission neutron a nucleus of U-239 will be formed. U-239 will, in general, not undergo fission but will undergo radioactive decay to form plutonium-239 (Pu-239). Consequently, when uranium is used as fuel in a nuclear reactor, plutonium, a man-made element found in only insignificant quantities in nature, steadily accumulates in the fuel elements as an inevitable by-product. Pu-239 is, like U-235, a fissile material and is a potential reactor fuel. It can also be used as the fissile material to fabricate nuclear weapons. When removed from the reactor, a fuel element contains unused uranium, plutonium, and fission products. These three substances can be chemically separated from each other

The link between peaceful and military nuclear programmes

10

in what is called a reprocessing plant. The fission products include isotopes of elements with atomic numbers ranging from 30 (zinc) to 66 (dysprosium). A relatively large amount of fission products is produced. For each megawatt of heat (or a third of a megawatt of electricity) produced, roughly a gramme of uranium fuel is consumed a day and the same amount of fission products produced. The bulk of these isotopes is radioactive. In fact, so much radiation is emitted by the fuel elements when they are removed from the reactor that they are dangerous to handle, even with remote handling equipment. The elements are, therefore, stored in a water-filled ‘cooling tank’ near the reactor typically for between three and five years before being sent to the reprocessing plant. During storage, the radioactivity decays by a factor of several thousand, and so does the heat, making the spent fuel elements easier to handle. Nevertheless, the fuel elements still contain much radioactivity. The main chemical operations in the reprocessing plant must, therefore, be performed by remote control behind thick shielding. According to present plans, only about 30 to 40 per cent of the fuel discharged from the western world’s reactors up to the year 2000 will be reprocessed. The rest must be stored securely until it can be disposed of permanently. So far, no publicly or politically acceptable method of disposal has been developed. In a typical commercial reprocessing plant, the chemical process used is based on an organic solvent, tributyl phosphate (TBP) and is called the Purex (Plutonium-URaniumEXtraction) process. The Purex process uses TBP dissolved in a kerosene hydrocarbon as the separating agent. The process depends on the fact that when uranium and plutonium are highly oxidized they are more soluble in the TBP-kerosene solution than they are in an aqueous (water) solution, whereas the fission products are more soluble in a strongly acid aqueous solution than in the organic solution (Bebbington 1976). The first reprocessing operation is to cut off the outer covers of the spent reactor fuel elements (an operation called decanning) to expose the bare fuel element. The element is then dissolved in nitric acid, forming nitrate solutions of the uranium, plutonium and fission products. Separation of the three substances takes place in a number of extraction ‘cells’ in which organic (TBP) and aqueous solutions, which do not mix, travel countercurrently so that substances more soluble in one solution than in the other are separated out. The TBP solvent is forced through each cell and takes the uranium and plutonium with it. The fission products are removed by a counterflow of aqueous nitric acid and are taken out of the cell. As the process is repeated in successive cells, the solution becomes increasingly free of radioactive fission products. The aqueous solution containing the fission products is sent to a radioactive-waste treatment plant. The uranium and plutonium are separated by using a reducing agent which converts the plutonium into the trivalent state which is not soluble in the TBP solvent. The reducing agent does not affect the uranium which is soluble in the TBP. The TBP solution containing the plutonium and uranium, is fed into a partitioning column. The plutonium is separated out of the TBP solvent by a downflowing stream of nitric acid containing the reducing agent and taken out of the bottom of the column. An upflowing countercurrent stream of TBP solvent removes the uranium from the top of the

How nuclear weapons spread

11

column. Uranium is then removed from the TBP solvent by dilute nitric acid. The plutonium solution and the uranium solution are further purified by similar processes. The plutonium and uranium emerge in nitrate solutions. The plutonium is normally converted to plutonium dioxide (PuO2), a convenient form for storage.

2 NUCLEAR POWER IN ELECTRICAL ENERGY SUPPLY NUCLEAR PROGRAMMES During the 1950s, there were high hopes that nuclear technology, turned from military to civilian purposes, would—by providing large amounts of cheap electrical energy, ‘too cheap to meter’ it was said—both hasten economic growth in the developed countries and enable Third World countries to develop rapidly. President Dwight Eisenhower articulated these hopes in a speech to the United Nations General Assembly on 8 December 1953: The United States pledges before you—and, therefore, before the world—its determination to help solve the fearful atomic dilemma—to devote its entire heart and mind to find the way by which the miraculous inventiveness of man shall not be dedicated to his death, but consecrated to his life. The project to beat nuclear swords into ploughshares was christened ‘Atoms for Peace’. But, sincere though the effort undoubtedly was, it did not bring the promised blessings. In the early 1970s, it was expected that there would be a continuing dramatic growth in the demand for electricity. In the USA, for example, the demand was projected to increase by 7 per cent a year, doubling every ten years. Much of the new electricity generating capacity would, it was thought, be nuclear. It was estimated that at least 700 large nuclear-power reactors would be operating by the year 2000. The 1973 oil-price increase drastically reduced the demand for electricity. In the USA, the demand typically grew by only 1 or 2 per cent. Many orders for nuclear reactors were cancelled. In the USA, for example, no new nuclear-power reactors have been ordered since 1978, and one has to go back to 1973 to discover a nuclear-power reactor that was ordered and not later cancelled. The experience of the oil-price rise, however, persuaded countries like France and Japan to reduce their dependence on oil imports by installing nuclear-power reactors. A future boost for nuclear power may be concern over the contribution to the greenhouse effect, and hence to global warming, from the atmospheric pollution produced by the burning of fossil fuels in power stations. Nuclear electricity was (and still is) promoted as being environmentally friendly. In the meantime, the prospects for nuclear power do not look bright. Nuclear electricity is proving to be relatively very expensive. So expensive, in fact, that the British

Nuclear power in electrical energy supply

13

government left nuclear power out of its electricity privatization programme. In the USA, for example, the average construction cost of twenty-one new nuclearpower reactors completed in the late 1980s and early 1990s was $3,700 per kilowatt or more than $4,000 million for a single reactor. In the early 1970s, the cost was about $200 per kilowatt and in 1980 was $750 per kilowatt. In the past twenty years, the construction cost of nuclear-power reactors has, in real terms to take inflation into account, increased sixfold (Greenpeace 1992). Typical operating costs of power reactors have also significantly increased. Over the past decade, these costs in the USA have increased threefold, in real terms. Combined operating and construction costs in the USA, for example, make nuclear electricity more expensive than coal and gas power plants. The 1986 Chernobyl nuclear accident was a major set-back for nuclear power, increasing considerably anti-nuclear feeling among the public. Reactor safety and economics are, of course, intimately linked. The incorporation of expensive reactor safety measures inevitably increases the cost of nuclear electricity. Finally, the nuclear industry has yet to come up with a politically and publicly acceptable way of disposing of highlevel radioactive waste. Until it does so, doubts about the wisdom of expanding nuclearpower programmes will continue. As of the beginning of 1992, the world’s nuclear-power reactors were generating a total of about 340,000 MW(e) (or 340 gigawatts of electricity, GW(e)), generated by 421 power reactors (Greenpeace 1992). They provided about 17 per cent of the world’s electricity. The world’s nuclear generating capacity will be less than 380 GW(e) in the year 2000 and probably about 450 GW(e) in 2010. So far, about seventy-five reactors, with a capacity of about 17 GW(e), have been shut down, after an average operating life of about seventeen years. Currently, twenty-nine countries are operating nuclear-power reactors. The nuclear share of electricity generation varies considerably in these countries, from about threequarters of the total electricity generated to less than 1 per cent. The percentages for individual countries for 1991 are: France 73; Belgium 59; Sweden 52; Hungary 48; South Korea 48; Switzerland 40; Spain 36; Bulgaria 34; Finland 33; the former Czechoslovakia 29; Germany 28; Japan 24; USA 22; UK 21; Argentina 19; Canada 16; CIS 13; the former Yugoslavia 6; South Africa 6; the Netherlands 5; Mexico 4; India 2; Pakistan 0.8; and Brazil 0.6 (IAEA 1992a). The USA operates 110 nuclear-power reactors, generating about 100 GW(e), about a third of the world total. France is operating 56 nuclear-power reactors, generating 17 per cent of the world total; Japan has 42 reactors, generating 9 per cent of the world total; and Germany has 21 reactors, generating 7 per cent of the world total. In the CIS, Russia operates 31 reactors, the Ukraine operates 14 reactors, Lithuania operates 2 reactors and Kazakhstan operates 1 reactor. Canada operates 20 power reactors; the UK 37; Sweden 12; the former Czechoslovakia 9; India 9; South Korea 9; Spain 9; Belgium 7; Bulgaria 6; Taiwan 6; Switzerland 5; Finland 4; Hungary 4; Argentina 2; the Netherlands 2; South Africa 2; Brazil 1; China 1; Mexico 1; Pakistan 1; and the former Yugoslavia 1. Excluding the Commonwealth of Independent States (CIS), there are 48 nuclear-power

How nuclear weapons spread

14

reactors under construction in fifteen countries. These will add about 38 GW(e) to the world’s total electricity generation capacity. The construction of 25 power reactors has begun in the CIS but only a few, perhaps four, are likely to be completed in the foreseeable future. One nuclear-power reactor is under construction in Argentina; one in Brazil; two in Canada; two in China; two in Cuba; six in the former Czechoslovakia; six in France; five in India; two in Iran; ten in Japan; one in Mexico; five in Romania; three in South Korea; one in the UK; and one in the USA. The Austrian government has abandoned its only nuclear-power station at Zwentendorf. The Italian government has decided not to install more nuclear power and to dismantle the country’s three reactors. These have, in fact, been shut down since the 1986 Chernobyl accident. Greece has decided not to go ahead with its plan to build a nuclear-power station. Belgium, the Netherlands, Spain, Switzerland and the UK have declared moratoriums on new nuclear capacity. In 1980, Sweden announced its decision to phase out nuclear power by the year 2010 even though nuclear power produces nearly a half of the nation’s electricity. East European countries, except Romania, bought nuclear-power reactors from the former USSR. There is now considerable concern in these countries about the safety of these reactors and much public opposition to nuclear power. Bulgaria has five operating reactors generating 36 per cent of the country’s electricity and two more reactors are under construction. Technical problems and local opposition are embarrassing the new government. The former Czechoslovakia planned to close its two nuclear-power stations and may greatly reduce its dependence on nuclear electricity. Hungary has no plans to expand its nuclear-power programme and the future of its four operating power reactors is uncertain. Poland has placed a moratorium on nuclear power until 2000. All of the nuclear-power plants in what was the German Democratic Republic have been, or will soon be, closed down. Romania started constructing five Canadian-supplied nuclear reactors in 1974 but none has yet been completed. Serbia plans to close its only nuclear plant by 1995 and has declared a moratorium on new plants until 2000. The post-Chernobyl situation in the former Soviet Union is unclear. Glasnost has allowed public opposition to nuclear power to make itself felt for the first time ever. The future of the fourteen nuclear-power reactors of the Chernobyl (RBMK) type is, not very surprisingly, under consideration. There have been a series of accidents involving RBMK reactors and there are growing demands that these reactors be closed down before another serious accident occurs. There were four RBMKs operating at Chernobyl when the accident occurred there. All of them were closed down, but two have been restarted. In addition, there are currently thirteen RBMKs operating in the CIS, generating nearly a half of the Commonwealth’s nuclear-generated electricity. The RBMK is not the only reactor type in the former Soviet Union reckoned by many to be unsafe. Another is the first generation of the VVER reactors, called the VVER-440 model 230 reactors. No country outside the CIS operates RBMKs but model 230s are

Nuclear power in electrical energy supply

15

operated in Bulgaria and the former Czechoslovakia, as well as in Russia. In spite of considerable public opposition to nuclear power in virtually all the republics of the CIS, the nuclear industry is still very influential. Before the Chernobyl accident, the industry’s plans were very ambitious. It was planned to install enough nuclear reactors to provide about a half of the total electricity production by the year 2000. The reactors now operating provide about 13 per cent of the total electricity generated in the CIS, although in the industrial regions nuclear power plays a greater role, providing, for example, about a third of the electrical power in the north-west region and about a quarter of Ukraine’s electricity The new generation of VVER-1000 reactors are of a safer design than the older reactors and Russia hopes to export them to a number of countries—including Finland, India, Iran, North Korea, Pakistan and Syria. Moreover, given the demands for economic development and the difficulties of finding, in a short time, alternative sources of power, we must expect new nuclear-power stations in areas of rapid economic development, like Siberia and the far east of Russia. In Latin America, Argentina is constructing its third nuclear-power reactor but it may never be completed because of escalating construction costs. Brazil is building its second power reactor but has had second thoughts about its ambitious nuclear programme, the future of which is very uncertain. Mexico’s experience with nuclear power has been discouraging mainly because of the huge cost of building its single power reactor, and the future of Mexico’s nuclear programme is much in doubt, although its second reactor will be completed. Cuba now has no nuclear power but is constructing two Soviet-designed nuclear-power reactors. Delays in construction and rising costs put their completion in doubt. Iran started to construct three power reactors when the Shah was in power. The Shah hoped to have twenty reactors operating by the year 2000. But the Khomeini government stopped the programme in 1979. The Iranian government now says it intends to complete the reactors but whether or not it will do so is uncertain. Apart from the former Czechoslovakia and France, Japan is the most enthusiastic constructor of nuclear-power reactors, with ten reactors under construction. These will add about 25 per cent to Japan’s nuclear electrical generating capacity. Elsewhere in Asia, especially in South Korea and Taiwan, ambitious nuclear plans have been dramatically curtailed by public opposition to nuclear power, particularly after Chernobyl. A nuclear reactor was built at Bataan but has not yet been operated. It seems that it might never be. Nuclear power is expanding faster in the Third World than in developed countries. Currently, eight Third World countries (Argentina, Brazil, India, Mexico, Pakistan, South Africa, South Korea and Taiwan) are operating twenty-seven nuclear-power reactors, generating about 18 GW(e), or about 5 per cent of the world total. Eight Third World countries (Argentina, Brazil, China, Cuba, India, Iran, Mexico and South Korea) are constructing nineteen nuclear-power reactors, with a capacity of about 11 GW(e). When these new reactors are operational, the Third World will generate a total of about 29 GW(e), about 8 per cent of the world total. Whereas the reactors under construction in Third World countries will add about 60 per cent to their current nuclear electricity

How nuclear weapons spread

16

generating capacity, those under construction in developed countries will add only 10 per cent to their current nuclear electricity generating capacity.

3 THE PLUTONIUM ECONOMY AND HIGHLYENRICHED URANIUM The current and planned use of nuclear-power reactors to generate electricity inevitably produces large amounts of plutonium. What is done with this material is a crucial issue, with important ramifications for global security.

PLUTONIUM INVENTORIES The world’s stock of plutonium arises from two main sources. One is the plutonium produced for military purposes. This includes plutonium in nuclear weapons; plutonium removed from dismantled nuclear weapons and now in stores; and plutonium in military reserve stocks. The second source of plutonium is that produced in civilian nuclear reactors. This plutonium is in fuel elements in reactors; in spent reactor fuel elements in stores waiting for reprocessing or disposal; going through reprocessing plants; and in civilian plutonium stores. Official figures of the amounts of military and civilian plutonium so far produced have not been made publicly available. The best non-governmental estimates are those published by the Stockholm International Peace Research Institute (SIPRI) (Albright et al. 1993). According to Albright et al., the total global inventory of plutonium at the end of 1990 amounted to about 1,000 tonnes. By the end of 1992, this total will have increased by about 150 tonnes. Of the 1990 total, about 260 tonnes was produced for military purposes and about 740 tonnes was produced in nuclear reactors. Military plutonium is owned by the declared nuclear-weapon powers - the Commonwealth of Independent States (CIS) (Russia, the Ukraine, Belarus and Kazakhstan), China, France, the UK and the USA. In addition, plutonium for military purposes is owned by Israel and India. The amounts of military plutonium in these countries is: CIS, 125 (+ or −25) tonnes; the USA, 97 (+ or −8) tonnes; the UK, 2.8 (+ or −0.7) tonnes; France, 6 (+ or −1.5) tonnes; China, 2.5 (+ or −1.5) tonnes. According to Albright et al., Israel had accumulated about 325 kilogrammes of military plutonium by the end of 1991 and India had about 290 kilogrammes. Other sources give higher figures for the amount of plutonium produced by Israel. Figures supplied by Mordechai Vanunu, for example, about the rate of plutonium production at Israel’s Dimona reactor suggest that Israel may have produced about 600 kilogrammes of plutonium. All this military plutonium contains

The plutonium ecnomy and highly enriched uranium

18

at least 93 per cent of the isotope Pu-239, known as weapons-grade plutonium. Roughly 100 tonnes of the military plutonium in the CIS at the end of 1990 was inside nuclear weapons and roughly 25 tonnes was outside weapons. The corresponding figures for the USA were roughly 70 tonnes inside weapons and roughly 30 tonnes outside; for the UK, China and France, the total figures were roughly 5 tonnes inside weapons and 10 tonnes outside. At the end of 1990, the world’s civilian nuclear-power reactors had produced about 650 tonnes of plutonium. By the end of 1992, another 140 tonnes or so had been produced. By the year 2000, the world’s civilian reactors will have produced a total of about 1,700 tonnes of plutonium and will then be producing some 100 tonnes a year (Albright and Feiveson 1991). By the end of 1990, about 530 tonnes of plutonium were in spent reactor fuel elements, and about 122 tonnes of civilian plutonium had been separated from spent reactor fuel elements. At the end of the year 2000, according to current plans, some 300 tonnes of plutonium will have been separated. And, at the end of 2010, some 550 tonnes will have been separated. Of the 122 tonnes of separated plutonium at the end of 1990, about 72 tonnes were in store, 37 tonnes were being used to fuel breeder reactors, and 13 tonnes were in mixed oxide reactor fuels. The stored plutonium was in the UK (about 33 tonnes); the CIS (about 25 tonnes); Germany (about 8 tonnes); France (about 4 tonnes); Japan (about 2 tonnes); and Belgium (about 0.9 tonnes). By the year 2000, the amounts of separated plutonium in store will probably be: in the UK, about 62 tonnes; in Japan, about 43 tonnes; in the CIS, about 40 tonnes; in Germany, between 14 and 40 tonnes; in France, about 14 tonnes; in Belgium, between 1 and 5 tonnes; and in Switzerland, about 3 tonnes. At the end of the year 2000, according to present reprocessing plans, about 165 tonnes of surplus plutonium will be in store. And, by the end of 2010, about 250 tonnes of plutonium will be in store. Plutonium is being, or soon will be, separated in civilian reprocessing plants mainly in the UK (the B205 and THORP plants at Sellafield); France (the UP3 and UP2–800 plants at La Hague); Russia (at the Mayak-RTl at Chelabinsk); and Japan (at the Rokkashomura plant). By the end of 1990, the UK had separated about 47 tonnes of plutonium, France had separated about 42 tonnes and Russia about 25 tonnes. Of this plutonium, about 4 tonnes separated in the UK, about 19 tonnes separated in France and roughly 3 tonnes separated in Russia, were from foreign reactor fuel. Smaller amounts had been separated in Japan (about 3.6 tonnes); the USA (about 1.5 tonnes); Germany (about 1.2 tonnes); Belgium (about 0.7 tonnes); and India (about 0.15 tonnes). In addition, some 2.8 tonnes of Japanese plutonium was separated abroad, as was about 14.6 tonnes of German and 1.2 tonnes of Belgian plutonium.

CIVIL USE OF HIGHLY-ENRICHED URANIUM As already described, nuclear-power reactors use low-enriched uranium, in which the concentration of U-235 is increased to 3 or 4 per cent. Much higher enrichments, with

How nuclear weapons spread

19

concentrations of U-235 of up to 90 per cent or more, are used to fuel research reactors and in nuclear weapons. Research reactors are used for: teaching and research in universities and other institutions; producing radioisotopes for medical, agricultural and industrial purposes; and material and test purposes. (Civilian research reactors are often called material testing reactors—MTRs.) Of the 320 or so civilian research reactors (operated in fiftythree countries), about 150 (operated in thirty-six countries) are fuelled by highlyenriched uranium. Highly-enriched uranium fuel for civilian research reactors is supplied by the USA, Russia, the UK and France. The bulk of the highly-enriched uranium fuel for civilian research reactors is supplied by the USA. According to David Albright, the USA has exported about 24 tonnes of highly-enriched uranium (to over forty countries) and about 20 tonnes of highly-enriched uranium have been used in American civil research reactors. Of this, 12 tonnes of the exported and all the domestic highly-enriched uranium has been enriched to over 90 per cent in U-235. (For comparison, the amount of highly-enriched uranium in the American nuclear-weapon arsenal is about 500 tonnes.) Most of the highly-enriched uranium exported from the USA is fabricated into fuel elements in Canada, France, Germany and Japan. The Soviet Union probably supplied about 8 tonnes of highly-enriched uranium to fuel domestic and exported civilian research reactors. The fuel cycle for civilian research reactors includes the enrichment of uranium, the fabrication of highly-enriched uranium into fuel elements, the transportation of the fuel elements to the research reactor, the storage of fresh and spent reactor fuel elements, the transportation of spent fuel elements to the reprocessing plant, and the reprocessing of the spent fuel elements. On average, highly-enriched uranium fuel elements take five years to go through the cycle. About 6 tonnes of uranium enriched to over 90 per cent in U-235 are in continuous circulation in the cycle. The use of highly-enriched uranium to fuel research reactors will decrease as new low-enriched uranium fuels are used instead. Typically, about a third of the U-235 in the fuel elements is burned in the research reactors. About 10 per cent of the exported highly-enriched uranium has been returned to the supplier for reprocessing. About a half of the exported highly-enriched uranium used to fuel civilian research reactors is still overseas, either in civilian research reactors or stored as unused or spent reactor fuel elements. This means that about 16 tonnes of exported highly-enriched uranium remains abroad. Most of the highly-enriched uranium used domestically in civilian research reactors has been burned up or reprocessed. All in all, the amount of highly-enriched uranium in civil use is about 20 tonnes.

THE REPROCESSING OF HIGHLY-ENRICHED URANIUM SPENT REACTOR FUEL ELEMENTS The USA, the UK, France, China and Russia have facilities to reprocess highly-enriched uranium fuel. After reprocessing, the recovered highly-enriched uranium is typically used in military reactors to produce plutonium and tritium for use in nuclear weapons. The former Soviet Union supplied the enriched uranium fuel for the civilian research

The plutonium ecnomy and highly enriched uranium

20

reactors which it exported. The spent fuel elements were returned to the former Soviet Union for reprocessing. The future status of ex-Soviet reprocessing is unclear. The other major reprocessor was the USA which, as part of America’s policy of discouraging the proliferation of nuclear weapons, reprocessed spent research reactor fuel very cheaply. But in 1988 the US Department of Energy suspended reprocessing until an environmental impact statement is issued, a process that could take several years to complete. In the meantime, the main reprocessing in the west will be done by the Atomic Energy Authority Technology’s Fuel Services’ reprocessing plant now operating at Dounreay, Scotland, and probably by the Cogema company at the UP-1 facility at Marcoule in France. If the French company succeeds in winning contracts to reprocess MTR fuel it will, of course, be in commercial competition with AEA Technology. The Dounreay plant began operating in July 1958, mainly to reprocess spent fuel elements from the Dido, Pluto, and DMTR at Harwell. The capacity of the plant is about 1,000 elements a year (typically aluminium alloy fuel elements). A standard tributylphosphate (TBP)/ kerosene reprocessing system is used. The Harwell reactors provided about 300 elements a year. Now they have stopped operating, Dounreay will have spare capacity which AEA Technology is anxious to fill by reprocessing non-UK fuel elements. Dounreay has, in fact, been reprocessing foreign fuel elements since 1962. Up to now, about one-quarter of the 11,400 or so research reactor elements (containing about 1,700 kilogrammes of uranium) reprocessed at Dounreay have been foreign. The biggest customers for MTR reprocessing were France and Denmark, closely followed by Germany and Japan. Other customers were Australia, Greece, India, South Africa and Sweden. The 1992 reprocessing campaign at the Dounreay plant included some elements from Dido and Pluto and elements from Indian, German and Iraqi research reactors. Western research reactors unload the equivalent of only roughly 450 standard fuel elements (each weighing about 3.5 kilogrammes) a year. But there is also a large number of spent fuel elements in storage at reactor sites. The stored elements are estimated to contain about 1,500 kilogrammes of uranium and the reprocessing market is estimated at about $50 million. It is because of this backlog that Cogema believes it can compete successfully with AEA Technology. But this may prove to be optimistic because many MTR operators may keep spent elements in interim storage for some time for economic reasons. Reprocessing research reactor fuel is much more expensive than reprocessing power reactor fuel because there is much less of it and because the contents of research reactor fuel varies considerably, with enrichments ranging from about 20 to 93 per cent.

USE OF HIGHLY-ENRICHED URANIUM IN NUCLEAR WEAPONS To obtain uranium that can be used to construct a nuclear weapon, the amount of U-235 in natural uranium is increased by enrichment. The proportion of U-235 is normally

How nuclear weapons spread

21

enriched from its natural value of 0.7 per cent to more than 40 per cent, preferably to over 95 per cent. The greater the amount of U-235 in the uranium, the less will be the critical mass. The 50,000 or so nuclear weapons in the arsenals of the declared nuclear-weapon powers (the USA, the UK, China, France, and the four CIS republics of Russia, the Ukraine, Belarus, and Kazakhstan) contain about 1,280 tonnes of highly-enriched uranium (compared with about 260 tonnes of weapons-grade plutonium) (Albright et al. 1992). The amount of highly-enriched uranium in nuclear weapons, or stock-piled for nuclear weapons, in the CIS is 700 (+ or −150 tonnes), of which roughly 500 tonnes are in nuclear weapons. The amount of highly-enriched uranium in nuclear weapons, or stockpiled for nuclear weapons, in the USA is 550 (+ or −50 tonnes), of which roughly 300 tonnes are in nuclear weapons. The UK, France and China are thought to have, respectively, roughly 10, between 10 and 20 and about 15 tonnes of highly-enriched uranium in their nuclear arsenals. Some undeclared nuclear-weapon countries have chosen enriched uranium rather than plutonium as the fissile material for nuclear weapons. Pakistan is using this route to produce fissile material for nuclear weapons, South Africa did so, and Iraq was in the process of doing so before the 1991 Gulf War. Pakistan may have produced about 175 kilogrammes of highly-enriched uranium and South Africa may have about 400 kilogrammes (Albright et al. 1992). It is, however, far more difficult to produce highly-enriched uranium than weaponsgrade plutonium. Other Third World countries developing uranium-enrichment capabilities include Argentina, Brazil, India, Iran and Israel. In U-235, the average time between spontaneous fissions is much greater than it is in Pu-239 and the so-called ‘gun’ method can be used to assemble a critical mass of U-235 in a nuclear weapon (see p. 36). A less than critical mass of U-235 is ‘fired’ (as if from a gun) into another less than critical mass of U-235. When the two masses come together they form a super-critical mass. About 60 kilogrammes of uranium enriched to over 90 per cent in U-235 were used in the Hiroshima bomb. About 700 grammes were fissioned. The average time between spontaneous fissions was about one-fiftieth of a second—quite adequate for the gun technique. The yield of the Hiroshima bomb was about 12.5 kilotonnes. A fission weapon using U-235 can, however, also be made using the implosion technique. An implosion weapon requires considerably less highly-enriched uranium than a weapon using the gun technique. If surrounded by a reflector made from natural uranium 15 centimetres thick, 100 per cent pure U-235 has a critical mass of 15 kilogrammes (compared with 4.4 kilogrammes for Pu-239). With uranium enriched to 40 per cent U-235, the critical mass increases to 75 kilograms; with 20 per cent U-235, it is 250 kilogrammes. High concentrations of U-235 are, therefore, highly desirable if the material is to be used to produce nuclear weapons. Uranium enriched below 20 per cent is referred to as low-enriched uranium and is generally reckoned to contain too large a fraction of U-238 to be used for a nuclear explosive. Highly-enriched uranium is used to fuel reactors in naval warships. American naval reactors use uranium enriched to about 97.3 per cent in U-235. It appears that ex-Soviet

The plutonium ecnomy and highly enriched uranium

22

naval reactors are fuelled with uranium enriched to less than 10 per cent. It is reasonable to assume that about 50 tonnes of U-235 are in the world’s naval fuel cycles, with about 3 tonnes of new fuel being introduced each year.

THE PROBLEMS OF REPROCESSING Large commercial reprocessing plants are currently operating in the UK, France and Russia. Other plants are under construction in Japan and the CIS. If operated commercially, for a profit, a plutonium reprocessing plant is a complex and costly chemical establishment and, because the capital cost is relatively independent of the size of the plant, economic reprocessing can only be achieved if a large-scale plant is used to serve many reactors. A typical modern commercial reprocessing plant will reprocess about 800 tonnes of spent reactor fuel a year, and service about thirty modern nuclearpower reactors. The British plants are the B205 plant at Sellafield, which reprocesses spent reactor fuel elements from reactors fuelled with uranium in metal form, and the THORP plant at Sellafield, which reprocesses uranium oxide fuel. The French plants are the UP1 plant at Marcoule, which reprocesses uranium metal fuel, and the UP2 and UP3 plants at La Hague, which reprocess uranium oxide fuel. The Russian plant is the Mayak plant at Chelyabinsk, which reprocesses uranium oxide fuel. The Japanese are operating a smaller reprocessing plant for uranium oxide fuel—the PNC plant at Tokaimura—and are constructing a new reprocessing plant at Rokkashomura, for uranium oxide fuels, scheduled for operation in 1998. India operates a small reprocessing plant at Tarapur and is building another at Kalpakkam, both for uranium oxide fuels. Russia is constructing a large reprocessing plant at Krasnoyarsk. The capacity of a reprocessing plant is measured in the amount of spent reactor elements it can reprocess, measured in tonnes of heavy metal (uranium) a year. The British B205 plant can reprocess up to 1,000 tonnes of heavy metal a year and THORP can reprocess up to 700 tonnes a year. The capacity of the French UP1 plant is 500 tonnes a year; the capacities of each of UP2 and UP3 are 800 tonnes a year. The Japanese Tokaimura plant has a capacity of 90 tonnes a year and that of Rokkashomura will be 800 tonnes a year. The capacity of the Russian Mayak plant is 600 tonnes of heavy metal a year. Each of the Indian plants at Tarapur and Kalpakkam has a capacity of less than 200 tonnes of heavy metal a year. Currently operating reprocessing plants have a maximum capacity of about 4,000 tonnes of heavy metal a year, producing about 14 tonnes of fissile plutonium a year. If present plans are realized, the global reprocessing capacity in the year 2000 will be up to about 6,000 tonnes of heavy metal a year, producing about 22 tonnes of fissile plutonium a year. This reprocessing capacity will only be able to handle a fraction of the spent fuel discharged from reactors. Currently, the amount of spent reactor fuel arising from the world’s nuclear-power reactors is about 10,000 tonnes of heavy metal a year. By the year 2000, it will be about 11,000 tonnes a year. Today’s reprocessing plants can handle a maximum of about a half

How nuclear weapons spread

23

of the spent reactor fuel produced; by 2000, the fraction will be only just over a half. Reprocessing was originally justified on the grounds that there would eventually be a shortage of uranium and to make nuclear power viable in the longer term it would be necessary to use the plutonium produced in reactor fuel elements as nuclear fuel. Hence the need to extract the plutonium in reprocessing plants and recycle the plutonium in ordinary reactors, as mixed-oxide fuel, or in breeder reactors. Reprocessing enthusiasts argue that a plutonium fuel cycle would conserve uranium and fossil fuel resources. But, as it has turned out, there is no foreseeable shortage of uranium mineable at economic prices. Using a cunning design, breeder reactors can produce more nuclear fuel than they consume. Their preferred fuel is plutonium containing more than 90 per cent of the isotope Pu-239. But, because breeder reactors have not yet proved that they can produce electricity economically, only a few have been built. These are currently operated by France, Japan, Russia, Kazakhstan and the UK (Dounreay). For use as fuel in thermal reactors, plutonium oxide is mixed with natural uranium oxide to produce a mixed-oxide (MOX) blend which is then made into fuel elements for use in, for example, light-water reactors. Typical MOX fuel contains about 3.5 per cent by weight of fissile plutonium. For safety reasons, reactors are generally restricted to using MOX fuel in only one-third of the reactor core. MOX fuel fabrication facilities, for breeder and light water reactor fuel, are operating in Belgium, France and Japan. MOX fuel fabrication plants are planned in Russia and the UK. According to present plans, the total commercial MOX reactor fuel production and use in breeder and light-water reactors during the 1990s is likely to be about 65 tonnes of plutonium. Another argument used in support of reprocessing is that it makes it easier the management of radioactive waste. Some plutonium isotopes have long half-lives (the time taken for the radioactivity of a given amount of a radioactive isotope to decay to half its value). The half-life of Pu-239 is about 24,000 years; and that of Pu-240 is about 6,500 years. Removing the plutonium from the spent fuel makes easier, it is said, the disposal of reactor radioactive waste products. Reprocessing separates the highly radioactive fission products. They can then be concentrated into a relatively small volume. Glassifying this waste (vitrification into a glass-like material) is said to encapsulate the highly radioactive waste into a form which is much more resistant to corrosion, and therefore more suitable for burial than spent reactor fuel elements. The spent metallic fuel from Magnox reactors corrodes rapidly in water and is, therefore, difficult to store for long periods. Reprocessing is regarded as necessary for this type of fuel. In today’s world, commercial reprocessing is justified on the grounds that it can make a profit. Generally, contracts for reprocessing are made on a long-term basis. Reprocessing plants, like the British THORP plant, are valuable earners of foreign exchange. The main argument against commercial reprocessing is that it makes available large amounts of plutonium in a form which could be illegally acquired by governments or

The plutonium ecnomy and highly enriched uranium

24

sub-national groups. Many argue that the potential disadvantage of making easier the spread of nuclear weapons, and possibly encouraging nuclear terrorism, outweighs the benefits of earning foreign exchange. A major problem with a large reprocessing plant is the difficulty of detecting the diversion (i.e. theft) of an amount of plutonium sufficient to produce a nuclear explosive. A typical commercial reprocessing plant handling, for example, 800 tonnes of heavy metal a year, would produce about 3 tonnes of fissile plutonium a year. Even with the best available and foreseeable safeguards technology, safeguards on reprocessing plants are unlikely to be better than 98 per cent effective. This means that 2 per cent of the plutonium—or 60 kilogrammes a year—could be removed illegally without detection. Given that about 10 kilogrammes of reactor-grade plutonium metal or 35 kilogrammes of plutonium oxide are enough to make a nuclear explosive, the possible diversion of plutonium from commercial reprocessing plants is a serious problem. Spent reactor fuel elements are so radioactive, because of the fission products in them, that they are self-protecting. But if the plutonium is removed from the fuel elements and separated from the fission products, in a reprocessing plant, it is in a form that can be relatively easily handled. As the amount of plutonium produced worldwide in civilian nuclear-power reactors and separated from spent reactor fuel elements increases it will become easier for governments (and sub-national groups) to obtain it illegally. Whether or not to reprocess is, therefore, a crucial political decision. Plutonium is probably most likely to be stolen when being transported. This is why the increasing transportation of plutonium is of concern. When plutonium has been reprocessed it will be transported back to the country that owns it. This will involve global transportation systems—on road, rail and sea, and in the air. It will, therefore, be more vulnerable to theft. MOX fuel will also be vulnerable to theft by terrorist or other sub-national groups as it is transported from the MOX fuel fabrication plant to the various reactors which will use it as fuel. In fact, the theft of plutonium in MOX fuel elements is probably the most likely way in which such groups will, in the future, obtain plutonium. Although nuclear terrorism is likely to become an increasing risk, the more immediate problem is the spread of nuclear weapons to countries that do not yet have them. How easy is it for a country to design and fabricate nuclear weapons?

4 THE COMPONENTS OF NUCLEAR AND THERMONUCLEAR WEAPONS A country intending to fabricate nuclear weapons must acquire or produce a wide range of components. The major components for a pure fission weapon include: very high quality conventional high explosives; detonators for these explosives; electronic circuits to fire the detonators; a tamper and reflector; a core of fissile nuclear material, mainly weapon-grade plutonium or highly-enriched uranium; a neutron source to initiate a fission chain reaction. If the explosive yield of a fission weapon is to be ‘boosted’ by some fusion a tritium source will be required. A thermonuclear weapon requires a quantity of lithium deuteride to provide fusion and highly-enriched uranium to ignite the fusion process. The components of nuclear and thermonuclear weapons are now described in some detail.

FISSION WEAPONS The basic nuclear weapon is the fission weapon (originally called the A-bomb) which relies entirely on a fission chain reaction to produce a very large amount of energy in a very short time—roughly a millionth of a second—and therefore a very powerful explosion. The fission weapons built so far have used the U-235 or Pu-239 as the fissile material. Thorium (Th) could, in theory, also be used. When the nucleus of the isotope Th-232 captures a neutron it becomes Th-233 which undergoes radioactive decay to U233, which is fissile material like U-235 or Pu-239. Thorium has, however, not been used as the fissile material in nuclear weapons. Critical mass The smallest amount of fissile in which a self-sustaining chain reaction is just sustained is called the critical mass. The critical mass is that from which just as many neutrons escape per unit time as are released by fission. If this mass of material is increased, the number of neutrons produced by fission builds up, and considerably more fissions occur in each successive generation of fission. A ‘super-critical’ mass is created and a nuclear explosion takes place. In a super-critical mass the rate of production of fission neutrons exceeds all neutron losses and a rapid and uncontrollable increase in the number of neutrons within the mass occurs. The critical mass depends on a number of factors. First, the nuclear properties of the

The components of nuclear and theronuclear weapons

26

material used for the fission, whether it is U-235 or Pu-239. Second, the shape of the material—a sphere is the optimum shape because for a given mass the surface area is minimized which, in turn, minimizes the number of neutrons escaping through the surface per unit time and thereby lost to the fission process. Third, the density of the material (the higher the density the shorter the average distance travelled by a neutron before causing another fission and therefore the smaller the critical mass). Four, the purity of the material (if materials other than the one used for fission are present, some neutrons may be captured by their nuclei instead of causing fission). Five, the physical surrounding of the material used for fission (if the material is surrounded by a medium like beryllium, which reflects neutrons back into the material, some of the reflected neutrons may be used for fission which would otherwise have been lost, thus reducing the critical mass). Plutonium metal occurs in six phases, each having a different density (ranging from 19.8 to 15.92 grammes per cubic centimetre) and crystalline form. As normally produced, plutonium metal is brittle and difficult to machine into precise shapes. To make plutonium more machinable, it is alloyed, usually with gallium or indium. A typical alloy for use in nuclear weapons would probably contain 2 per cent by weight (8 per cent by atoms) of gallium. Alloying plutonium in one phase prevents it changing to another phase. This is important because a phase change, and the associated change in density, will change the volume of the mass of plutonium which may distort it. It would be very undesirable if this occurred in the carefully machined plutonium pieces in a nuclear weapon. The critical mass of, for example, a sphere of pure Pu-239 metal in the alpha phase, which has a density of 19.8 grammes per cubic centimetre and is the densest form of the metal, is about 10 kilogrammes. The radius of the sphere is about 5 centimetres, about the size of a small grapefruit. If the plutonium sphere is surrounded by a natural uranium neutron reflector, about 10 centimetres thick, the critical mass is reduced to about 4.4 kilogrammes, a sphere of radius of about 3.6 centimetres, about the size of an orange. A 32-centimetre thick beryllium reflector reduces the critical mass of alpha-phase Pu-239 to about 2.5 kilogrammes, a sphere of radius of 3.1 centimetres, about the size of a tennis ball. Using a cunning technique called implosion, in which conventional chemical explosives are used to produce a shock wave which uniformly compresses the plutonium sphere, the volume of the plutonium sphere can be reduced and its density increased. If the original mass of the plutonium is just less than critical it will, after compression, become super-critical and a nuclear explosion will take place. In practice, Pu-239 in the delta phase, which has density of 15.92 grammes per cubic centimetre, is used. The delta phase, which has a cubic crystalline form, would be typically stabilized as a gallium alloy. Using implosion the density of the plutonium can be roughly doubled so that a nuclear explosion could, with the best modern design including an effective, but practicable, reflector, be achieved with about 3 kilogrammes of delta-phase Pu-239. The trick is to obtain very uniform compression of the sphere. In such a weapon, the implosion will liquefy the plutonium before the explosion blows it apart.

How nuclear weapons spread

27

Nuclear-weapon designers prefer the concentration of Pu-239 to be as high as possible. The larger the concentration, the smaller the critical mass and, hence, the bigger the explosive yield for a given weight of plutonium. Plutonium containing more than 93 per cent of Pu-239 is called weapons-grade. Nominal weapons-grade plutonium contains 0.05 per cent Pu-238; 93 per cent Pu-239; 6.4 per cent Pu-240; 0.5 per cent Pu-241; and 0.05 per cent Pu-242. For the best yield-to-weight ratio, super-grade plutonium is used, containing 98 per cent Pu-239 and 2 per cent Pu-240. The complete fission of 1 kilogramme of Pu-239 would produce an explosion equivalent to that of 18,000 tonnes (18 kilotonnes, or kt) of TNT. Modern fission bombs have efficiencies approaching 40 per cent, giving yields of 7 kilotonnes or so per kilogramme of plutonium present. It is this high yield-to-weight ratio that makes nuclear weapons so special. Implosion In an implosion design, the plutonium would be surrounded by a spherical shell, made from a heavy metal, like natural uranium, which acts both as the tamper and reflector. The conventional explosive used to compress the plutonium sphere is placed outside the tamper. The tamper has two functions. First, because the tamper is made of heavy metal, its inertia helps hold together the plutonium during the explosion to prevent the premature disintegration of the fissioning material and thereby obtain a greater efficiency. Second, the tamper

The components of nuclear and theronuclear weapons

28

Figure 3 Direct routes to nuclear weapons. All these operations may be done under military control on a small scale and secretly. Nuclear explosives may be produced as by-products of a peaceful nuclear-power programme.

How nuclear weapons spread

29

converts the divergent detonation wave into a convergent shock wave to compress the plutonium sphere. The tamper may also serve to reflect back into the plutonium some of the neutrons which escaped through the surface of the plutonium core to minimize the mass of plutonium needed. In some designs the reflector is of a different material from the tamper, in which case the plutonium sphere is surrounded by another spherical shell, situated between the plutonium and the tamper. Beryllium is a good neutronreflecting material. An excellent description of a first-generation plutonium fission nuclear weapon is given by Margaret Go wing. Her description of an early British weapon (which she calls ‘the gadget’) follows. The first nuclear weapons developed by a country are likely to be of this type. An implosion design has been chosen, in which the mass of high explosive, surrounding a sphere containing both the fissile material and a tamper, was so arranged as to produce a shock wave travelling radially inwards and thus compressing the material. (Author’s note: The high explosive was arranged in a number of shaped charges, called ‘lenses’.) The design had the advantage of high velocities, which reduced the chance of pre-detonation despite the many background neutrons present in plutonium; at the same time the material was compressed to such density that super-critical masses were obtained with comparatively little material. It had been realised at Los Alamos (the Manhattan Project) that performance could be improved by using explosive lens to turn the divergent waves, which started from detonators, into parts of a common spherical wave converging on the centre of the sphere. The main components of the gadget can be listed, working from outside to the centre. First came the detonators, which operated from an impulse from a firing device and involved other auxiliaries like safety switches and arming circuits. The detonation had to be started simultaneously in all the lenses; the lenses themselves were carefully calculated shapes, containing a combination of fast and slow explosive so that transit from the detonator to every point on the inner spherical surface of lens was simultaneous. The detonation from the lenses then reached a spherical shell of homogeneous high explosive called the supercharge. Within the supercharge was the tamper, which converted the divergent detonation wave into a convergent shock wave, reflected some of the neutrons back into the fissile material and generally increased the efficiency of the explosion. Within the tamper was the plutonium and within that the initiator. The last component was necessary because, although the implosion resulted in a powerful compression of the fissile material and the surrounding tamper, the material would stay compressed only for a few microseconds and would then expand again very quickly. It was therefore essential to make sure that the chain started at the right moment. This could be done by creating at the centre of the fissile material an intense neutron source. (Gowing 1974)

The components of nuclear and theronuclear weapons

30

The compression of the plutonium sphere can be improved if the sphere is suspended within the tamper, with an air gap between the sphere and the tamper. This space allows the tamper to gain momentum before hitting the sphere, considerably increasing the compressive shock, a technique called ‘levitation’ (Hansen 1988). Pre-detonation In a nuclear explosion exceedingly high temperatures (hundreds of millions of degrees centigrade) and exceedingly high pressures (millions of atmospheres) build up very rapidly (in about one-half of a millionth of a second, the time taken for about 55 generations of fission). The mass of the material used for fission expands at very high speeds—initially at a speed of about 1,000 kilometres a second. In much less than a millionth of a second the size and density of the material have changed so that it becomes less than critical and the chain reaction stops. The designer of a nuclear weapon aims at keeping the fissionable material together, against its tendency to fly apart, long enough to produce an explosion powerful enough for his purpose. A major problem in designing implosion fission weapons for maximum efficiency is to prevent the chain reaction from being started before the maximum achievable supercriticality is reached—an eventuality called pre-detonation. Pre-detonation is most likely to be caused by a neutron from spontaneous fission—fission that occurs naturally without the stimulation of an external neutron—in the material used for fission. In 6 kilogrammes of Pu-239, for example, the average time between spontaneous fissions is only about three-millionths of a second. To prevent pre-detonation and loss of efficiency, the assembly of a plutonium bomb must be very rapid. Implosion is necessary. The spontaneous fission rate in the plutonium used to fabricate a fission weapon is clearly important. The fewer spontaneous fissions the better. In super-grade plutonium the rate is about 20 spontaneous fission neutrons per gramme per second whereas it is 66 spontaneous fission neutrons per gramme per second in weapons-grade plutonium. This rate is very small compared with the huge number of neutrons produced in a fission chain reaction. About 1023 nuclei of Pu-239 are fissioned to produce each kilotonne of explosive yield. This number of fissions would produce about 2.5×1023 neutrons. Reactor-grade plutonium The isotopic composition of the plutonium produced in reactors operated for different purposes varies. The plutonium produced specifically for military purposes is, as we have seen, rich in the isotope Pu-239, typically containing more than 93 per cent of Pu-239. Plutonium produced in nuclear-power reactors operated to produce electricity in the most economical way, known as reactor-grade plutonium, typically contains only about 60 per cent Pu-239. About 25 per cent is Pu-240 (in weapons-grade plutonium the amount is typically about 7 per cent) and about 10 per cent is Pu-241. If the reactor fuel is burnt at a very fast rate, the plutonium will contain about 40 per cent Pu-239, about 30 per cent Pu240, about 15 per cent Pu-241, and about 15 per cent Pu-242.

How nuclear weapons spread

31

Can reactor-grade plutonium be used to produce nuclear explosions? This is an important question because, if it can, countries operating nuclear-power reactors for peaceful purposes have access to plutonium that could be used to produce nuclear weapons. And, as the quantity of reactor-grade plutonium in the world increases, it becomes easier for a country to acquire it illegally and produce nuclear weapons. That reactor-grade plutonium can be used to produce a nuclear weapon has been shown in the USA, where at least two such devices have been built and tested. The critical mass of typical reactor-grade plutonium in the form of a bare metal sphere surrounded by a natural uranium reflector, about 10 centimetres thick, is about 7 kilogrammes. Reactor-grade plutonium is usually stored, after reprocessing, in the form of plutonium oxide and is, therefore, most likely to be available in this form. The oxide can, however, be easily converted to the metal form. Amory Lovins (1980) explains that the view that reactor-grade plutonium cannot be used in nuclear weapons is based on the following assumptions: 1 that reactor-grade plutonium is far more hazardous than weapon-grade plutonium to people dealing with it; 2 that a nuclear explosive device made from reactor-grade plutonium is much more likely to explode unintentionally; 3 that such a device, if it explodes at all, will not explode violently enough to do much damage, nor to accomplish the main aims of the makers; and 4 that its explosive yield is too unpredictable to be acceptable to its makers. Lovins concludes that ‘each of these assumptions contains, in certain circumstances, an element of truth’ but, he adds, ‘each is generally, or can by plausible counter-measures be rendered, false. Their implication that reactor-grade plutonium is not very dangerous is wishful thinking, and causes the proliferation risks of civil nuclear activities to be gravely underestimated.’ The conventional high explosives The timing of the detonations of the chemical explosives to produce the shock wave to compress the plutonium sphere is crucial for the efficient operation of an implosion atomic bomb. Microsecond (a milli-onth of a second) precision is essential. The shapes of the explosive lenses are rather complex and must be carefully calculated. The high explosive must be chemically extremely pure and of constant constituency throughout its volume. The conventional high explosives used to compress the spherical fissile core of a nuclear weapon are one of the most crucial components. If the compression is not symmetrical or rapid enough the nuclear explosion will not reach its predicted explosive yield. So important is the constituency of the high explosives that details of the chemicals used, the methods of their preparation, and the size and shape of the charges are closely guarded secrets. The Nagasaki bomb used high-explosive charges of Composition B, a mixture of cyclotrimethylenetrinitramine (RDX)—(CH2)3N3 (NO2)3—and trinitrotoluene (TNT)—

The components of nuclear and theronuclear weapons

32

C6H2 (NO2)3CH3. Composition B is a fastburning explosive more effective than TNT on its own. More modern implosion charges use diaminotrinitrobenzene (DATB)—C6H3 (NO2)3NH2—or triaminotrinitrobenzine (TATB)—C6H3 (NO2)3NH3. DATB and TATB are relatively insensitive to shocks. Pentaerythrtoltetranitrate (PETN)—C(CH2O(NO2)4 and cyclotetramethylenetetranitramine (HMX)—(CH2)3N3 (NO2)3—have also been used in nuclear weapons. The Iraqis, for example, were experimenting with HMX for use in their nuclear weapons. The amount of high explosive used in a fission weapon has decreased considerably since 1945—from about 500 kilogrammes to as little as about 45 kilogrammes in modern nuclear weapons (Hansen 1988). Normally, the more explosive charges there are the more perfect is the spherical symmetry of the shock wave. Forty or so detonations would be typical. Getting the timing of the detonation sequence—milli-microsecond (a thousandth of a millionth of a second) precision is essential—and the chemistry and geometrical shapes of the explosive lenses right are the most difficult problems in designing an efficient implosion-type nuclear fission weapon. The most sophisticated fission weapons use geometries other than the one described above. For example, the high explosive is arranged in an ellipsoid geometry, and only two detonators are used, one at each end of the ellipsoid. This arrangement is used, for example, in nuclear artillery shells. Firing the detonators A typical circuit to fire the detonators uses krytrons to generate short, high-current pulses with amplitudes of about 4,000 volts and rise-and-fall times of a few milli-microseconds. The krytron is a cold-cathode, gas-filled switch using an arc discharge to conduct high peak currents for short times. The energy in the current pulse used to fire the detonators in a nuclear weapon is normally produced by charged capacitors. Because the rate of change of current is very large, the capacitors must have a very low self-inductance. This is why the manufacture of such capacitors, rugged enough for military use, requires special attention. The neutron source For maximum efficiency, the chain reaction in an atomic bomb must be initiated at precisely the right moment—the moment of maximum super-criticality. The initiation is achieved by a pulse of neutrons. In earlier fission weapons, the neutron pulse was produced from a polonium-beryllium source. When alpha-particles from the polonium bombard the beryllium, neutrons are produced. In a fission weapon, the polonium and beryllium are contained in a hollow sphere placed at the centre of the plutonium sphere. The polonium and beryllium are placed on opposite sides of the hollow sphere. When the high-explosive lenses are detonated, the shock wave crushes the hollow sphere and mixes the polonium and beryllium, producing a pulse of neutrons. In today’s nuclear weapons, the neutron pulse is produced by a small electronic device

How nuclear weapons spread

33

called a neutron ‘gun’, which is a much more effective neutron generator than a polonium-beryllium source. In a neutron gun a high voltage is used to accelerate small amounts of tritium and deuterium. When the tritium and deuterium nuclei collide (and fuse—in this case, a difference in electrical potential is used to give momentum to the nuclei, instead of the high temperature of a fission explosion), neutrons (on the order of several tens of millions) are released. These large numbers of initial neutrons create many fission reactions at the start of the fission chain reaction explosion, thus increasing the efficiency of the explosion and allowing more of the plutonium to fission before it is blown apart (Hansen 1988). The neutron gun can be used to vary the explosive yield of a fission nuclear weapon. Variable yields can be obtained either by varying the voltage across the device—which alters the velocity of the tritium and deuterium nuclei and hence the efficiency of the fusion process—or by varying the amount of tritium or deuterium used. The problem of getting the timing of the shaped-charge detonations and the injection of the neutron pulse right is mainly theoretical, in calculating the timing sequence for optimum efficiency. The practical problems of manufacturing the electronic components and building the circuits to produce the calculated sequence of triggering pulses are much less difficult. Highly-enriched uranium as fissile material The alternative to Pu-239 as the fissile material in a nuclear weapon is U-235, although some of the most advanced types of nuclear weapons contain both materials arranged in thin concentric shells, rather than a solid sphere. Plutonium undergoes fission faster than uranium, and placing it inside a shell of enriched uranium makes more efficient use of its fission neutrons. In this way a greater explosive power can be achieved for a given mass of fissile material. The amount of highly-enriched uranium in, for example, the American nuclear arsenal is about 500 tonnes, five times the amount of plutonium in the arsenal. But the more modern nuclear weapons tend to use relatively more plutonium than earlier models. In U-235, the average time between spontaneous fissions is much greater than it is in Pu-239 and the so-called ‘gun’ method can be used to assemble a critical mass of U-235 in a nuclear weapon. In the Hiroshima bomb, for example, a less than critical mass of U235 was fired down a ‘cannon barrel’ (the barrel from a naval gun) into another less than critical mass of U-235 placed in front of the ‘muzzle’. When the two masses came together they formed a super-critical mass which exploded. About 60 kilogrammes of U-235 were used in the Hiroshima bomb. About 700 grammes were fissioned. The average time between spontaneous fissions was about onefiftieth of a second—quite adequate for the gun technique. The yield of the Hiroshima bomb was about 12.5 kt. A fission weapon using U-235 can, however, also be made using the implosion

The components of nuclear and theronuclear weapons

34

technique. If surrounded by a reflector made from natural uranium 15 centimetres thick, 100 per cent pure U-235 has a critical mass of 15 kilogrammes (compared with 4.4 kilogrammes for Pu-239). With uranium enriched to 40 per cent U-235, the critical mass increases to 75 kilogrammes; with 20 per cent U-235, it is 250 kilogrammes. High concentrations of U-235 are, therefore, highly desirable if the material is to be used to produce nuclear weapons. Designs based on the Hiroshima and Nagasaki bombs are likely to be used by countries beginning a nuclear-weapon programme. But even the first weapons now produced by a country would probably be more sophisticated than these early, primitive weapons. The Nagasaki bomb, for example, was about 3 metres long, 1.5 metres wide, and weighed about 4.5 tons. A modern fission weapon, even the first produced in a nuclear-weapon programme, should weigh no more than a few hundred kilogrammes. The difficulty of designing and fabricating a nuclear weapon from either Pu-239 or U235 is often exaggerated. A competent group of nuclear physicists, and electronic and explosive engineers, given adequate resources and access to the literature, would have little difficulty in designing and constructing such a weapon from scratch. They would not need access to any classified literature.

BOOSTED FISSION WEAPONS Although very large explosions—equivalent to the explosion of 100 or 200 kt of TNT— can be obtained from nuclear weapons based on pure fission, there is a limit to the explosive power that can be obtained from a militarily operational one. The maximum explosive power of a militarily usable fission weapon is 50 kt. Higher explosive power than can be achieved by a pure fission nuclear device can be obtained by ‘boosting’. In a boosted weapon, some fusion material is placed at the centre of the plutonium sphere in a fission weapon. When the fission weapon explodes, the temperature and pressure at the centre of the core are such that nuclear fusion can take place (see page 38). The neutrons produced during the fusion process produce additional fissions in the plutonium in the weapon before it disintegrates, increasing its efficiency. In an unboosted fission weapon, the rate of production of generations of fissions is about 100 per microsecond, for a boosted weapon it is about 1,000 per microsecond. Boosted weapons are, therefore, about ten times more efficient than unboosted ones. The fusion in a boosted weapon is used mainly as an additional source of neutrons to help the fission process, rather than as a direct source of energy. Because the efficiency of the weapon is increased, a higher explosive yield is achieved for a given weight of plutonium. Boosted weapons are essentially sophisticated fission weapons. Using boosting, a much higher explosive power is obtained from a given amount of plutonium. Militarily usable boosted weapons have explosive powers of up to about 500 kt, i.e. about ten times the power of non-boosted operational weapons. The yields of the most powerful boosted weapons are equal to those of low-yield thermonuclear weapons. In a typical boosted weapon a mixture of deuterium and tritium gases (heavy isotopes

How nuclear weapons spread

35

of hydrogen) is used as the fusion material. Deuterium is found in natural hydrogen at a concentration of 0.015 per cent. Tritium is normally produced by bombarding lithium-6 with neutrons in a nuclear reactor. A pressurised deuterium-tritium mixture is injected from a reservoir, placed outside the fission-weapon core, into a space at the centre of the plutonium sphere after the fission process has begun. Because the centre of the sphere is needed for the fusion mixture, a boosted weapon must be initiated by an external neutron gun. The pressure in the boosting system is typically about 20 million N per m2 and about 5 grammes of the deuterium-tritium gas mixture are injected into the centre of the plutonium sphere. The timing of the injection is crucial for maximum efficiency. The explosive yield of a boosted weapon can be varied by varying the amount of tritium and deuterium injected onto the core of the weapon. Alternatively, as described above, the yield can be varied by varying the voltage on the neutron gun. Because tritium has a relatively short half-life of 12.3 years, the tritium in the reservoir has to be replaced regularly. For ease of replacement the reservoir is fixed on the exterior of the weapon.

THERMONUCLEAR WEAPONS If explosions in the range of a few thousand kilotonnes are required, extra energy must be obtained from fusion. The fusion process is the opposite of fission. In fission, heavy nuclei are split into lighter ones. In fusion, light nuclei are formed (i.e. fused) into heavier nuclei. In nuclear weapons, the heavier isotopes of hydrogen—deuterium and tritium—are fused together to form helium. The fusion process, like the fission process, produces energy and is accompanied by the emission of neutrons. There is no critical mass for the fusion process and therefore, in theory, there is no limit to the explosive yield of fusion weapons—or H-bombs (H for hydrogen) as they are often called. Fission is relatively easy to initiate—one neutron will start a chain reaction going in a critical mass of a fissile material, such as Pu-239 or U-235. But fusion is possible only if the nuclei to be fused together are given a high enough energy to overcome the repulsive electric force between them due to their positive electric charges. In H-bombs, this energy is provided by raising the temperature of the fusion material. Because H-bombs depend on heat they are also called thermonuclear weapons. In a typical thermonuclear weapon, deuterium and tritium are fused together. But to get this fusion reaction to work, the deuterium-tritium mixture must be raised to a temperature of a hundred million degrees centigrade or so. This can be provided only by a pure fission nuclear weapon (atomic bomb) in which such a temperature occurs at the moment of the explosion. An H-bomb, therefore, consists of a fission stage, which is an atomic bomb acting as a trigger, and a fusion stage, in which hydrogen isotopes (tritium and deuterium) are fused by the heat produced by the trigger. Normally, the fusion material is in the form of a cylinder. The cylinder is made out of lithium-6 deuteride. When neutrons from the fission explosion bombard lithium-6 nuclei

The components of nuclear and theronuclear weapons

36

in the lithium deuteride, tritium nuclei are produced. The tritium nuclei fuse with deuterium nuclei in the lithium deuteride to produce fusion energy. These nuclear processes, employing lithium in a breeding cycle to produce tritium, and deuterium and tritium to produce fusion, can be described as follows: lithium-6+neutron→tritium+helium-4+4.8 Mev of energy deuterium+tritium→helium-4+neutron+17.6 Mev of energy (Mev=million electron volts of energy) It is very advantageous to use lithium-6 deuteride as the fusion material because it is a solid at normal temperatures whereas tritium and deuterium, the fusion materials used in boosted weapons, are gases at normal temperatures. It is, of course, much easier to construct nuclear weapons from solid materials than gases. Lithium, the lightest known solid, occurs in nature as the mixture of two isotopes lithium-6 and lithium-7. Most, 92.58 per cent, of natural lithium is lithium-7—only 7.42 per cent is lithium-6. Lithium-6 is separated from lithium-7 in natural lithium by the electrolysis of an amalgam of lithium and mercury. Lithium hydroxide is passed through the lithium-mercury amalgam and the separation of lithium-6 from lithium-7 occurs by chemical exchange by passing the mixture through exchange columns. Chemical exchange takes place between the amalgam and the aqueous solution of lithium hydroxide. The lithium-7 is concentrated in the amalgam phase and the lithium hydroxide becomes enriched in lithium-6. By repeating the process through each of a number of columns, the proportion of lithium-6 is typically increased from the natural proportion to about 85 per cent. The energy released from such a thermonuclear weapon comes from the fission trigger and the fusion material. But, if the fusion device is surrounded by a shell of uranium metal, the high-energy neutrons produced in the fusion process will cause additional fissions in the uranium shell. This technique can be used to enhance considerably the explosive power of a thermonuclear weapon. Such a weapon is called a fission-fusionfission device. On average, about half of the yield from a typical thermonuclear weapon will come from fission and the other half from fusion. H-bombs are much more difficult to design than fission nuclear weapons. The problem is to prevent the fission trigger from blowing the whole weapon apart before enough fusion material has been ignited to give the required explosive yield. Sufficient energy has to be delivered to the fusion material to start the thermonuclear reaction in a time much shorter than the time it takes for the explosion to occur. This means that the energy must be delivered with a speed approaching the speed of light. This is achieved using the Teller-Ullman technique, invented by Edward Teller and Stanislaw Ullman. Rotblat has described the technique used: The solution to the problem lies in the fact that at the very high temperature of the fission trigger most of the energy is emitted in the form of X-rays. These Xrays, travelling with the speed of light, radiate out from the centre and on reaching the tamper (surrounding the fusion material) are absorbed in it and

How nuclear weapons spread

37

then immediately re-emitted in the form of softer X-rays. By an appropriate configuration of the trigger and the fusion material it is possible to ensure that the X-rays reach the latter almost instantaneously. If the fusion material is subdivided into small portions, each surrounded with a thin absorber made of a heavy metal, the bulk of the fusion material will simultaneously receive enough energy to start the thermonuclear reaction before the explosion disperses the whole assembly. (Rotblat 1981) Although essentially weightless, X-rays can exert great pressure. In an H-bomb, the pressure (several million pounds per square inch) is exerted uniformly on the fusion material and long enough for the fusion process to work before the material is blown apart. Because the radiation travels at the speed of light, it arrives at the fusion material about a millionth of a second before the much slower moving shock wave from the trigger explosion. The X-rays also arrive before any particles, including neutrons, produced in the fission explosion. When the shock wave arrives, and blows the assembly apart, the fusion explosion has occurred. The fusion process in a thermonuclear weapon is initiated by a so-called ‘sparkplug’, a thin sub-critical cylindrical rod of weapons-grade U-235 or Pu-239 placed at the centre of the cylinder of fusion fuel (Hansen 1988). When the fusion fuel has been compressed, by radiation from the explosion of the fission trigger, neutrons from the trigger penetrate into the sparkplug. The sparkplug begins to fission and the fission reaction, in the middle of the highly compressed fusion fuel, initiates the main fusion explosion. A thermonuclear weapon, therefore, uses two fission explosions—one in the trigger and another in the sparkplug. Very large explosive yields have been obtained with thermonuclear weapons. Typically, each stage of a thermonuclear explosion explodes with a power roughly ten times that of the preceding stage. If the fission trigger explodes with an explosive yield of a few tens of kilotonnes, the first fusion stage would explode with a yield of several hundred kilotonnes, and the second fusion stage, if present, would yield several megatonnes. For example, the Soviet Union exploded an H-bomb in 1962 with a yield equal to that of 58 million tonnes of TNT—equivalent to about 3,000 Nagasaki bombs. This was probably a three-stage device, with a fission trigger which exploded with a power of several hundred kilotonnes, and two fusion stages. Even higher yields could be obtained. The fusion process in a thermonuclear weapon is probably about 30 per cent efficient so that an explosion equivalent to that of about 25 kt of TNT is produced for each kilogramme of lithium-6 deuteride in the weapon. A 500-kt weapon would, therefore, contain about 20 kilogrammes of lithium-6 deuteride. Having constructed nuclear or thermonuclear weapons is it necessary to test them before deploying them in the military arsenal?

5 NUCLEAR-WEAPON TESTING Now that we are in a period in which the two major nuclear-weapon powers—the USA and Russia—are busily reducing their nuclear arsenals the need to continue testing nuclear weapons arises. The issue is important because a number of non-nuclear-weapon parties to the Nuclear Non-Proliferation Treaty (NPT) argue that by continuing testing the nuclear-weapon parties are violating their obligation under Article VI of the Treaty to ‘pursue negotiations in good faith on effective measures relating to cessation of the nuclear arms race at an early date and to nuclear disarmament’. The negotiation of a Comprehensive Test Ban Treaty is seen as an essential step towards fulfilling this obligation. In 1995, a conference will be held to decide how long the NPT will be renewed for. A significant extension may depend on progress towards a CTBT. In any case, testing nuclear weapons is a declining activity. In 1991, fourteen nuclear tests were made, the lowest number since the Partial Test Ban Treaty came into force in 1963. In 1991, the USA conducted seven nuclear tests, France six and the UK one. Russia and China conducted no tests in 1991 although China conducted one on 21 May 1992 and another in September 1992. The former Chinese test was estimated at 1 megatonne, the largest ever test by China. This explosive yield is nearly seven times greater than the 150-kt yield limit agreed in the Threshold Treaty signed by the USA and the CIS. Until the beginning of the 1990s, the declared nuclear-weapons powers actively tested nuclear weapons. Before the 1963 ban on testing nuclear weapons in the atmosphere, ratified by the USA, the UK and the USSR, 578 nuclear tests were conducted, 457 of them in the atmosphere. Between 1963 and 1992, 1,344 nuclear tests were made, for a grand total of 1,923 since the first test on 16 July 1945. Of these, the USA conducted 936 tests, the USSR 717, France 189, the UK 44, China 36 and India 1 (SIPRI 1992). The decline in nuclear testing is due to reduced military budgets, the end of the Cold War, reductions in the number of programmes to develop new types of nuclear weapons, and public pressures to restrict or ban testing, mainly because of their environmental impact. The closure in August 1991 of the ex-Soviet nuclear test site at Semipalatinsk in Kazakhstan, for example, was brought about by public pressure over health and environmental hazards. On 2 October 1992, US President Bush signed a bill containing provisions on limiting nuclear-weapon testing. The provisions of the new law are: a maximum of fifteen tests for the period 1 July 1993 to 30 September 1996, with a maximum of five of these tests in any one year, the tests relating to the safety of nuclear weapons; a maximum of one test relating to the reliability of American nuclear weapons; and no testing after 30 September 1996 unless a foreign state conducts a test after this date. The UK is allowed

Nuclear-weapon testing

39

one test a year, to be included in the safety total. The Bush Administration disapproved of these restrictions. And so does the British government. All the nuclear-weapon powers are, in fact, under pressure from their nuclear-weapon communities, particularly the nuclear-testing teams, to continue testing. The argument used is the need to continue modernizing nuclear weapons. But the real reason is probably the desire to keep the nuclear-weapon laboratories in business. In October 1991, Russian President Boris Yeltsin declared a one-year moratorium on testing at Russia’s other test site at Novaya Zemlya but there are indications that the Russians may resume limited testing at Novaya Zemlya in mid-1993, possibly at the rate of about three tests a year. France suspended nuclear testing for a nine-month period in 1992.

THE REASONS FOR TESTING What did all these tests achieve? Most tests are associated with the development of new types of nuclear weapons. Others are random tests to maintain confidence in weapons in the nuclear arsenals. Without confidence testing the military would, it is said, eventually lose confidence in the reliability of their nuclear weapons. It is also claimed that nuclear weapons must be tested to improve their safety. There is some controversy among nuclear-weapon experts about the need for confidence testing. American ex-weapon designers—for example, Hans Bethe, Richard Garwin, and J.Carson Mark—argue that testing is not necessary to maintain confidence in the nuclear arsenal, whereas ex-directors of the nuclear-weapon laboratory—Harold Agnew, Roger Batzel, and Donald Kerr—argue that testing is essential (Fetter 1988). The point is, however, that the military leaders are likely to lose confidence, sooner or later, in deployed nuclear weapons as nuclear war fighting weapons (targeted on an enemy’s military forces) unless weapons are tested from time to time. Put another way, the military must be convinced that nuclear war fighting weapons are so reliable that almost all will reach their targets. The military will only be so convinced if the effectiveness of nuclear weapons is demonstrated by testing. It should be noted, however, that most nuclear-weapon tests have been for the development of new types of nuclear weapons. In the USA, for example, in recent years about 80 per cent of tests have been related to new weapons and only about 15 per cent for stockpile reliability.

BRITISH NUCLEAR TESTING The USA and the USSR have developed a large number of types of nuclear weapons. The two powers have each developed about 100 types of nuclear warheads since 1945. Each main type has involved several (about seven or eight) tests. But that the testing programmes of the smaller nuclear-weapon powers have been considerably less effective is shown by the case of the UK.

How nuclear weapons spread

40

On 3 October 1952, Operation Hurricane—the explosion of Britain’s first nuclear device—took place near the Monte Bello Islands, a barren, uninhabited group of islands about 120 kilometres off the north-western coast of Australia. The prototype atomic device was exploded, 2.6 metres below sea-level, in the antiquated Royal Navy frigate HMS Plym. It had an explosive power equivalent to that of about 25,000 tonnes of TNT, roughly the same as the atomic bomb that destroyed Nagasaki. The device exploded in Operation Hurricane was very similar to the one dropped on Nagasaki. It showed that the British could produce weapon-grade plutonium; could design the chemical explosive lenses needed to compress uniformly a sub-critical mass of plutonium into a super-critical mass; had perfected the electronics to fire the detonators of the explosive lenses at the precise times; and could construct a polonium-beryllium initiator to produce a burst of neutrons at the moment of maximum super-criticality. While scientists at the Atomic Weapon Research Establishment at Aldermaston were developing the nuclear component, engineers at the Royal Aircraft Establishment at Farnborough were designing an aircraft bomb, code-named Blue Danube, to carry the nuclear warhead. After Operation Hurricane, progress in warhead design was much more rapid than expected. Within months, further nuclear tests, specifically of the Blue Danube design, were planned. These were to be landbased. A site was chosen in the area called Emu Field, about 480 kilometres north-west of Woomera, in the Great Victorian Desert of South Australia. And, on 15 and 27 October 1953, two nuclear weapons were exploded in Operation Totem. Totem-1 had a yield of 9.1 kt and Totem-2 of 7.1 kt. By this time, British scientists were looking forward to producing a thermonuclear weapon and were anxious to see if they could fuse nuclei of hydrogen in the vicinity of a fission explosion. To this end, the British Prime Minister (Eden) asked the Australian Prime Minister (Menzies), to agree to two nuclear tests involving some fusion. The explosive power of one of these tests was expected to be about 60 kt. It was felt that the Australians would be unhappy to have such a large nuclear explosion inland. Also, the British were anxious to push ahead with the development of an operational H-bomb and did not want to wait for the completion of a test range being prepared at Maralinga, some 200 kilometres south of Emu. And so the Monte Bello Islands were again chosen for the tests, code-named Operation Mosaic. Mosaic-1 was exploded, on a tower, on 16 May 1956 and had a yield of about 15 kt. Mosaic-2, also exploded on a tower, had a yield of 60 kt, the most powerful device exploded by the British in Australia. The mushroom cloud rose to an altitude of 14,000 metres. The Mosaic tests were particularly controversial because, in the words of the report of the Australian Royal Commission into British nuclear tests in Australia, ‘the theoretical predictions were incorrect for both Mosaic tests and parts of the (radioactive) clouds passed over the mainland of Australia. The presence of Aborigines on the mainland near the Monte Bello Islands and their extra vulnerability to the effect of fallout was not recognised’ by those responsible for safety at the tests. Seven more British nuclear weapons were tested in Australia between 27 September 1956 and 9 October 1957, all at Maralinga, four in Operation Buffalo and three in

Nuclear-weapon testing

41

Operation Antler. The yields of these explosions were, respectively: 15 kt; 1.5 kt; 3 kt; 10 kt; 1 kt; 6 kt; and 25 kt. The Buffalo explosions were designed to allow the effects of nuclear explosions to be measured, including those on military weapons—tanks, aircraft, and so on. Live animals were used in Operation Buffalo for the first time to measure the direct effects of heat and blast and the biological effects of the ingestion of radioactive products. Buffalo-2 was exploded at ground level and Buffalo-3 was dropped by the Royal Air Force and exploded 150 metres above the ground. Buffalo-3 was a complete test of the Blue Danube bomb system—the weapon was released from a Valiant bomber flying at an altitude of about 11,000 metres. Antler-3 was fired from a balloon at an altitude of about 300 metres. The other four nuclear devices were exploded on towers. The Antler explosions, like the Mosaic ones, were, according to the Royal Commission, ‘designed to evaluate components used in thermonuclear weapons although neither series involved thermonuclear explosions. High yield lightweight warheads for ballistic rockets and low yield lightweight weapons using plutonium-239 were tested, with particular emphasis being given to their triggering mechanisms’ The time between Britain’s first nuclear explosion and first full-scale thermonuclear test explosion was four and a half years, about the same time as it took the Soviets. The USA took seven years and France took over eight years. China holds the record at two and a half years. On 15 May 1957, Britain began testing thermonuclear devices and weapons having explosive powers in the megatonne (million tons) range. In 1957 and 1958, nine thermonuclear explosions were conducted at Christmas Island in the Pacific. Seven were dropped from Valiant bombers flying at about 12,000 metres—they exploded at an altitude of about 4,500 metres. Three of these had explosive yields of about 10 megatonnes. The first twenty-three British nuclear tests were, then, in the atmosphere. Between 24 September 1958 and 1 March 1962, Britain made no nuclear tests. It then began testing underground at the American test site in Nevada. Britain has made eighteen nuclear explosions at Nevada; the latest British test was in December 1991. The Nevada series of tests have been mainly devoted to the development of a warhead for the Skybolt missile (which was never deployed); the Chevaline advanced warhead for the Polaris submarine-launched ballistic missile; and, most recently, the warhead for the new Trident submarine-launched ballistic missile. The total number of nuclear warheads in Britain’s arsenal is about 400. But, in spite of all the nuclear testing, Britain has only a narrow range of types of nuclear weapons in its arsenal. British strategic nuclear forces consist of Polaris submarine-launched ballistic missiles, equipped with the Chevaline system which has a manoeuvring capability to evade Soviet anti-ballistic missiles defending Moscow. The explosive yield of the Chevaline warhead may be about 40 kt. The British have also begun to deploy Trident strategic nuclear submarines equipped with Trident submarine-launched ballistic missiles. The only RAF bomb in service is the WE-177 (probably with a variable yield of up to about 20 kt). The Royal Navy had a free-fall naval version of the WE-177 and a depth

How nuclear weapons spread

42

charge, probably also based on the WE-177; these were withdrawn from service early. British nuclear testing doesn’t seem to have been very cost-effective—forty-four tests for three types of deployed nuclear weapons.

IS NUCLEAR TESTING NEEDED? An important question is whether or not new nuclear-weapon powers will deploy nuclear weapons before a sample has been tested. Will the political leaders be prepared to accept nuclear weapons unless and until the designs have been tested with full-scale nuclear tests? In many countries, the political leaders will get different advice from military scientists, on the one hand, and military leaders, on the other hand. Today’s nuclear-weapon designers would be so confident that they could design and construct an ordinary nuclear fission weapon, using implosion to compress a sub-critical solid sphere of weapon-grade plutonium or enriched uranium, that they would not need a full-scale test of the device. Nor would testing be requested if enriched uranium were used in a gun-type design of the Hiroshima type. The designers would be very much less confident if reactor-grade plutonium were used and would probably want to test such a weapon. Provided that the fissile material used was weapon-grade uranium or plutonium, the designers could predict the explosive power rather precisely, within a narrow range. They would probably be sure that the weapons would produce explosive yields within their predicted range. This is important because the military will almost certainly want to know the precise explosive yield of their nuclear weapons. Will they demand tests to check that estimated yields can, in practice, be achieved within relatively narrow limits? Whether or not the military will take the word of the nuclear scientists and engineers that the nuclear weapons they design and build will work reliably according to their predictions will depend on the attitude of the military to science and technology. If a significant fraction of the senior military officers are technically minded it is likely that the military will accept the word of the scientists. But the designs of nuclear weapons that include an element of nuclear fusion are much more complex than first-generation nuclear fission weapons. The designers of boosted nuclear weapons and of full-scale thermonuclear weapons will, therefore, not be prepared to guarantee that these weapons will work well without a testing programme. The design of a boosted or thermonuclear weapon will require access to, for example, high-capacity computers. The test of a thermonuclear weapon need not involve testing the entire assembly at full explosive power. It would normally be enough to test the fission trigger plus a small section of the fusion component to test that the fusion process was set off. The explosive yield of such a test may be relatively low. If the scaled-down device produces some fusion, the designer will justifiably assume that the full-scale weapon will work effectively. To hide a nuclear explosion with an explosive power greater than, say, a few kt is a difficult task, even if it is set off deep below ground. Many seismic monitoring stations

Nuclear-weapon testing

43

have been established around the world to detect and measure the magnitude of earthquakes. A nuclear explosion with a yield above a few kilotonnes will be detected by seismic monitoring equipment even though it is operated outside the country in which the nuclear explosion takes place. The Indian explosion in May 1974 of a 12-kt nuclear device was, for example, easily detected. The explosion, set off at a depth of about 100 metres underground in the Rajasthan desert, produced a crater in the ground of 150 metres in diameter which was easily observed by US and Soviet reconnaissance satellites. The key components of a nuclear fission weapon could be tested without a full-scale nuclear test. Pakistan is said to have done so. The effectiveness of an implosion system, for use in a weapon based on a solid sphere of fissile material, could be tested using a sphere of the same size and shape made from a non-nuclear material. By photographing the explosion of the conventional high explosives with flash X-ray equipment the dynamics of the shock wave could be followed and the symmetry of the compression measured. Reports that Pakistan, for example, was buying flash X-ray machines from a Swedish firm in 1986 enhanced suspicions that Pakistan was planning to test its nuclearweapon design without a nuclear explosion. The design of a fission weapon could be tested by, for example, using only a small sphere of fissile material at the centre of the imploding system. The sphere could be so small that the fission yield would produce an explosion of a power about the same as that of the conventional explosive (i.e. 200 kilogrammes or less). This amount of fission would be sufficient to be detected by radiation detectors placed around the test assembly. The detection of a burst of radiation, particularly neutrons, would show that an effective implosion had taken place. But the explosion would be insufficiently powerful to be detected by seismic monitors outside the country concerned. Although it would be difficult to hide the test of a nuclear device with significant explosive yield, an attempt may have been made to do so. On 22 September 1979, an American Vela satellite recorded a double flash of light originating from the South Atlantic/Indian Ocean area. Vela satellites, operated by the US Air Force, specialized in the detection of nuclear explosions in the atmosphere and outer space. Up to 22 September 1979, Vela satellites had sighted forty-one nuclear explosions in fifteen years of operation. This was a perfect record. The scientists at the Los Alamos nuclear-weapon laboratory in New Mexico, who operated the Vela satellites, are particularly confident that the signal from the Indian Ocean came from a nuclear explosion—the satellite’s equipment had been calibrated just a week before. The signal produced in a Vela satellite’s equipment by a nuclear explosion in the atmosphere is a very characteristic one—a double pulse in which the heights of the two pulses are in a specific ratio. The Los Alamos scientists were not alone in believing that a nuclear weapon had been tested over the Indian Ocean. They were backed up by the American Defense Intelligence Agency, the CIA and the US Naval Research Laboratory. The CIA was specific about it. The explosion, it said, was a joint South African-Israeli nuclear test. A scientific panel was set up by the Carter Administration to review the evidence. The panel came out with the surprising conclusion that the evidence was not conclusive. In its

How nuclear weapons spread

44

report it said: ‘Although we cannot rule out the possibility that this signal was of nuclear origin, the panel considers it more likely that the signal was one of the zoo events, possibly a consequence of the impact of a small meteoroid on the satellite.’ A ‘zoo event’ is one that cannot be adequately explained! The panel reached its conclusion because this particular signal was bigger than previous Vela signals from nuclear explosions, suggesting that the event occurred ‘close to the satellite rather than near the surface of the earth’. Not a very convincing explanation. Particularly considering the evidence, other than that from the Vela satellite, that a nuclear explosion had taken place. Apart from evidence of radioactive fall-out which could have come from a nuclear test, the White House panel of scientists also discounted ionospheric evidence provided by the Arecibo Laboratory in Puerto Rico and the Los Alamos weapons laboratory. And the panel did not even consider new ionospheric evidence provided by the US Naval Research laboratory and radar detection by the US Air Force, all of which indicated a nuclear-weapon test over the Indian Ocean. The observatory at Arecibo, which has the world’s most sophisticated radio telescope, saw an unusual ripple in the ionosphere that could have been caused by a nuclear blast in the atmosphere. The shock wave from a nuclear explosion would travel through the atmosphere and produce disturbances in the ionosphere. The scientists at the laboratory reckoned that the ripple they observed occurred at the same time and in the same general area as the flash detected by the Vela satellite. Early-warning radars operated by the US Air Force picked up signals on 22 September 1979, of what some analysts believed was a nuclear test. This trio of events—the Vela double flash, the ionospheric ripple, and the radar signals—together with the measurements of radioactivity, are strong evidence. A task force of South African warships was conducting a secret exercise at sea the very night and at the same latitude and longitude as the nuclear explosion is believed to have taken place. South African warships visit the area extremely rarely. The fact that ships and aircraft avoid the area makes it attractive for a clandestine nuclear test. In summary, there are different interpretations of the event over the Indian Ocean of 22 September 1979. If it was a test of an Israeli or South African nuclear weapon, it was probably a low-yield device, producing an explosion with an explosive power equivalent to that of 2,000 or 3,000 tonnes of TNT.

6 DISMANTLING NUCLEAR WEAPONS PLANNED REDUCTIONS IN THE NUMBERS OF NUCLEAR WEAPONS The nuclear arsenals of the USA and the former Soviet Union contain a total of about 47,000 nuclear weapons, tactical and strategic. The American tactical nuclear arsenal currently contains about 7,000 nuclear weapons. The number of ex-Soviet tactical nuclear weapons is about 15,000 (but, according to some published figures, the number may be up to 20,000). As of the end of 1991, the American strategic nuclear arsenal contained about 13,000 strategic nuclear warheads. The ex-Soviet strategic nuclear arsenal then contained about 12,000 nuclear warheads. (For comparison, China has about 250 nuclear weapons, France has about 600, Israel has about 150 and the UK has about 400.) The Strategic Arms Reduction Treaties (START I and II) and Soviet-American announcements of unilateral reductions in numbers of tactical nuclear weapons will, if they are all carried out, cut the total number of nuclear weapons operationally deployed by the United States and the former Soviet Union from about 47,000 to less than 10,000.

STRATEGIC NUCLEAR WEAPON REDUCTIONS The START I reductions will reduce the number of American strategic nuclear warheads by about 2,500 and the number of ex-Soviet strategic warheads by about 3,700. The START II agreement signed by President Bush and President Yeltsin during the Russian summit in January 1993 will reduce the total number of nuclear weapons in the arsenals of the two powers to about 7,000, roughly equally divided between them. This reduction is to be achieved by the year 2003 at the latest (Factfile 1992). The 3,500 or so American strategic nuclear warheads will include: 500 warheads on long-range land-based missiles; 1,728 on submarine-launched ballistic missiles; and about 1,250 cruise missiles and bombs carried by strategic bombers. The 3,500 or so Russian strategic nuclear warheads will include: 500 warheads on mobile long-range land-based missiles; 1,744 on submarine-launched ballistic missiles; and about 1,250 cruise missiles and bombs carried by strategic bombers. Although the sizes of the two arsenals are almost the same, the Americans will, in practice, have the advantage because their submarine-launched ballistic missiles are much more accurate than their Russian counterparts.

Dismantling nuclear weapons

46

TACTICAL NUCLEAR WEAPON REDUCTIONS The greatest reduction in numbers of nuclear weapons will be in tactical ones (those whose delivery systems have ranges of less than 5,500 kilometres). American unilateral reductions will lead to the destruction of 1,300 artillery shells, 850 Lance surface-tosurface missiles and 900 naval nuclear depth-charges. In addition, 1,275 more (350 sealaunched cruise missiles (SLCMs) and 925 naval nuclear bombs) are to be withdrawn and stored. The number of American tactical nuclear aircraft bombs deployed in Europe will be reduced from 1,400 to about 700 and all nuclear weapons will be removed from South Korea; these weapons will presumably be stored. According to General Colin Powell, chairman of the Joint Chiefs of Staff, the USA will eventually reduce its tactical nuclear arsenal to about 1,600 tactical nuclear weapons. Officially announced unilateral reductions could reduce the number of ex-Soviet tactical nuclear weapons by 10,000 land-based weapons (artillery shells, surface-tosurface missiles, nuclear land mines and surface-to-air missiles), and 2,000 sea-based weapons (naval depthcharges, torpedoes, ship-to-air missiles, sea-launched cruise missiles, anti-submarine warfare missiles and mines). It is proposed that some of these weapons will be destroyed and some put into storage. Gorbachev also proposed the removal, by bilateral negotiations with the Americans, of all nuclear bombs and air-to-surface missiles carried on tactical aircraft. This would remove from operational deployment the remaining 3,000 ex-Soviet tactical nuclear weapons. Recent statements by President Yeltsin suggest that all the ex-Soviet tactical nuclear weapons will be destroyed. Whether or not this will require some reciprocal action from the United States is not clear. We are, therefore, entering a period in which thousands of nuclear weapons will be dismantled because of START, new bilateral agreements, unilateral initiatives and the retirement of aged and obsolete weapons.

THE LOCATION OF EX-SOVIET NUCLEAR WEAPONS Soviet nuclear weapons were deployed in all the republics. The 3,000 or so Soviet tactical nuclear weapons that were deployed in Eastern European countries—Czechoslovakia, Hungary, Poland and East Germany—have been removed. The last were removed from the former East Germany in August 1992, ten months after German unification. The 800 or so nuclear weapons, all of them tactical, deployed in the three Baltic republics of Lithuania, Latvia and Estonia, were withdrawn to central storage sites after the Baltic republics became independent. It is claimed that all the ex-Soviet tactical nuclear weapons that were deployed outside Russia were transferred to ‘central storage’ in Russia by 1 July 1992. Intercontinental ballistic missiles are scheduled to be removed by the end of 1994 and, according to some military officials, eliminated by 1996. But the Ukraine and Kazakhstan are apparently

How nuclear weapons spread

47

loath to fulfil these schedules, probably mainly because of distrust of Russia. Russia, the Ukraine, Kazakhstan, and Belarus contain all the 12,000 or so ex-Soviet strategic nuclear warheads. About 80 per cent of the strategic nuclear force, including all six of the Soviet strategic nuclear submarine ports, are in Russia. If all the ex-Soviet tactical and strategic nuclear weapons are concentrated in the four republics of Russia, the Ukraine, Kazakhstan, and Belarus, there may be about 21,000 nuclear weapons in Russia, about 4,000 in the Ukraine, about 1,400 in Kazakhstan, and about 800 in Belarus.

TYPES OF NUCLEAR WEAPONS The nuclear arsenals contain a large variety of types of nuclear weapons. Strategic missiles carrying nuclear warheads include: landbased intercontinental ballistic missiles; submarine-launched ballistic missiles; air-launched cruise missiles; and short-range attack missiles. Free-fall strategic bombs are carried on strategic bombers. Tactical missiles carrying nuclear warheads include: sea-launched cruise missiles; surface-to-air missiles; surface-to-surface missiles; and anti-submarine warfare missiles. In addition, tactical nuclear weapons include: artillery shells; land mines; tactical aircraft bombs; naval aircraft bombs; naval depth-charges; anti-submarine warfare bombs; and torpedoes. Typically, strategic nuclear weapons have larger explosive yields than tactical nuclear weapons, as well as longer ranges. Virtually all strategic and many tactical nuclear weapons are thermo-nuclear. Many of the remaining tactical nuclear weapons are boosted weapons. Low-yield tactical weapons, such as artillery shells and land mines, are pure fission weapons.

THE DISMANTLING OF NUCLEAR WEAPONS Dismantling involves: removing the nuclear weapons or nuclear delivery systems from their deployed positions and transporting them to central storage areas; where necessary, removing the warhead from its delivery system (such as a missile); cutting open and removing the outer casing of the nuclear weapon. If the explosive power of the warhead is ‘boosted’ by fusion material, the container of tritium is removed. If the weapon is a non-boosted nuclear fission weapon, the conventional high explosives surrounding the core of fissile material are removed. The beryllium reflector and the uranium tamper surrounding the fissile material are then removed. Finally, the fissile material itself is removed. If the weapon is thermonuclear, the lithium deuteride and the highly enriched uranium in the fusion stage is removed. Then the conventional high explosives surrounding the core of fissile material in the fission trigger are removed, followed by the beryllium reflector and uranium tamper. Finally, the fissile material in the trigger is removed. As the weapon is being dismantled the smaller non-nuclear components, such as the electronic circuits used to fire the detonators of the conventional high explosives,

Dismantling nuclear weapons

48

batteries, and sensors, are removed as they are exposed. Non-nuclear components can be destroyed; for example, the conventional high explosive can be burned and electronic components discarded. Normally, the nuclear components will be melted down to change their form to keep secret details of the pit— the original machined size and shape of the plutonium and/or highly-enriched uranium pieces, the amounts of non-nuclear metals alloyed with the fissile material in the pit; the amounts of fissile material used; and the isotopic composition of the fissile material. Because of the presence of radioactive decay products, such as isotopes of americium; non-nuclear alloy material, such as gallium; and non-nuclear material used to coat (and make airtight) the fissile material, such as gold or copper, the pits will be chemically reprocessed to separate out and purify the plutonium and highly-enriched uranium. These fissile materials will then either be stored, in metallic or oxide form, or permanently disposed of. Nuclear weapons must be dismantled with great care. Each ex-Soviet weapon is, like most American ones, probably a sealed unit, filled with inert gas. The gas will probably be radioactive and its release would be a hazard to workers. Also, some of the nonnuclear material in the weapons, such as beryllium, is highly toxic. The conventional high explosive in the warhead must, of course, be handled with care. Many ex-Soviet, and some American, weapons do not have insensitive high explosives (less prone to shock than ordinary explosives). There is, therefore, a danger with these weapons of a conventional explosion during dismantling, a danger which will increase with the age of the weapon. Also, the safety devices in typical ex-Soviet weapons are much less sophisticated than those in their American counterparts. Some of the older exSoviet weapons probably do not have safety devices. Modern nuclear warheads contain Permissive Action Links, which are coded switches. The warhead cannot be armed without possession of the code. Warheads with Permissive Action Links are safer to dismantle than those without them because there is an additional stage in the arming process. The process of dismantling and destroying nuclear weapons has yet to begin in earnest. When it does, it will take place at the facilities in which the weapons were originally assembled—in the USA, at the Pantex plant in Texas, and in Russia, at a plant at Nizhnyaya Tura, on the eastern edge of the Urals, 200 kilometres north of Yekaterinberg (formerly Sverdlovsk) and possibly at plants at Penza and Novosibirsk (although the existence of warhead-fabrication facilities at the latter location is uncertain). The dismantling of nuclear weapons will involve the removal of large amounts of fissile material. As has been seen, the total amounts of fissile material in the American and ex-Soviet nuclear arsenals are between 200 and 250 tonnes of weapon-grade plutonium and between 1,200 and 1,600 tonnes of highly-enriched uranium. If the two nuclear arsenals are cut to a total of, say, 10,000 nuclear warheads, roughly 80 per cent of these totals will become available. Dismantling nuclear weapons is a lengthy process. If, for example, the ex-Soviet arsenal is to be reduced from the current 27,000 or so nuclear weapons to, say, 5,000, at a rate of about 3,000 a year (apparently the capacity of Russian dismantling facilities), the process will take about eight years. In the meantime, the weapons will have to be stored.

How nuclear weapons spread

49

Keeping large numbers of stored nuclear weapons secure from theft and adequately maintained, to minimize the risk of accidental explosion, is a difficult and expensive business. The Russians will not want foreigners involved in the actual dismantling of ex-Soviet nuclear weapons to avoid revealing top-secret design information. The Americans have offered Russia $400 million to help them build modern facilities to store nuclear weapons before dismantling and to store fissile materials after dismantling. Much more assistance will be required if the ex-Soviet nuclear weapons removed from operational deployment are to be stored securely and dismantled safely during the 1990s. CIA Director Robert Gates is concerned that ‘one or another of the older, less sophisticated very small tactical devices might be stolen or slip out of control of the central authorities’. And Colonel Viktor Alksis, parliamentary leader of the Soyuz group, has stated: It is impossible to predict the future of our nuclear weapons today. I talked to a commander of a submarine equipped with nuclear arms. He told me that he would immediately carry out the maximum nuclear strike possible if anyone should try to put our nuclear weapons under international control. He will not wait for an order from Moscow, he said. (Garwin 1992) The period of social, political and economic turmoil which many of the republics face will be a particularly dangerous time. The nuclear danger will be somewhat reduced if the proposed reductions in the numbers of nuclear weapons are implemented by both sides as quickly as possible. In the meantime, the weapons must be securely stored.

7 DISPOSING OF PLUTONIUM The world faces a crucial problem. What should be done with the huge, and increasing, stockpile of plutonium created by the civilian reprocessing of spent reactor fuel and by the dismantling of nuclear weapons as disarmament agreements lead to considerable reductions in the nuclear arsenals? Until the end of the Cold War, civil and military stocks of plutonium were usually regarded as distinct. But as plutonium is removed from nuclear weapons, transferred to the civilian sector, and brought under international safeguards, the distinction disappears. The problem with plutonium is its nature. It is an exceedingly toxic material—the inhalation of just a minute particle can cause lung cancer (about 27 millionths of a gramme in the lungs is extremely carcinogenic). With a half-life of about 24,000 years, it remains, for all intents and purposes, permanently in the environment. Contaminated areas would be uninhabitable until decontaminated. But it is the possible use of plutonium to make nuclear weapons that is the most serious problem. Most of the plutonium produced in civilian nuclear-power reactors will be left in spent reactor fuel elements. Eventually, these will have to be permanently disposed of. Some of the plutonium separated from spent reactor fuel elements and removed from nuclear weapons will be burnt as fuel in reactors. How should the remaining plutonium be handled?

WAYS OF DISPOSING OF PLUTONIUM Burning plutonium in reactors As has been described, some of the separated plutonium will be used as fuel in breeder reactors and in ordinary (light water) reactors. A breeder reactor is able to produce more plutonium than it burns as fuel. This extra plutonium can then be used to fuel new breeder reactors. In theory, a series of breeder reactors eventually becomes self-sufficient in fuel. But there are two major problems. The electricity produced by breeder reactors will be very expensive for the foreseeable future, so expensive that breeder reactors are much less economically viable even than ordinary reactors. Breeder reactors will produce electricity at prices competitive with ordinary reactors only when uranium is five times as expensive as it is today. This will not happen for decades into the future. (The price of uranium today is only about $20 per kilogramme, and is likely to remain low for the

Disposing of plutonium

51

foreseeable future.) The other problem with breeder reactors is that the type of plutonium they produce, and the type preferred as fuel, is the same as the type most suitable for fabricating nuclear weapons. Not very surprisingly, very few countries are, in fact, interested in developing breeder reactors. Two have been built in France but only one is in commercial operation; a prototype is about to begin operating in Japan; a small one is operating in Scotland; one is operating in Kazakhstan; and one is operating in Russia (four others are planned in Russia but are likely to be abandoned). Only Japan seriously plans to build a series of breeders, the first to start operating in 2005, but how realistic this Japanese plan is remains to be seen. The global future of breeder reactors is bleak indeed, at least for the foreseeable future. Unlikely to be taken seriously until they are proved to be economically viable, they are clearly no solution to the plutonium-stockpile problem. For use as fuel in light-water reactors, plutonium oxide is mixed with uranium oxide to produce a mixed oxide fuel, called MOX. The problem of using plutonium in MOX fuel is the relatively high cost of manufacturing MOX reactor fuel elements. It is much cheaper to make standard uranium fuel elements. Light-water reactors use low-enriched uranium fuel which costs about $750 per kilogramme. A realistic price for MOX fuel today is about $1,500 per kilogramme, excluding the cost of plutonium (Berkhout et al. 1992). The French and British price for reprocessing one kilogramme of spent reactor fuel is apparently about $1,000. This will produce about 5 grammes of plutonium. The use of plutonium fuel in ordinary (lightwater) reactors is not, therefore, economically viable. Nevertheless, the use of MOX as fuel in light-water reactors is planned in Belgium, France, Germany, Japan and Switzerland. About 30 per cent of the plutonium separated so far (about 40 tonnes) has been used in the breeder-reactor fuel cycle; about 10 per cent (about 14 tonnes) in the MOX fuel cycle; and the remainder (about 90 tonnes) is in store. Up to the year 2000, according to present plans, about another 8 tonnes will be used in breeder reactors, and about another 50 tonnes will be used as MOX fuel. By the year 2000, therefore, the amount of civilian plutonium in store will have been increased to about 200 tonnes (Albright et al., forthcoming). Of this, about 60 tonnes will be stored in the UK, about 50 tonnes in Japan, about 40 tonnes in each of Germany and Russia, and about 15 tonnes in France. Plutonium is being separated from spent reactor fuel elements much faster than it is being, or will be, used as fuel in breeder and ordinary reactors. Mixing plutonium with high-level radioactive waste Other suggested methods of disposing of plutonium include: firing it into the sun using rockets; transmuting it into other elements in special reactors or particle accelerators; mixing it with high-level radioactive waste. The risk that a rocket might accidentally fall back to earth with its plutonium payload is environmentally unacceptable. Machines for

How nuclear weapons spread

52

transmuting large amounts of plutonium have not yet been developed. A number of countries plan to put high-level radioactive waste into a form suitable for permanent disposal by glassification—converting it from liquid to solid form by incorporating it into blocks of glass (borosilicate glass) by a chemical process. Classified high-level waste is being produced in significant quantities in France (at La Hague and Marcoule); in Russia (at Chelabinsk); and in the UK (at Sellafield) and is planned in Japan (at Tokai) and the USA (at Hanford, Savannah River, and West Valley). Plutonium could be included in glassified high-level waste for permanent disposal, with very little extra cost. Alternatively, plutonium could be incorporated into silica glass without wastes. It would be more difficult to extract plutonium from silica glass than from borosilicate glass, making it more proliferation resistant. It has been estimated that the extra cost of incorporating plutonium into glassified high-level radioactive wastes would be about $1 million a tonne of plutonium, much less than disposal in MOX fuel. The cost of glassification of plutonium without high-level wastes would, it is estimated, be about $1.5 million per tonne (Berkhout et al. 1992).

INTERNATIONAL PLUTONIUM STORES Because of the toxicity of plutonium and its potential use in nuclear weapons it would be best if the material wasn’t produced in the first place. Such draconian measures may eventually become necessary to protect plutonium in storage against theft or illegal diversion, and particularly while it is being transported, that it is conceivable that democratic societies may be threatened if large stockpiles of plutonium are allowed to accumulate. There are, therefore, good arguments in favour of phasing out nuclear power as soon as possible. At the very least, commercial reprocessing should be stopped as soon as practicable. Because of the difficulty of storing spent reactor fuel elements from Magnox reactors and the need to fulfil existing contracts, it may be impossible to stop reprocessing immediately. But reprocessing is so uneconomical that it is hard to see it surviving for much longer than another decade. There is now a consensus that reprocessing does not offer better management of the high-level radioactive wastes in spent reactor fuel compared to the storage and direct disposal of spent reactor fuel elements without reprocessing. Present and planned reprocessing capacity is able to remove only about 20 per cent of the plutonium in discharged reactor fuel elements. The fact that the bulk of spent reactor fuel elements will have to be permanently disposed of must be faced and suitable geological repositories found. The disposal of separated civilian plutonium and military plutonium from dismantled nuclear weapons would be best achieved by incorporating the plutonium into glass and permanently disposing of it in geological depositories. But this process will inevitably take time. In the meantime, large amounts of plutonium will have to be stored. Unilateral and bilateral disarmament agreements may lead to the removal during the 1990s of about 90 tonnes of plutonium from dismantled American and ex-Soviet nuclear

Disposing of plutonium

53

weapons. With the 200 tonnes or so in civilian stores, this will give a total of nearly 300 tonnes of plutonium in store by the year 2000. Stores containing civilian and/or military plutonium must be: totally secure, able to resist forced entry by any method, including the use of explosives, and to suppress fire; equipped with the most modern systems to detect intruders; and provided with an effective system for cooling rooms in which plutonium in metal form is stored. Plutonium should preferably be stored in oxide form (PuO2) rather than as a metal; the oxide is more stable and less easily usable in nuclear weapons. Because of these requirements, plutonium storage is not cheap. A study by the US Nuclear Regulatory Commission estimated that the capital cost of a 25,000-square-metre storage facility containing 50 tonnes of plutonium oxide in 12,500 shipping containers, each containing 4 kilogrammes of PuO2, would be $240 million in 1990 prices. Assuming a real interest rate of 4 per cent and that the store operates for thirty years, the annual capital charge is $14 million. With a staff of 200, at a cost of $110,000 per person, and annual maintenance and operating costs of $6 million, the total annual operating plus capital costs are $42 million or $840 per tonne of plutonium per year (Berkhout et al. 1992). The cost of a store with a much larger capacity would be significantly reduced because labour costs would not increase proportionally and economies of scale would reduce marginal capital costs. Doubling the capacity could reduce the cost of storage to less than $500 per tonne of plutonium per year. Nevertheless, the cost of plutonium storage will make more attractive the permanent disposal of plutonium. The crucial question is: who should own, operate, and safeguard plutonium stores, while preparations are being made for the permanent disposal of the plutonium? Because plutonium is most vulnerable to theft and hijacking during transportation, there is a strong argument that civilian plutonium should be stored regionally, at six sites in five countries, close to commercial reprocessing plants—at La Hague and Marcoule in France, Sellafield in England, Chelabinsk in Russia, Kalpakkam in India and Tokaimura in Japan. Military plutonium should be stored near the nuclear-weapon dismantling plants in the USA and Russia—at the Pantex plant near Amarillo, Texas in the USA and at Nizhnyaya Tura in the Urals and Penza, south-east of Moscow, in Russia. The international community will be confident that plutonium is being stored securely only if the stores are under strict international safeguards. These could be provided by the International Atomic Energy Agency (IAEA). Under Article XII.A.5 of its Statute, the Agency has the right to require that any surplus of fissile material above the civilian needs of its members be deposited with the Agency, to prevent the national stockpiling of, for example, plutonium. But this concept of the IAEA as a depository for plutonium has lain dormant. The IAEA did, however, organize a study in 1982 on the international management and storage of plutonium, in which twenty countries participated. The participants could not agree over the procedures for withdrawing stored plutonium and the IAEA took no action. However, the 1985 NPT Review Conference called on the IAEA to ‘establish an internationally agreed effective system of international plutonium storage in accordance with Article XII.A.5 of its Statute’. In April 1992, William Dircks, Deputy Director of

How nuclear weapons spread

54

the IAEA, said, in a speech in Japan, that ‘The IAEA is ready to participate in organising the international storage or disposition of plutonium at the request of Member States wishing to place their plutonium under “international supervision”’ (Dircks 1992). The main opponents to the concept of international plutonium storage under the IAEA are Argentina, India and Pakistan. Australia, Canada, the Netherlands, Sweden, the UK and the USA are among the strongest supporters. Germany, Italy and Japan are lukewarm supporters of a minimal storage scheme (Fischer and Szasz 1985). The main reason for opposing, or being lukewarm about, the concept is the argument that countries which are not parties to the NPT will not join an international plutonium storage scheme. The only countries that will join, it is said, are those who have renounced nuclear weapons. Under these circumstances, many NPT parties are unwilling to accept the additional international obligations that would accompany international plutonium storage. Another reason why countries have opposed international plutonium storage is the fear that plutonium would not be released from storage as soon as the owners requested it. The objections to international plutonium storage are becoming weaker with time. It is now widely realized that there is no good reason for wanting plutonium for peaceful purposes. Countries might, therefore, just as well transfer it to international storage, particularly if there is an economic incentive to do so. Moreover, the difficult problems, and high cost, of physically protecting and safeguarding the plutonium would be the responsibility of an international agency rather than a national responsibility. International storage is likely increasingly to suit the main reprocessors—particularly France, the UK and Russia. They would then have an agreed and reliable international regime to look after the plutonium. International plutonium storage would help but not solve the nuclear-weapon proliferation problem. Countries could still construct clandestinely reactors to produce plutonium for military purposes. But the international management and storage of plutonium would be much preferable to the current chaotic situation of national ownership and storage, and give some confidence that plutonium was not being diverted to military uses (Scheinman and Fischer 1992). There is no doubt that regional security would be improved. Korean concerns about Japanese plutonium, for example, would be alleviated. Given that objections to international plutonium storage are weakening and that the main reprocessors are likely to see advantages in it, the time is ripe to elaborate the details of adequate international plutonium storage. But incentives will be needed to persuade countries to agree to international plutonium storage. The main one would be economic. For example, existing plutonium could be purchased from its owners at an attractive price. But to discourage the reprocessing of plutonium, there should be economic incentives to dispose permanently of spent reactor fuel elements. A step in this direction would be the establishment of international spent fuel centres which would store and arrange for the disposal of spent fuel. For a number of reasons, it would be best if the world surplus of civilian and military plutonium was managed, stored, and safeguarded by the United Nations, under the Security Council, rather than by the IAEA. The adequacy of IAEA safeguards is

Disposing of plutonium

55

questioned by some governments, even though the Agency is responsible for safeguarding the NPT. This is particularly so after Iraq, a party to the NPT, demonstrated that a party could take advantage of its membership of the Treaty to obtain assistance to acquire nuclear technology that could then be used to help develop nuclear weapons. Iraq did this even though it was regularly visited by IAEA inspectors. The Agency is responsible for promoting nuclear energy and many ask whether a single agency can effectively promote and safeguard nuclear activities at the same time, given the intimate relationship between military and peaceful nuclear programmes. With an increasing amount of plutonium becoming available, and governments and sub-national (including terrorist) groups being prepared to pay large sums of money for it, the emergence of a flourishing nuclear black market is a very real possibility. A number of recent incidents suggests that such a black market already exists. There is, therefore, a clear need for an international body to deal with nuclear smuggling. Such a body should have severe powers—like a nuclear Interpol. For all these reasons, a new international agency should be established, reporting regularly and directly to the United Nations Security Council: to be responsible for the management and storage of the world’s surplus plutonium, with the power to require that any surplus of plutonium be deposited with it, thereby ensuring that no nation stockpiles plutonium; to establish and operate a credible, investigative, and intrusive nuclear safeguards system; and to have the authority to control the nuclear black market and the clout to react immediately to incidents of nuclear smuggling.

8 THE PROSPECTS FOR THE NUCLEAR ARSENALS Nine countries are known to have nuclear weapons. They are the USA, the UK, China, France, Israel and four republics of the Commonwealth of Independent States—Belarus, Kazakhstan, Russia and the Ukraine. In addition, India and Pakistan either have nuclear weapons or could assemble some very quickly indeed. The world’s nuclear arsenals contain a total of nearly 50,000 nuclear weapons. The exSoviet arsenal contains about 27,000 nuclear weapons, the USA has about 20,000, France about 600, China about 300, the UK about 400 and Israel about 150 nuclear weapons (SIPRI 1992). As we have seen, the officially stated aim is to reduce American and ex-Soviet strategic nuclear arsenals to between 3,000 and 3,500 warheads each, the reductions to be completed before the year 2003. The American tactical nuclear arsenal may be reduced to about 1,600 weapons and the Russians may get rid of virtually all the ex-Soviet tactical nuclear weapons. If all these promises are actually implemented, the American and ex-Soviet arsenals will be eventually reduced from about 47,000 nuclear warheads to 9,000 or so. Such deep cuts in the ex-Soviet and American nuclear arsenals will have ramifications for the smaller nuclear-weapon powers. As the ex-Soviet and American nuclear arsenals are reduced, those of China, France and the UK will become relatively more significant. And Israel’s nuclear arsenal will obviously be a crucial issue in any arms control negotiations that emerge from the Middle East peace process. Sooner or later, then, there will probably be negotiations about the future size of all the nuclear arsenals. In the meantime, China, France, the UK and Israel intend to go on modernizing their nuclear weapons. What are the nuclear plans of these countries? The British strategic nuclear force consists of four Polaris strategic nuclear submarines, each carrying 16 submarine-launched ballistic missiles. Each submarine-launched ballistic missile carries two nuclear warheads, for a maximum total of 128 warheads. It is, however, thought that Britain has produced only enough warheads for three full boatloads of missiles, giving a total of 96 warheads. The British have, in addition, about 150 tactical nuclear weapons—including about 125 land-based aircraft bombs and 25 naval nuclear depth bombs to attack submarines. The naval weapons are no longer routinely carried on warships but are stored ashore. The British plan to replace their Polaris submarines with four new strategic nuclear submarines. The first, called the Vanguard, was launched on 4 March 1992 and is now

The prospects for the nuclear arsenals

57

undergoing sea trials. Vanguard is expected to enter operational service in late 1994. The second and third new submarines, the Victorious and the Vigilant, are being built and the fourth is on order. They are expected to enter service after 1995. Each Vanguard submarine will carry 16 American Trident submarine-launched ballistic missiles. Each Trident missile may carry eight independently targetable warheads, for a total of 512 nuclear warheads, a fourfold increase in the size of the British strategic nuclear arsenal (IISS 1992). But the Ministry of Defence may decide to deploy less than eight warheads on each missile and may produce only enough warheads for three full boat-loads of missiles. The total number of Trident warheads deployed may, therefore, be considerably smaller than the maximum of 512. The British are also considering a new tactical air-to-surface missile, to be developed in collaboration with France but with a British-designed nuclear warhead, to replace obsolete free-fall nuclear bombs on aircraft. French strategic nuclear forces consist of five strategic nuclear submarines, each carrying 16 submarine-launched ballistic missiles. Sixteen of these 80 missiles carry one warhead each. But each of the other 64 carries six independently targetable warheads. The French strategic nuclear forces, therefore, are equipped with a total of 400 nuclear warheads. In addition, the French have 18 land-based strategic missiles, each carrying one warhead, and 18 Mirage strategic bombers, each carrying one air-to-surface missile equipped with a nuclear warhead. French tactical nuclear weapons include 45 Mirage and 20 Super Etendard aircraft, each carrying one air-to-surface missile equipped with a nuclear warhead, and 44 Pluton launchers capable of firing nuclear surface-to-surface missiles. The Pluton tactical missiles will be phased out by 1994. France’s plans to modernize its nuclear forces have yet to be finalized. It is probable that land-based strategic ballistic missiles will be withdrawn. But France plans to build four new ballistic missile submarines and to develop a new submarine-launched ballistic missile, which may carry 10 independently targetable nuclear warheads. The first of the new Triomphant-class submarines will be commissioned in 1995; the second is being built. The Mirage strategic bomber may be replaced with the Raphael bomber. Early next century the French strategic forces could contain 800 warheads on submarine-launched ballistic missiles and British strategic nuclear forces could contain 512 warheads on submarine-launched ballistic missiles. The combined British and French nuclear arsenals could, therefore, contain a total of over 1,300 strategic nuclear warheads by this time. China has eight intercontinental ballistic missiles and between 70 and 100 intermediate-range ballistic missiles, each equipped with one nuclear warhead. Each of two strategic nuclear submarines carries 12 submarine-launched ballistic missiles, each with one warhead. China also has up to 120 medium-range bombers, each carrying one or two nuclear bombs. China is slowly modernising its strategic missiles, equipping them with independently targetable warheads. It may also be developing a more modern strategic nuclear submarine; a new bomber; and a mobile intercontinental ballistic missile.

How nuclear weapons spread

58

Israel is extraordinarily secretive about its nuclear weapons. Even though it is well known that Israel has nuclear weapons, Israeli officials continue to deny their existence. Government Ministers regularly repeat the standard Israeli formula that ‘Israel will not become the first country to introduce nuclear weapons into the Middle East’, a direct denial that Israel has nuclear weapons. The point of the denial is that other governments, particularly the US government, can maintain the myth that Israel doesn’t have nuclear weapons. This is crucially important for Israel because the US government cannot, under American law, continue giving economic aid to Israel if it acknowledges that Israel has nuclear weapons. Israel’s nuclear weapons are deliverable by aircraft and Jericho-II surface-to-surface missiles. The 1,500-kilometre range Jericho missile is being further developed. All the nuclear-weapon powers are very anxious to prevent other countries acquiring nuclear weapons. On 9 March 1992, China joined the NPT. And on 19 June 1992 the French National Assembly approved a law enabling the government to join the Treaty. Belarus, Kazakhstan and the Ukraine are committed to do so. But Israel, India and Pakistan refuse to join the Treaty. Israel showed its attitude to the international non-proliferation regime when it bombed Iraq’s Osirak nuclear reactor in June 1981. Israel said that, even though Iraq is a party to the NPT, Iraq would have used the reactor to produce plutonium for nuclear weapons. We now know that Israel was correct in its belief about Iraq’s nuclear-weapon intentions. Israel has never believed that the NPT, or any other international measure, is reliable enough to satisfy Israel that its national security is being adequately protected. Israel intends, therefore, to take any measure necessary, including military action, to stop other Middle East countries acquiring nuclear weapons. India and Pakistan also believe that the Treaty would not effectively protect their security. Until recently, South Africa was another country which refused to join the Treaty. But early in 1992 it joined. Many believe that South Africa has produced a stock of highlyenriched uranium for nuclear weapons and may have actually manufactured some nuclear weapons.

9 INDIA’S NUCLEAR PROGRAMME India has a long history of activities in nuclear science and technology. India’s early nuclear start was due to the enthusiasm of Homi J. Bhabha, who headed the Indian nuclear programme for nearly twenty years. Bhabha began promoting nuclear research three years before India became independent in 1947. One of the first actions taken by the Indian government after independence was to set up the Atomic Energy Commission in 1948. India’s first research reactor, Apsara, designed and built indigenously with the fuel supplied by the UK, became operational in 1956 at the Bhabha Atomic Research Centre at Trombay.

INDIA’S NUCLEAR FUEL CYCLE For a Third World country, India has an ambitious nuclear programme. It has developed and constructed a self-sufficient nuclear fuel cycle, with uranium mines and mills, a uranium purification (UO2) plant, fuel fabrication plants, plutonium reprocessing plants, nuclear-power reactors and research reactors. In addition, it has a small uranium conversion UF6 plant, a pilot uranium-enrichment plant, and heavy-water production plants. The Indian government is very secretive about its nuclear programme, presumably because of the military aspects of it. India has significant reasonably assured reserves of about 50,000 tonnes of uranium. Uranium is mined and milled at Narwapahar and Turamdih in Bihar and at Jaduguda. Uranium production is about 200 tonnes a year. India also has substantial resources of thorium, estimated at about 320,000 tonnes, much of it in the state of Kerala. India is developing a nuclear fuel cycle for thorium. The fuel fabrication plants are at Hyderabad, Tarapur and Trombay. The Hyderabad plant makes about 80 tonnes of fuel elements a year for India’s nuclear-power reactors and for the blanket of the Fast Breeder Test Reactor. Its capacity is being expanded to about 225 tonnes a year. The Trombay plant produces annually 135 tonnes of fuel elements for research reactors, including elements for the Fast Breeder Test Reactor. The plant also produces thorium oxide which is used for the production of U-233 reactor fuel. The Tarapur plant produces about 20 tonnes a year of MOX reactor fuel, involving the use of about 65 kilogrammes of plutonium a year. Plutonium reprocessing plants, using the Purex process, are operating at Tarapur and Trombay. The Tarapur plant can reprocess about 100 tonnes of spent fuel elements a year; producing about 140 kilogrammes a year when operating at full capacity. It has

India's nuclear programme

60

been operating at full scale since 1982. The Trombay plant can reprocess about 30 tonnes of spent fuel elements a year from research reactors—specifically from the CIRUS and Dhruva reactors. A larger reprocessing plant is under construction at Kalpakkam. It should have the capacity to reprocess about 125 tonnes of spent reactor fuel elements a year and will have a separate stream to reprocess fuel elements from a fast breeder reactor. India has eight plants producing heavy water, some producing heavy water in connection with a fertilizer plant. The heavy water is used in pressurized heavy-water reactors. The total capacity of the plants is about 660 tonnes of heavy water a year, although the actual production is significantly lower. The plants are sited at Baroda (45 tonnes a year); Hazira (110 tonnes a year); Kota (85 tonnes a year); Manuguru (185 tonnes a year); Nangal (14 tonnes a year); Talcher (65 tonnes a year); Thal-Vaishet (110 tonnes a year); and Tuticorin (50 tonnes a year). Two more plants are under construction (Spector 1990). India is experimenting with gas centrifuges for the enrichment of uranium. It has a small UF6 plant producing enough material for the experimental enrichment programme. Both activities are taking place at the Bhabha Atomic Research Centre at Trombay. The pilot plant has had 100 centrifuges operating since the mid-1980s. It is reported that a larger enrichment plant, containing several hundred centrifuges, is operating near Mysore (Albright and Hibbs 1992f).

INDIA’S NUCLEAR-POWER PROGRAMME Until recently, the government’s plan was to have enough nuclear-power reactors operating by the year 2000 to generate 10,000 megawatts of electricity (MWe). This has been cut back to a more modest 5,770 MWe. At the end of 1991, India had seven nuclear-power reactors in operation, generating about 1,400 MWe, and another seven under construction, which will generate abut 1,500 MWe for a total of 2,900 MWe. The government plans to build another five reactors with a total generating capacity of 940 MWe. It will be very difficult, to say the least, for India to have more than these nineteen reactors, with a combined capacity of 3,840 MWe, operating by the year 2000. India will, therefore, fall far short of its target of 5,770 MWe by the year 2000. India apparently has ambitions to develop and construct fast breeder reactors and is now operating a small Fast Breeder Test Reactor. But the future of this programme is very doubtful. Indian nuclear-power reactors, which currently generate less than 2 per cent of India’s total electricity needs, are relatively small reactors. The first two, which began operating in 1969, were supplied by the USA. They are boiling-water reactors, fuelled with lowenriched uranium, each generating 160 MWe, and sited at Tarapur, Maharashtra. The third reactor, which began operating in 1972, was supplied by Canada and is a pressurized heavy-water reactor (cooled and moderated by heavy water) generating 207 MWe. The fourth reactor is of the same type and generating capacity as the third but was

How nuclear weapons spread

61

supplied by an Indian company, Larson & Toubro. These two heavy-water reactors are at Kota, Rajasthan. The other three operating Indian nuclear-power reactors are all pressurized heavy-water reactors, each generating 220 MWe, and are of Indian design. Two, supplied by Larson & Toubro, are operating at Kalpakkam, Tamil Nadu. The other, supplied by Walchandnagar Industries of India, is sited at Narora, Uttar Pradesh. A second 220-MWe pressurized heavy-water reactor, supplied by the Indian firm Richardson & Cruddas, will begin operating at Narora in 1993. The construction of two more of these reactors, at Kakrapar, Gujarat, is well advanced; they are being supplied by Larson & Toubro, and by Walchandnagar Industries. Another two 220-MWe pressurized heavy-water reactors are being built at Kaiga, Karntaka, and two more are being built at Kota, Rajasthan. They are being supplied by a number of Indian firms. The two Tarapur light-water reactors use low-enriched fuel, originally supplied by the USA. In 1982, the USA demanded full-scope safeguards as a condition of supply; France then took over as the supplier. But India will receive no more fuel from France after the present contract ends in 1993 because France is now a party to the NPT and requires fullscope safeguards on fuel supplies, a demand India rejects. The Indian capacity to produce enriched uranium is far less than is required to fuel the two Tarapur reactors, and will probably remain so for the foreseeable future. The reactors may, therefore, have to be shut down. All the Indian pressurized heavy-water reactors are fuelled with natural uranium. The fuel elements for them are made in India using Indian uranium. These reactors are, therefore, independent of foreign fuel supplies. If all goes according to schedule, all the Indian reactors under construction should begin operating before the end of the 1990s. The plan is then to construct larger reactors, beginning with two 470-MWe pressurized heavy-water reactors at Tarapur, Maharashtra.

RESEARCH REACTORS India operates a number of research reactors. Four are operating at Trombay (Apsara, CIRUS, PURNIMA and Dhruva) and two at Kalpakkam (the Fast Breeder Test Reactor and Kamini). The Apsara reactor is a light-water pool reactor using 80 per cent enriched uranium as fuel. The reactor contains about 4 kilogrammes of fuel and generates 400 kilowatts of thermal power—kW(th). It began operating in 1956. CIRUS is a heavy-water tank reactor, fuelled with 10 tonnes of natural uranium and generating about 40 megawatts of thermal power—MW(th). Supplied by Canada, it became fully operational in 1960. Originally, Canada provided the fuel and the USA the heavy water. The reactor is now fuelled from Indian sources. PURNIMA is an indigenously designed and built fast reactor, generating only a few watts of power. It began operating in 1972, originally fuelled with plutonium dioxide pellets clad in stainless steel. It was converted to use U-233 fuel, produced in India, and began operating with this fuel in 1989. Dhruva is a heavy-water tank reactor, fuelled with 6.5 tonnes of natural uranium, indigenously supplied, and generating 100 MW(th) of power. The reactor has been

India's nuclear programme

62

operating since 1985. The heavy water has been supplied by India, Norway, the ex-Soviet Union and possibly Romania. India has had to import heavy water for this reactor and for the Madras nuclear-power reactors because its domestic heavy-water plants have produced less heavy water than expected. The supply of heavy water produced in Indian plants lags behind demand. Since the early 1980s, India has been buying heavy water abroad. The Fast Breeder Test Reactor is a fast breeder reactor, generating about 40 MW(th) or 15 MWe. It was supplied by India but was based on a French design. The fuel, developed and manufactured in India, is a mixture of plutonium and natural-uranium carbide. The reactor began operating in 1985. Kamini is a small reactor fuelled with U-233, generating about 30 kW(th). Supplied by India and fuelled with Indian fuel, it began operating in 1988. India hopes to use in the future U-233, obtained from thorium, as a major reactor fuel; the Kamini and PURNIMA research reactors are part of this programme.

INTERNATIONAL SAFEGUARDS India has not signed the NPT. There are, therefore, no safeguards agreements with the IAEA in connection with the Treaty. But India is a member of the IAEA and accepts IAEA safeguards on some of its nuclear facilities, which are regularly inspected to verify that they and the nuclear materials they produce are not used for any military purpose. Safeguards agreements between India and the IAEA cover imported materials, facilities and equipment, particularly those imported from Canada and the USA, but not those produced by India. This means that only four of India’s nuclear-power reactors— two at Rajasthan and two at Tarapur—and none of the research reactors are under safeguards. Safeguards are applied to: the Hyderabad fuel fabrication plant only when it is making fuel from safeguarded material; the Tarapur fuel fabrication plant only when it is making fuel from plutonium produced in safeguarded reactors; and to the Tarapur reprocessing plant only when it is reprocessing spent fuel from safeguarded reactors. For a decade or more, India has been buying heavy water from abroad clandestinely so that the reactors using it do not have to be put under IAEA safeguards. In short, India has a complete fuel cycle not open to IAEA inspection. India is under no obligation to demand full-scope safeguards to its nuclear exports, although it has said that it would require that IAEA safeguards would apply. India could export small research reactors and the technology to produce heavy water and to reprocess spent reactor fuel elements. India has offered to sell a small reactor to Iran. If it does so it will be the first Indian export of nuclear technology. Because it is developed indigenously, the reactor would not have to be subject to IAEA safeguards.

How nuclear weapons spread

63

PLUTONIUM PRODUCTION As described above, except for the Tarapur plant, all of India’s nuclear-power reactors, in operation and under construction, use heavy water to moderate them. Heavy-water reactors can be operated to produce weapon-grade plutonium. The Dhruva research reactor is estimated to be capable of producing about 25 kilogrammes a year, if operated just to produce plutonium. The CIRUS reactor, which produced the plutonium for India’s nuclear test in 1974, can produce about 10 kilogrammes of plutonium a year. The CIRUS reactor was supplied by Canada on the understanding that it was used only for peaceful purposes, but it did not require IAEA safeguards on the reactor. India claimed that it did not violate its understanding with Canada because its nuclear device was for ‘peaceful’ purposes. India has three heavy-water nuclear-power reactors operating without IAEA safeguards. The plutonium produced in these reactors could be used to fabricate nuclear weapons. The amount of plutonium produced in the three unsafeguarded reactors is not publicly known. Heavy-water reactors are efficient producers of plutonium and each 220MWe reactor could produce about 55 kilogrammes of plutonium a year, for a total of 165 kilogrammes of plutonium a year, enough for about thirty nuclear weapons a year. The research reactors can produce enough plutonium for another seven nuclear weapons a year. In practice, the Indian reactors have not operated at anything like their design capacity nor have the reprocessing plants. Also, some plutonium has been used in the Fast Breeder Test Reactor and for other research purposes. Taking these factors into account, it is estimated that India may have in stock in mid-1992 about 300 kilogrammes of weaponsgrade plutonium, enough to make about sixty nuclear weapons. About two-thirds of this plutonium has come from the CIRUS reactor, and most of the rest has come from the Dhruva reactor. By the end of 1995 the amount of weapons-grade plutonium in stock may have increased to about 420 kilogrammes, enough for about eighty-five nuclear weapons. The extra 100 kilogrammes will come from the Dhruva reactor (Albright et al. 1993). All of this plutonium is free from IAEA safeguards and from any restrictions from foreign nuclear suppliers. According to official Indian statements, however, India’s plutonium will be used in India’s breeder-reactor programme. It is, however, very unlikely, to say the least, that India will construct commercial breeder reactors in the foreseeable future. Indian officials consistently claim that India does not currently have any nuclear weapons. They suggest, however, that it might make them if Pakistan acquires a significant nuclear force. If India does not have nuclear weapons, it could produce a significant nuclear arsenal in, at most, a few years. The Indian nuclear test demonstrated conclusively India’s capability to design and fabricate nuclear weapons. Indian nuclear scientists have been pressing for a nuclear test for years. The first proposal was apparently made to the government, and agreed to, in January 1965. The

India's nuclear programme

64

Indian nuclear test took place on 18 May 1974 about 100 metres underground in the Rajasthan desert in north-western India. The design of the nuclear device was similar to that of the Nagasaki bomb. It probably used about 6 kilogrammes of plutonium and had an explosive power equivalent to that of 12 kilotonnes of TNT. The Atomic Energy Commission of India described it as ‘a peaceful nuclear explosion experiment using an implosion device’. There is, however, no fundamental difference between the technology of ‘peaceful’ and military nuclear explosives.

10 PAKISTAN’S NUCLEAR PROGRAMME Pakistan has had ambitions to deploy nuclear weapons for a long time. In 1969 Pakistan’s former Prime Minister, Ali Bhutto, wrote: Our plans should include the nuclear deterrent…. If Pakistan restricts or suspends her nuclear programme, it would not only enable India to blackmail Pakistan with her nuclear advantage, but would impose a crippling limitation to the development of Pakistan’s science and technology. Our problem, in its essence, is how to obtain such a weapon in time. (Mochaver 1991) Pakistan has, in fact, had a significant nuclear programme since the 1960s and has built up a group of very competent nuclear scientists and technologists, including the Nobel Prize winner Abdul Salem. Whereas India chose the plutonium route to nuclear weapons, Pakistan decided to use highly-enriched uranium. In February 1992, Pakistan’s Foreign Secretary Shahryar Khan stated that Pakistan had enough components for one or more nuclear weapons.

PAKISTAN’S NUCLEAR ACTIVITIES Pakistan’s uranium resources are estimated to be 20,000 tonnes but this is uncertain. The Pakistan Atomic Energy Commission operates a uranium mine at Bagalchur near Multan and two uranium mills. How much uranium is produced is not publicly known. Pakistan has a nuclear-power reactor—the KANUPP reactor located at Paradise Point on the Arabian Sea, about 50 kilometres from Karachi. KANUPP is a heavy-water reactor supplied and constructed by Canadian General Electric with a generating capacity of 125 MWe. It began operating in 1972 with Canada supplying the natural uranium fuel. But the supply of fuel and spare parts for the plant was stopped by Canada in 1976. Since then KANUPP has been operating on fuel and spare parts produced in Pakistan. Pakistan is buying a 300-MWe pressurized-water reactor from China, to be constructed at Mianwali, in the Punjab, 150 kilometres north of Islamabad. Pakistan was hoping to buy a 900-MWe reactor from France and the sale was approved by the French government in 1990. But negotiations were suspended after the Gulf War. The later decision by France to join the NPT has, to say the least, complicated the negotiations because France is now obliged to apply full-scope safeguards to its nuclear exports.

Pakistan's nuclear programme

66

China has also joined the NPT. How China’s reactor deal with Pakistan will be affected remains to be seen. In addition to its power reactor, Pakistan operates a research reactor at the Pakistan Institute of Nuclear Science and Technology (Pinstech) at Rawalpindi. The pool-type reactor, supplied by the USA, went into operation in 1965. Its power output was originally 5 MW(th) but it has recently been upgraded with indigenous Pakistani resources to 10 MW(th). The core of the reactor has been redesigned so that its fuel, which was uranium enriched to 93 per cent U-235, is now uranium enriched to 20 per cent U-235. The old fuel was supplied by the USA. The new fuel is said to have been fabricated in China, presumably using enriched uranium produced in Pakistan. Pinstech also operates another small research reactor, having a power output of about 400 kW(th). This pool-type reactor uses enriched-uranium fuel and was installed by China in 1990. Pakistan makes the natural-uranium fuel elements for the KANUPP reactor at the fuel fabrication plant at Chasma. The capacity of the plant is not publicly known but the reactor needs about 15 tonnes of fuel a year and presumably the plant can provide this. The heavy water for the KANUPP reactor was originally supplied by Canada and the USA. Pakistan has built two heavy-water plants, one at Multan with a production capacity of 13 tonnes a year and the other at Karachi with an unknown capacity. Pakistan has built a small processing plant at Pinstech, Rawalpindi, said to be capable of reprocessing about 15 tonnes of fuel a year, producing between 10 and 15 kilogrammes of plutonium a year. Whether or not the plant is currently operating is not known. A larger reprocessing plant has been partially constructed at Chasma. This plant should, according to an agreement signed in 1976, have been built by the French company Saint-Gobain Techniques Nouvelles. But France withdrew from the project in 1977, after heavy pressure from the USA. But its construction may be continuing with French, Italian, Turkish and Swiss firms helping the Pakistanis. The plant is said to be capable of reprocessing about 100 tonnes of spent reactor fuel a year, separating about 150 kilogrammes of plutonium. To produce plutonium for a clandestine nuclear-weapon programme, Pakistan would need a reactor, which is not under safeguards, capable of producing significant amounts of plutonium for reprocessing. Although there is speculation that Pakistan has built such a reactor there is no reliable information that it has done so. Pakistan is, however, capable of building and operating such a plutonium-production reactor, with an output of about 50 MW(th), perhaps fuelled with natural uranium and moderated with graphite, able to produce enough plutonium for two or three nuclear weapons a year.

PAKISTAN’S URANIUM-ENRICHMENT PROGRAMME Pakistan refuses to join the NPT or accept full-scope safeguards because it operates a secret uranium-enrichment plant, producing material for nuclear weapons. This plant is at Kahuta, about 30 kilometres south-east of Islamabad. A small experimental plant has been operating at Sihala, near Kahuta, since 1979, giving Pakistani scientists and

How nuclear weapons spread

67

engineers experience in constructing and operating gas centrifuges for enriching uranium. The Kahuta and Sihala establishments were built using materials, technology, and equipment obtained secretly from some European countries, including Switzerland, the UK, the Netherlands and West Germany; some was smuggled from the USA. The designs of gas centrifuges were obtained from the URENCO uranium-enrichment plant at Almelo in the Netherlands. Knowledge about URENCO centrifuges was reportedly acquired by the Pakistani metallurgist Abdul Qadeer Khan, when he worked between 1972 and 1975 for a Dutch company associated with the URENCO plant. On leaving the Dutch firm, Dr Khan returned to Pakistan and started the Pakistani uranium-enrichment programme. That the Kahuta centrifuges are based on Khan’s knowledge of URENCO centrifuges is the conclusion of an investigation made by the Dutch Ministry of Foreign Affairs. No important details of the heavily guarded Kahuta plant have been publicly released. But it is thought that two types of centrifuges are used, based on URENCO centrifuges designed by German scientists—the G-1 and G-2 types. The amount of separation an enrichment plant can achieve is measured by separative work units (SWUs). SWUs are complicated units because the amount of work needed to enrich uranium to a given concentration of U-235 depends on the concentration of U-235 in the uranium hexafluoride gas fed into the centrifuge plant and the concentration of U235 in the uranium hexafluoride gas at the end of the centrifuging operation. For example, if natural uranium hexafluoride (containing 0.7 per cent U-235) is fed into the plant and the concentration of U-235 in the waste gas is 0.3 per cent, it takes 200 SWUs of separation work to produce a kilogramme of uranium enriched to 94 per cent, a concentration suitable for use in nuclear weapons. The amount of separative work needed to produce weapon-grade uranium depends greatly on the concentration of U-235 in the feed material. The separative work needed to increase the concentration of U-235 in uranium from its natural value of 0.7 per cent to 5 per cent is about 70 per cent of that needed to go from natural uranium to 94 per cent enrichment, i.e. to weapon-grade uranium. If, therefore, the feed material is somewhat enriched (perhaps acquired from a foreign supplier), the separative work needed to produce weapon-grade uranium is considerably less than that required if the input is natural uranium hexafluoride. The number of centrifuges operating in the Kahuta plant is a closely guarded secret. According to Albright and Hibbs (1992e) the plant contains 14,000 centrifuges. But even the most advanced operators find it very difficult to keep centrifuges operating continuously; they break down quite often and need much maintenance. It is, therefore, likely that although Pakistan may have this number of centrifuges, only about 3,000 are actually in operation at Kahuta at any one time. Albright and Hibbs (ibid.) believe the centrifuges in the Kahuta plant are of mixed types, based on the G-l and G-2 centrifuges, with most of them being based on the G-2 type. The capacity of a G-l centrifuge is about 2.5 SWUs a year and that of a G-2 is about 5 SWUs a year. It is reasonable to assume that, on average, the capacity of each centrifuge is about 4 SWUs a year. If 3,000 centrifuges are operating in the Kahuta plant at any one time, the total capacity will be about 12,000 SWUs a year.

Pakistan's nuclear programme

68

It is also reasonable to assume that natural uranium hexafluoride is the feed material for Kahuta and that about 0.3 per cent of the U-235 is left in the waste. The Kahuta plant will then produce about 60 kilogrammes of weapon-grade uranium a year. Pakistani nuclear-weapon designers could probably fabricate a nuclear weapon with between 15 and 20 kilogrammes of weapon-grade uranium. The Kahuta plant could, therefore, produce enough weapon-grade uranium to produce three or four nuclear weapons a year. Kahuta has been producing weapon-grade uranium since 1986. But, according to Albright and Hibbs (ibid.): Kahuta has not operated at nominal capacity for most of its history. By the end of 1991, Pakistan had probably produced between 100 and 200 kilogrammes of weapon-grade uranium, based on a variety of tails assays (i.e. concentration of U-235 in the wastes) and separative capabilities. This is enough material for roughly 6 to 13 nuclear explosive devices. Pakistan is apparently building another uranium-enrichment plant using gas centrifuges at Golra, some 10 kilometres west of Islamabad. Construction is said to have begun in 1987 but it seems that Pakistan is having problems acquiring equipment for Golra from abroad so that the construction of the new plant is a slow process (Albright et al. 1993). The uranium hexafluoride for Pakistan’s gas centrifuges is produced at a plant at Multan which has a capacity of about 200 tonnes a year. It was supplied by the German firm CES Kalthof GmbH, Freiburg, and has been operating since 1982. According to US officials, Pakistan has received a copy of a nuclear-weapon design from China. This was probably an implosion-type weapon requiring about 15 kilogrammes of weapon-grade uranium. There have been a number of reports claiming that Pakistan has tested non-nuclear components of nuclear weapons, including tests of assemblies of conventional highexplosive lenses for implosion. A useful test of an implosion assembly would use a dummy sphere of natural or depleted uranium. There are reports that Pakistan is developing boosted fission weapons and even thermonuclear weapons. In 1987, Pakistan acquired illegally a small quantity of pure tritium from West Germany. This could be used in neutron initiators for nuclear weapons. And Pakistan’s attempts in the late 1980s to build a tritium purification plant with a capacity of between 0.5 and 1 gramme of tritium a day, with West German assistance, may be related to a wish to develop boosted fission weapons.

INTERNATIONAL SAFEGUARDS Pakistan is a member of the IAEA. The KANUPP nuclear-power reactor and the PARR research reactor at Rawalpindi are under IAEA safe-guards. No other Pakistani facilities are. There is an agreement between Pakistan, France and the IAEA to safeguard the Chashma reprocessing plant which has been concluded but is not yet in force.

11 ISRAEL’S NUCLEAR PROGRAMME In spite of the official secrecy, we know quite a lot about Israel’s nuclear-weapon programme. Much of this knowledge comes from the revelations of Mordechai Vanunu who gave the London-based Sunday Times top-secret information about Israel’s production of plutonium at the Dimona nuclear establishment in the Negev desert. Vanunu worked as a technician at Dimona between 1976 and 1985. His story was told in an article, entitled ‘Revealed: the secrets of Israel’s nuclear arsenal’, published in the Sunday Times on 5 October 1986 (Barnaby 1989). Vanunu was abducted from London, lured to Rome, and forcibly taken to Israel and, at the end of 1987, tried in the Jerusalem District Court for treason and the collection of secret information with intent to impair the security of the state. He was sentenced to eighteen years’ imprisonment, which he is now serving. The nature of the charges can be seen as an admission that the information given by Vanunu, or at least the bulk of it, is true. The credibility of Vanunu’s information was much enhanced by a collection of fifty-seven photographs of various processes and equipment taken clandestinely by Vanunu in Dimona.

ISRAEL’S NUCLEAR ACTIVITIES Israeli efforts to acquire the technology, personnel, and materials to produce nuclear weapons date back almost as far as the birth of the state in May 1948. Much of the credit for this must go to Chaim Weizmann, the biochemist elected to be Israel’s first President. A Department of Isotope Research was set up before the end of 1948 at the Weizmann Institute, which soon became a world-renowned scientific institute. It is likely that a group of Israelis soon set their eyes on acquiring a nuclear-weapon force, believing that this would offset Israel’s inferiority in military manpower and quantity of conventional armaments, compared with its Arab enemies. To pursue this programme, the group knew that Israel would need a nuclear reactor and a reprocessing facility. The first step was to find the uranium needed to fuel a reactor. Surveys carried out by the Ministry of Defence found that uranium is mixed with phosphate deposits in the Negev desert. The deposits contain between 0.01 and 0.2 per cent of uranium and the Negev deposits are estimated to contain between 30,000 and 60,000 tonnes of uranium. It is estimated that about 100 tonnes of uranium are produced a year (Spector 1990). Research into the production of heavy water also began soon after Israel was born and

Israel's nuclear programme

70

a new method of production was developed. A heavy-water plant, built at Rehovot, was in operation by 1954. In 1955, Israel signed an agreement with the USA under which the USA supplied Israel with a small research reactor. Built at the Nuclear Research Centre at Nahal-Soreq, the 5-MW(th) reactor began operating in 1960. The pool-type light-water reactor is fuelled with highly-enriched (90 per cent) uranium. In 1957, the Israel Atomic Energy Commission signed an agreement with France’s Commissariat a Energie Atomique under which France agreed to build the nuclear facilities at the nuclear establishment near Dimona in the Negev desert. In return, Israel supplied France with the details of its special method of heavy-water production (probably by hydrogen distillation) (Hersh 1991).

THE DIMONA ESTABLISHMENT The Dimona Nuclear Research Centre is Israel’s most advanced nuclear institution. Whereas the Nahal-Soreq Nuclear Research Centre is a civilian institution, Dimona is a secret, heavily guarded military establishment. Dimona is defended against air attack with surface-to-air missiles and no unauthorized aircraft can fly over it. The Armament Research and Production Administration and the Armament Development Authority work with Dimona on nuclear-weapon research and development. The most important facilities at the Dimona establishment are the reactor and a reprocessing facility, both built with the assistance of French scientists and engineers. The Dimona reactor, moderated with heavy water and fuelled with natural uranium, is an efficient producer of plutonium. It came into operation in 1963. Yellow cake is converted into UO2, and the fuel elements for the reactor are manufactured at the Dimona establishment. It is generally believed that the original power output of the Dimona reactor was 26 MW(th). According to Vanunu, the power of the reactor was increased to 70 MW(th) at some time before 1976. It was apparently increased again, presumably to about 150 MW (th). The rate of plutonium production reported by Vanunu requires the latter power output (Barnaby 1989). To make the power output of the reactor five or six times greater without considerably increasing the physical size of the reactor would have been difficult. It is possible that the reactor may have been originally cooled using a gas, probably carbon dioxide. When the Dimona reactor was being built the French were constructing a gas-cooled, heavy-water moderated reactor (the EL-4 reactor) and may have used the same design for both reactors. The power of the Dimona reactor may have been increased by adding a heavywater pressure circuit to cool the reactor with heavy water rather than a gas. Cooling with a liquid is much more efficient than cooling with a gas so that a considerably greater output could be obtained with the same amount of uranium fuel. The power could have also been increased by using enriched uranium (the EL-4 used uranium enriched to about 1.5 per cent) instead of natural uranium, although this is unlikely. There are reports that the reactor was deliberately built with a greater volume

How nuclear weapons spread

71

than was necessary for a power of 26 MW(th). If this is so, more fuel elements were packed into the core to increase the power output. Pierre Pean reports that the cooling circuits supplied for Dimona were three times larger than needed for a 26 MW(th) reactor (Pean 1982). Whatever method, or combination of methods, was used in the late 1960s or early 1970s to boost the power of the Dimona reactor, it would have been a difficult task. That it was achieved testifies to the competence of Israeli nuclear technologists. Having done so, they would have found the task of designing and fabricating nuclear weapons not very taxing. About 22 tonnes of natural uranium, in the form of solid metal cylindrical rods, were needed to fuel the Dimona reactor. It is believed that about a half of this came from Israel’s own stocks, produced as a by-product of phosphate production. The other half was bought from foreign suppliers. Since France built the reactor it would be odd if it did not supply some of the fuel. In addition, it has been reported that other suppliers may have included Argentina, Belgium, the Central African Republic, Gabon, Niger and South Africa. Israel may also have received pirated uranium. In 1968, about 200 tonnes of uranium oxide (yellow cake) were being shipped from Antwerp to Genoa in The Plumbat. But the uranium never reached its destination. The ship was diverted while in the Mediterranean, reportedly to Israel. It is also reported that Israel received about 100 kilogrammes of highly-enriched uranium stolen from the Nuclear Materials and Equip-ment Corporation (NUMEC) between 1962 and 1968. The missing uranium was enough to make about five nuclear weapons. If Israel did get this uranium, it could have been used to produce a small nuclear force before plutonium from the Dimona reactor became available in significant quantities. Israel also needed to import some of the heavy water needed for the Dimona reactor. The Rehovot plant can almost certainly produce enough heavy water to make good the losses from the Dimona reactor. But the initial charge was imported. About 20 tonnes of heavy water was imported from Norway, more than enough to start Dimona operating. If the power output of the reactor was increased from 26 MW(th) to 70 MW(th) and then to 150 MW(th), Israel would have needed to import a total of about 80 tonnes of heavy water. Israel obtained about 4 tonnes of heavy water from the USA; the rest probably came from France. According to Vanunu, the reprocessing facility at Dimona, which began operating in 1966, uses the Purex process to separate plutonium from the fuel elements after they have been removed from the reactor. The chemical agent used for the separation is tributyl phosphate dissolved in a kerosene hydrocarbon. Tributyl phosphate is derived from phosphoric acid which Israel produces in significant quantities. A total of about 40 kilogrammes a year of weapon-grade plutonium is separated at Dimona’s reprocessing plant. The plutonium leaves the reprocessing facility as plutonium nitrate. Plutonium is precipitated as the oxalate which is thermally degraded to the dioxide (PuO2). The dioxide is then converted to plutonium tetrafluoride which is reduced with calcium to plutonium metal.

Israel's nuclear programme

72

The plutonium is machined into solid spheres, each weighing about 4.5 kilogrammes. The spheres, encased in airtight copper shells because plutonium may ignite spontaneously in air, are the fissile components of Israel’s nuclear weapons. The plutonium sphere is surrounded by a beryllium shell which acts as a tamper and neutron reflector. According to Vanunu’s information, the Israelis produced at Dimona about 400 kilogrammes of plutonium between 1976 and 1985; they probably produced about 150 kilogrammes before 1976. And since 1985 they may have produced about 250 kilogrammes. The total plutonium production up to 1992 may, therefore, amount to 800 kilogrammes, enough for about 150 nuclear weapons.

LITHIUM AND TRITIUM PRODUCTION Vanunu gave information about the production of lithium-6 at Dimona. The lithium-6 is used to produce tritium, by bombarding lithium-6 with neutrons in the reactor. Lithium-6 is also compounded with deuterium to produce lithium-6 deuteride which can be used as the fusion material in a thermonuclear weapon. Natural lithium contains two isotopes—lithium-6 and lithium-7. Lithium-6 is separated from lithium-7 at Dimona by producing an amalgam of lithium and mercury by electrolysis. Lithium hydroxide is then passed through the amalgam. Separation of lithium-6 from lithium-7 occurs by exchange between the amalgam and the aqueous solution of lithium hydroxide. Six vertical exchange columns, each 13 metres long, are used in the process. The lithium-7 is concentrated in the amalgam phase and the lithium hydroxide is enriched in lithium-6. By repeating the exchange process through each of the six columns, the concentration of lithium-6 is, according to Vanunu, increased from the natural value of 7.42 per cent to about 85 per cent. Vanunu said that lithium-6 production went on for three years, from 1984 to 1987. A total of about 170 kilogrammes of lithium-6 was produced. This amount of lithium-6 would produce about 220 kilogrammes of lithium-6 deuteride, enough to fabricate about thirty-five typical thermonuclear weapons. Small rods of lithium-6 are irradiated with neutrons in the Dimona reactor, producing tritium, helium, and hydrogen. Tritium and hydrogen are separated by passing them through a column of palladium asbestos; helium and tritium are separated by passing them through a column of palladium and mercury. The tritium gas is stored by absorbing it on powdered uranium. When the tritium is required, the uranium is heated and the tritium gas comes off. With supplies of tritium and lithium-6 deuteride, Israel could produce boosted fission weapons and thermonuclear weapons.

How nuclear weapons spread

73

URANIUM ENRICHMENT According to Vanunu, a secret unit in Dimona has been enriching uranium using gas centrifuges since about 1980. And, in 1981, the Israelis began using laser beams to separate uranium isotopes. How much enriched uranium is being produced at Dimona by these processes is not known but Vanunu stated that the laser-enrichment process was being expanded to production scale when he left the Dimona establishment in 1985. It is hardly surprising that the Israelis are using laser enrichment. They are leaders in the field.

HAS ISRAEL TESTED A NUCLEAR WEAPON? Israeli nuclear-weapon designers are likely to be very confident that they can design and construct ordinary fission weapons without full-scale nuclear tests. A boosted fission weapon, with some deuterium and tritium gases fed into the centre of a plutonium sphere, has the same basic design as a non-boosted weapon; the difference is technical rather than scientific. The Israeli military is sufficiently scientifically minded to accept the word of the nuclear scientists that they can predict quite precisely the explosive yields of any nuclear-fission weapons they design. The Israelis may, therefore, be prepared to deploy both types without a nuclear-testing programme. But they would not be prepared to deploy full-scale thermonuclear weapons (with a solid lithium-deuteride fusion component sited outside a fission trigger) without testing them. The design of these weapons is too complex to deploy them without several (probably between five and ten) tests. This is not to suggest that the test of a thermonuclear weapon need involve the whole assembly at the entire explosive yield. It is only necessary to test the fission trigger and a small section of the fusion stage to check that some fusion occurs. If some fusion takes place, it can be assumed that the whole weapon will work effectively. Whether or not Israel has conducted one or more nuclear tests is a controversial issue. An event took place on 22 September 1979 over the South Atlantic/Indian Ocean region which was detected by an American Vela satellite, which was designed to detect nuclear explosions in the atmosphere and outer space. The scientists at the Los Alamos nuclearweapon laboratory who analysed the information obtained by the Vela satellites, were, and still are, confident that the signal picked up from the Indian Ocean came from a nuclear explosion. Although there is other evidence of a nuclear explosion, a panel of scientists set up by ex-President Carter’s White House concluded that the evidence was not conclusive (see Chapter 6). If a nuclear explosion has taken place over the Indian Ocean, there are good reasons to believe that it was a joint Israeli-South African test involving an Israeli nuclear explosive and South African warships which were conducting a secret exercise at the same latitude and longitude as the nuclear explosion may have taken place. The Vela analysts estimated

Israel's nuclear programme

74

that, if a nuclear explosion took place, it was probably a low-yield device, producing an explosion of about 2 kt. This could have been a test of a trigger for a thermonuclear weapon.

12 IRAQ’S NUCLEAR-WEAPON PROGRAMME The case of Iraq shows how a country with a significant peaceful nuclear programme has a cadre of trained nuclear scientists and engineers able to design and fabricate nuclear weapons. Iraq’s peaceful nuclear programme became well known as a result of the worldwide publicity which followed the destruction by bombing of the Osiraq research reactor in June 1981. The bombing, carried out by Israeli aircraft, completely destroyed the reactor even though the core was enclosed in a thick concrete containment vessel. The Osiraq reactor, ordered from France in 1976, was being built at the Tuwaitha nuclear research centre located in the suburbs of Baghdad. It was scheduled to begin operating a few months after the bombing. Osiraq was to be a light-water reactor, fuelled with uranium enriched to a concentration of 93 per cent in U-235. The fuel for the reactor, which would have had a power output of 40 MW(th), was supplied by France. Israeli nuclear scientists believed that, although the reactor was not suited to produce militarily significant amounts of plutonium directly, it could have been modified to do so in a relatively short time by surrounding the reactor core with uranium. The neutron irradiation of the uranium would produce plutonium in the uranium. Israeli suspicions were aroused when Iraq ordered from a West German firm about 10 tonnes of depleted uranium. The uranium was to be supplied in the form of metal cylinders of such dimensions that they would have fitted into the reactor. The Israelis estimated that enough plutonium could, in theory at least, have been produced to fabricate one or two nuclear weapons a year.

IRAQ’S NUCLEAR PROGRAMME In August 1990, when Iraq invaded Kuwait, it was operating two research reactors at Tuwaitha. Both were light-water reactors fuelled with highly-enriched uranium and their main purpose was to produce radioisotopes for medical use, some of which were exported to other countries. Both reactors were destroyed by American bombers in January 1991. One of the reactors (IRT-5000) had been in operation since 1967. The pool-type reactor and its fuel were supplied by the Soviet Union. The fuel was mainly uranium enriched to 80 per cent in U-235 but smaller amounts of uranium enriched to 36 and 10 per cent were used. Its power output was 5 MW(th). The other reactor (Tammuz-2) was supplied by France, as was its fuel. This pool-type

Iraq's nuclear-weapon programme

76

reactor started operating in 1980, with a power output of 800 kW(th). It was fuelled with uranium enriched to 93 per cent in U-235. When the Tuwaitha establishment was bombed by American bombers during the 1991 Gulf War the Iraqis had a stock of fresh reactor fuel for the IRT-5000 reactor. Some was of 80 per cent enrichment containing 10.97 kilogrammes of U-235. The rest was of 36 per cent enrichment containing 1.27 kilogrammes of U-235. In addition, there was fresh fuel for the Tammuz-2 reactor with an enrichment of 93 per cent but containing only 372 grammes of U-235. There was also some fuel, enriched to 93 per cent, which had been irradiated in the reactors and which, therefore, could not readily be used in nuclear weapons. The amount of U-235 in this irradiated fuel was 35.58 kilogrammes. Iraq has been mining some uranium which occurs with phosphate deposits. These were being mined in the north. How much uranium was obtained is not publicly known. After the 1991 Gulf War, Iraq admitted to having some 400 tonnes of natural uranium in a number of forms including the metal and uranium oxide (yellow cake).

IRAQ’S NUCLEAR-WEAPON PROGRAMME There is no doubt that Iraq was developing nuclear weapons in a secret programme codenamed ‘Petrochemical Three’, set up both to design nuclear weapons and to produce fissile material for them. Petrochemical Three was given huge amounts of money and large numbers of scientists and engineers worked for it. The crucial question is: How long would it have taken Petrochemical Three to fabricate nuclear weapons? Two international agencies have been trying to answer this question —the United Nations Special Commission, based in New York, and the IAEA. United Nations Security Council Resolution 687, which ended the 1991 Gulf War, empowers the UN Special Commission to destroy Iraq’s weapons of mass destruction—biological and chemical as well as nuclear. Resolution 687 established procedures for the ‘destruction, removal, or rendering harmless’ of Iraq’s weapons of mass destruction and called for the development of a regime which would ensure over the longer term that Iraq’s capacity for manufacturing such weapons would not be redeveloped. The Special Commission has, up to September 1992, carried out about thirty inspections of Iraq to locate and dispose of these weapons and the facilities to produce them (IAEA 1992b). Iraq has, to say the least, not co-operated fully with the inspectors and, in some cases, has seriously harassed them. Moreover, Iraq has not yet completely disclosed all the details of its programmes for the development and procurement of prohibited weapons. Generally speaking, the Iraqis have co-operated with the destruction of weapons but not with the destruction of production facilities. The Special Commission and Iraqi political leaders are playing a cat-and-mouse game. Nevertheless, in the words of Ambassador Rolf Ekeus, the head of the United Nations Special Commission, ‘the greatest part of Iraq’s capability with regard to weapons of mass destruction and ballistic missiles has been accounted for and is being disposed of.

How nuclear weapons spread

77

And there is no reason to doubt that, within the foreseeable future, any remaining significant Iraqi capability will be searched out and destroyed. As Ekeus says: ‘there is not much left for the cat and mouse to play with’. Destroying Iraq’s chemical-weapon capability, for example, is a mammoth task, involving nearly 50,000 munitions containing chemical-warfare agents. The agents are Sarin nerve gas, mustard gas and CS gas. Iraq’s chemical stockpile will be destroyed in two plants built at al-Multhana. Destruction has just begun and will take between twelve and eighteen months. The IAEA was nominated in Resolution 687 to carry out ‘urgent on-site inspection and the destruction, removal, or rendering harmless’ of any nuclear material in Iraq usable in nuclear weapons and Iraq’s nuclear-weapon programme and then to develop a plan for the long-term monitoring and compliance by Iraq of its obligations under Resolution 687. Iraq’s nuclear-weapon programme has, so far as it is known, been virtually destroyed. The Special Commission tells the IAEA which Iraqi sites to visit and inspect. The IAEA does its inspections and reports back to the Special Commission. Between the end of the 1991 Gulf War and October 1992, the IAEA sent eleven inspection teams into Iraq. In these inspections of Iraqi nuclear facilities, the United Nations Special Commission has discovered that Iraq had been pursuing three different techniques to enrich uranium for use in nuclear weapons. Documents unearthed by inspectors show that Iraqi scientists were designing an implosion-type nuclear-fission weapon and a surface-to-surface missile, presumably as a delivery system for their nuclear weapon. The Iraqi nuclear-weapon programme involved a complex network of establishments for research into methods of producing nuclear material for use in nuclear weapons, the production of machines and equipment for uranium enrichment, the operation of machines for the production of enriched uranium for use in nuclear weapons, the design of nuclear weapons and research and testing of the high explosives needed for nuclear weapons. These establishments have been destroyed by the United Nations inspectors working for the Special Commission. Iraq faces an indefinite period of United Nations surveillance and inspections to prevent it re-establishing programmes to develop nuclear, chemical, and biological weapons and the prohibited ballistic missiles. Inspections will cover civilian as well as military establishments. Iraq will be prohibited from importing a comprehensive list of chemicals, other materials, and components which could be used to produce weapons of mass destruction or ballistic missiles. United Nations inspectors have the right to travel anywhere in Iraq, inspect any site, facility, activity, material or other item, and take any documents they consider relevant. Their inspections must not be hindered in any way and may be unannounced. Although there are suspicions that Iraq has hidden some SCUD ballistic missiles in defiance of Resolution 687, and Iraq has not co-operated fully with United Nations inspectors, it seems that the resolution has generally been reasonably successfully implemented. Moreover, the presence of inspectors will stop Iraq producing any of the prohibited weapons.

Iraq's nuclear-weapon programme

78

HOW LONG WOULD IRAQ HAVE TAKEN TO PRODUCE NUCLEAR WEAPONS? Separating fact from fiction about how long it would have taken Iraq to produce nuclear weapons is no easy task, even though the 1991 Gulf War has been over for some time. Getting at the truth is difficult because the Iraqis, on the one hand, and those countries which fought against Iraq, on the other hand, are anxious to manipulate the evidence about Iraq’s nuclear-weapon programme. The Iraqi authorities are, to say the least, evasive about their nuclear activities probably because they want to hide whatever details they can, in case they decide to continue to develop nuclear weapons at some future date. Governments which fought against Iraq, like the American and British ones, generally exaggerate Iraq’s nuclear capabilities for domestic political reasons. These governments want to maintain public support for their military action against Saddam Hussein and for possible future action against him. And they know that what concerns the public most is the thought that Iraq may acquire nuclear weapons. It therefore suits them to pretend that Iraq was very close to producing nuclear weapons. The stories that Iraq may have made a nuclear weapon from the enriched uranium it had acquired as fuel for its nuclear research reactors turned out to be untrue. The Iraqis would have needed enriched uranium containing 20 kilogrammes or so of U-235 to have fabricated a nuclear weapon. They had insufficient unirradiated enriched uranium for this purpose. Moreover, the uranium fuel they did have was under IAEA safeguards and regularly inspected. In fact, one inspection occurred in November 1990, after Iraq invaded Kuwait. The inspectors found that the uranium was still in the form of reactor fuel elements. One reason why there is confusion about the rate of progress of Iraq’s nuclear-weapon programme is the rivalry that has evolved between the IAEA and the UN Special Commission. A symptom of this rivalry is the leaking of information by officials of the Special Commission to the press about the findings of the IAEA inspectors. For example, the inspectors discovered that a West German firm supplied Iraq with 240,000 small ferrite spacers. This information was leaked to the press against the wishes of the IAEA (Albright and Hibbs 1992c). The Iraqis were working on a design for a gas centrifuge to enrich uranium for use in nuclear weapons. The design required twenty-four spacers per machine and the spacers supplied by the Germans could, therefore, have equipped 10,000 centrifuges. Some newspapers incorrectly and misleadingly reported that the spacers would have enabled Iraq to build quickly 10,000 centrifuges and produce enough enriched uranium for three or four nuclear weapons a year. A gas centrifuge to enrich uranium is a sophisticated machine, built from many special components. The Iraqis were unable to produce one effective centrifuge. They had simply acquired the spacers in case they needed them at some future date. A general policy of the Iraqis was to pick up bits and pieces wherever possible if there was any chance that

How nuclear weapons spread

79

they might at some time be useful in their programme. In its inspections of Iraqi facilities since the end of the Gulf War, the IAEA has discovered that Iraq was energetically pursuing two methods of enriching uranium for use as fissile material in nuclear weapons. In addition to gas centrifuges, they were working on devices called calutrons. Of the two uranium-enrichment methods, top priority was given to the calutrons. Apparently, the Iraqis began working on this project in 1982 at the Tuwaitha nuclear research centre. The Iraqis secretly built and operated eight calutrons at Tarmiya, 40 kilometres northwest of Baghdad, between February and September 1990. A second batch of seventeen calutrons was being installed when Tarmiya was bombed in January 1991, although some components had yet to be manufactured. A total of seventy calutrons was planned. These calutrons were capable of producing only low-enriched uranium, which could not have been used in a nuclear weapon. The plan was to further enrich this low-enriched uranium to weapon-grade uranium in more sophisticated calutrons, six of which were being built at Tarmiya when it was bombed. It would have taken some time—perhaps two years or more—to complete the calutron project. The IAEA calculates that if the Iraqis had eventually built and operated all these calutrons they could have produced about 13 kilogrammes of weapon-grade uranium a year, enough to produce a nuclear weapon every two years. But the assumptions made by the Agency were very optimistic. In practice, it is very unlikely that the Iraqi calutrons would have operated as efficiently as the Agency assumed. It is more likely that it would have taken the Iraqi calutrons at least three or four years to produce enough weapon-grade uranium for one nuclear weapon. Once operating experience had been gained, the Iraqis may, of course, have been able to have significantly increased the rate of production of weapon-grade uranium from calutrons. The Iraqis began their centrifuge project in 1987 at Tuwaitha. Two centrifuges were successfully tested at Tuwaitha; both used carbonfibre rotors acquired from abroad. Only one of the two tests actually enriched uranium and the rotor used in this centrifuge broke down during the test. The Iraqis showed UN inspectors ten carbon-fibre rotors, probably made abroad. As has been described, the performance of a gas centrifuge is measured in Separative Work Units. The Iraqis claimed to have achieved an enrichment rate of about 1.9 units a year but thought that they could have improved this to about 2.7 units a year. This is by no means an impressive performance—Pakistani centrifuges, for example, are twice as efficient (Albright and Hibbs 1992a). The Iraqis planned to mass-produce gas centrifuges using maraging-steel rotors, even though these are less efficient than carbon-fibre ones, at a factory near An Walid, 20 kilometres south of Baghdad. The Agency inspectors believe that about 600 centrifuges a year could have been produced using the machine tools available at the plant. But no centrifuge had actually been produced at An Walid. The Iraqis were unable to produce good enough maraging steel for centrifuges. Nevertheless, they planned to have a 500-centrifuge cascade in operation by 1996. Whether or not the Iraqis could have achieved this is very uncertain. About 5,000 Separative Work Units of enrichment are needed to produce enough

Iraq's nuclear-weapon programme

80

weapons-grade uranium for one nuclear weapon. Assuming optimistically that the Iraqis could have eventually made centrifuges capable of producing 5 units of enrichment a year, they would have needed 1,000 centrifuges operating continuously to produce one nuclear weapon a year. If one in three centrifuges were operating at any one time, another optimistic assumption, they would have needed a total of more than 3,000 centrifuges to produce enough weapon-grade uranium for one nuclear weapon a year. Yet they planned to have only 500 operating by 1996. Documents found by IAEA inspectors show that Iraqi scientists were indeed designing a nuclear weapon. Iraqi nuclear scientists were working on a design at a research centre at Al-Atheer, 50 kilometres south of Baghdad. High-quality conventional high explosives for an implosion-type nuclear weapon were, for example, being developed there. The IAEA has destroyed the buildings, equipment, and materials at the Al-Atheer centre thought by the IAEA to be related to nuclear weapons (Albright and Hibbs 1992d). Iraqi scientists were developing polonium-beryllium neutron initiators for nuclear weapons. It seems that some had been tested using high-explosive shaped charges but with what success is not known. The Iraqis were also investigating neutron guns to initiate the fission chain reaction in nuclear weapons. They were also trying to produce the high-quality electronic components for use in circuits to set off the detonators for the shaped charges in an implosion system. They had not succeeded in producing adequate electronic circuits for use in nuclear weapons before the invasion of Kuwait. The Agency inspectors discovered that Iraqi scientists had reprocessed plutonium at Tuwaitha. This is intriguing. Although only about 3 grammes of plutonium were found, the fact that the Iraqis were experimenting with reprocessing may suggest that they intended to build, or were building, a clandestine plutonium-production reactor. The Iraqis had acquired from Italy some hot cells for handling plutonium. The Iraqis had acquired at least 3 tonnes of heavy water (which, they say, was lost during the bombing), and they have lots of uranium. The production of plutonium for nuclear weapons—in, for example, a heavy-water (or graphite) moderated, naturaluranium fuelled reactor—is a lot easier and quicker than producing enriched uranium. A small plutonium-production reactor with a power output of about 25 MW(th), capable of producing about 6 kilogrammes of weapon-grade plutonium a year, enough for one nuclear weapon a year, could be easily hidden, in, say, an underground cavern dug out of a mountain. The Agency inspectors have found no evidence that such a reactor exists, or was under construction, but Iraq is a big country and the inspectors have only seen a fraction of it. Iraq has been a party to the NPT since 1969 and IAEA inspectors have been visiting Iraq about twice a year for many years. The inspectors visited the Tuwaitha centre, at which plutonium was separated and research on the enrichment of uranium using both centrifuges and calutrons was conducted. But the inspectors totally failed to discover any sign of Iraq’s nuclear-weapon programme, inevitably raising questions about the effectiveness of the Treaty. The example of Iraq shows that a country that takes the political decision to do so can illegally establish a nuclear-weapon programme while a party to the NPT, taking advantage of its membership of the Treaty to obtain assistance in acquiring nuclear

How nuclear weapons spread

81

technology from numerous foreign suppliers. Many believe that we need a much stronger international non-proliferation regime to prevent other countries doing what Iraq did. Misinformation about Iraq’s nuclear-weapon programme is wide-spread—a good example of the saying that ‘truth is the first casualty of war’. Some authorities want us to believe that Iraq was within months of making nuclear weapons when it invaded Kuwait in August 1990. But, on available evidence, the Iraqis would have taken at least four or five years to produce enough weapons-grade uranium for one nuclear weapon. They would not have had a militarily-significant nuclear force of, say, twenty nuclear weapons until well into the twenty-first century (Albright et al. 1993).

13 NORTH KOREA’S NUCLEAR PROGRAMME Over the past few years, suspicions that North Korea has ambitions to acquire nuclear weapons have steadily grown even though it ratified the NPT in 1985. North Korea’s announcement on 12 March 1993 that it was withdrawing from the Treaty has reinforced these suspicions. According to a March 1990 statement by Rear Admiral Thomas Brooks, Director of America’s Naval Intelligence, North Korea has an ‘advanced nuclear weapon development’ programme. And in mid-1992 the director of the US CIA publicly announced that North Korea was close to producing nuclear weapons. But there is little evidence that it is currently developing nuclear weapons. On 4 May 1991 North Korea gave the IAEA a list of its nuclear facility, including: one research reactor in operation and two under construction; a sub-critical assembly; a fuelfabrication plant; and two uranium mines and mills. The list excludes a small research reactor which is already under IAEA safeguards. Presumably, this reactor will be the only facility safeguarded by the IAEA while North Korea stays outside the NPT. A plant listed as ‘a radiochemical laboratory for research on the separation of uranium and plutonium and waste management and for the training of technicians’ is assumed to be the reprocessing plant that North Korea was suspected of operating. Dr Hans Blix, the Director General of the IAEA, visited the plant in mid-May, described it as 80 per cent complete and confirmed that, when completed, it would be a reprocessing plant (Christian Democrat International 1992). In Blix’s words, the Agency ‘would not have any hesitation in calling (the plant) a reprocessing plant in the terminology of the industrialized world’.

NORTH KOREA’S NUCLEAR PROGRAMME North Korea has large uranium deposits. The extent of these was discovered in the early 1960s when the North Koreans, assisted by the Chinese, made extensive surveys throughout the country. The size of the uranium reserves has not been publicly revealed. Uranium ore, mineable at economic prices, is probably located in the south of the country near Pyongsan and Hae Kumgang, at Unggi in the far north, and at Hamhung in the centre on the east coast. North Korea is believed to export uranium to the People’s Republic of China and the former USSR and possibly to other countries. It is believed that there is a uranium mill operating at Kusong, producing significant amounts of refined uranium since the early 1980s (Spector 1990). North Korea has had assistance on nuclear research from both the USSR and the

North korea's nuclear programme

83

People’s Republic of China. Agreements with the USSR date back to 1956 and with China to 1959. The former provided for Soviet assistance in the establishment of a nuclear research centre at Yongbyon, about 90 kilometres north of Pyongyang. Serious nuclear research activity began here in the early 1960s. In 1965, North Korea acquired from the USSR a 4-MW(th) pool-type research reactor and a l00-kW(th) critical assembly. The research reactor and the critical assembly were installed at Yongbyon. The reactor probably became operational in 1967. Fuelled with highly-enriched uranium supplied by the Soviet Union, it has been used to produce radioisotopes for use in medicine and industry, and for research. North Korea reportedly doubled the power output of the reactor using its own technology. In 1987 the North Koreans began operating a second research reactor at Yongbyon. The indigenously designed and constructed reactor is listed in the inventory given to the IAEA as having a power output of 5-MW(e) and is described as an experimental power reactor. It is a gas-cooled, graphite-moderated reactor, fuelled with natural uranium. This power output is much less than the 30-MW(th) normally quoted in the literature. Apparently, North Korea is able to produce graphite of sufficient purity to be used in a reactor. It also provides the natural-uranium fuel for the reactor but the location of the plant at which the reactor fuel elements are manufactured is not publicly known, although it is listed in the inventory of nuclear facilities given to the IAEA. The inventory lists two reactors under construction; one has a power output of 50 MW (e) and the other of 200 MW(e). Construction reportedly began in 1984. The 50-MW(e) reactor is under construction at Yongbyon and may be completed in 1995. The larger reactor is under construction at Taechon, 100 kilometres north of Pyongyang (Albright and Hibbs 1992g). Both reactors are gas-cooled, graphite-moderated, natural-uranium fuelled reactors. Like the smaller (5-MW(e)) reactor, they would produce plutonium efficiently, including weapon-grade plutonium. It seems, however, that electric power grids are being constructed at the sites at which the reactors are being built which suggests that they will be used to generate electricity. Until North Korea produced its inventory for the IAEA, it was believed that North Korea was constructing only one new reactor. This construction activity and the construction of a reprocessing plant was observed by US military surveillance satellites and considerably enhanced concern about North Korea’s nuclear-weapon ambitions. It was generally assumed that a new reactor would become operational in 1992. The construction of the reprocessing plant, at Yongbyon, is believed to have begun in about 1988; it is generally assumed that it could be separating plutonium in the mid-1990s. Although not yet completed, the plant has separated gramme quantities of plutonium from damaged reactor fuel elements. When fully operational, the plant could reprocess several hundred tonnes of spent reactor fuel elements a year, a large enough capacity to handle all the spent fuel from the three North Korean reactors. How much plutonium could North Korea have accumulated? The 5-MW(e) reactor should be able to produce about 4 kilogrammes of weapon-grade plutonium a year. Assuming that it began operating in 1989 it will by now (1992) have produced about 12 kilogrammes of weapon-grade plutonium, enough for two nuclear weapons. The 50-MW

How nuclear weapons spread

84

(e) reactor should be able to produce about 40 kilogrammes of weapon-grade plutonium a year and the 200-MW(e) reactor about 160 kilogrammes of weapon-grade plutonium a year. When all three reactors are operating, North Korea should have the capacity to produce about 200 kilogrammes of weapon-grade plutonium a year, enough to produce about 30 nuclear weapons a year. But North Korea is unable to separate plutonium in significant quantities from spent reactor fuel elements until its reprocessing plant begins operating, possibly in 1995–7. By that time the plant should be under IAEA safeguards. Albright et al. (1992) estimate that North Korea could separate about 200 kilogrammes of plutonium a year, if its reprocessing plant becomes operational. North Korea has plans to construct nuclear-power reactors. At the end of December 1986, North Korea established the Ministry of Atomic Power Industry. The ratification of the NPT by North Korea led to an agreement with the Soviet Union to construct four 440MW(e) power reactors at Sinpo, on the coast of the Sea of Japan in the east of the country. The site for the reactor has been selected and approved. Now that North Korea has completed a full-scope safeguards agreement with the IAEA there is presumably no political reason why the agreement should not go ahead. But whether the Russians will provide the power reactors remains to be seen.

INTERNATIONAL SAFEGUARDS On 31 December 1991, North Korea and South Korea signed a joint agreement under which they will not separate plutonium in a reprocessing plant and would not acquire uranium-enrichment plants. The agreement also provided for bilateral inspections of each other’s nuclear facilities. The future of this agreement is in doubt, however, because it may be regarded as being made redundant by withdrawal of North Korea from the NPT. South Korea, which ratified the NPT in 1975, has been under full-scope safeguards for many years. Although the Korean agreement bans reprocessing, North Korean officials, including the representative at the IAEA, have recently said that North Korea must complete the construction of its reprocessing plant, which it calls a ‘radiochemistry laboratory’, for its contribution to the economic and technological development of the country. But there is another reason why North Korea may need access to a reprocessing plant. The North Korean reactors are of a similar type to, for example, British Magnox reactors—gas-cooled, graphite-moderated reactors fuelled with natural uranium metal. The spent reactor fuel elements from this type of reactor cannot be stored, easily and safely, for long periods, and should therefore be reprocessed. The reason for this is that the material—normally, magnesium oxide mixed with zirconium, called ‘Magnox’—in which the natural uranium fuel is encased, or ‘clad’, corrodes in moist air or water. Storing Magnox fuel in water for long periods is, therefore, not possible because the corrosion will eat away the cladding and release the radioactivity from the spent fuel elements into the environment. Storing large amounts of spent Magnox fuel in air is also not practicable because there

North korea's nuclear programme

85

is bound to be some moisture which will eventually corrode the cladding and expose the metal core to air. When exposed to air, uranium metal may spontaneously ignite (rather like sodium) and burning will release some of the highly radioactive fission products in the spent fuel elements into the atmosphere. Because Magnox-type reactors produce weapon-grade plutonium very efficiently, they are normally used by the military. Civilian nuclear-power reactors—such as light-water reactors—are normally fuelled with uranium oxide encased in a zirconium alloy which does not corrode significantly. In contrast to Magnox fuel, oxide fuel can be stored for very long periods in water, until it is permanently disposed of in geological repositories. North Korean spent reactor fuel elements are likely to be reprocessed eventually either in the North Korean reprocessing plant, when it is completed, or sent abroad for reprocessing. Whether or not a foreign country would be willing to reprocess North Korean spent reactor fuel is doubtful. Reprocessing in North Korea would, of course, violate the bilateral agreement with South Korea, even if it were done under IAEA safeguards. How this dilemma will be resolved remains to be seen. North Korean officials apparently told Dr Hans Blix, during his visit to North Korea in May 1992, that they wanted to produce plutonium for later use in breeder reactors or in MOX fuel for light-water reactors. But they also said that North Korea would be willing to give up its reprocessing plant and its gas-cooled graphite reactors if it were provided with light-water reactors without cost. Even though Asian security would be considerably enhanced if there were no suspicions that North Korea (or South Korea) had ambitions to acquire nuclear weapons, it is very unlikely that any country will be prepared to give North Korea reactors. North Korean interest in reprocessing and in the use of plutonium has rekindled South Korean interest in reprocessing. South Korean interest, like North Korean interest, in reprocessing is also related to Japan’s intention to reprocess large amounts of spent reactor fuel and use the plutonium as reactor fuel. If one country in East Asia stockpiles plutonium, others get nervous, because of the possible use of plutonium to fabricate nuclear weapons, and want to acquire plutonium themselves. South Korea first tried to acquire reprocessing technology from the west a long time ago. In the early 1970s, for example, South Korea negotiated with France for the purchase of a small reprocessing plant. But under considerable US pressure the deal was cancelled in 1975. (In 1976, the US pressurized Taiwan to close down a small reprocessing plant which it had been secretly operating). More recently, in September 1991, South Korea tried to persuade the Americans to sell it reprocessing technology of the type developed for the Integral Fast Reactor. The request was refused. The efforts made by South Korea to acquire reprocessing technology and North Korea’s refusal to stop constructing its reprocessing plant cast doubt on the seriousness of the bilateral pledge by both countries that they would not reprocess plutonium. On 20 July 1977, North Korea signed a non-NPT agreement with the IAEA putting the 4-MW(th) pool-type research reactor and the 100-kW(th) critical assembly, supplied by the Soviet Union, under safeguards. Under the 1992 full-scope safeguards agreement, signed under its NPT commitments, North Korea accepted IAEA safeguards on all its

How nuclear weapons spread

86

nuclear facilities. This has, however, not prevented some observers believing that North Korea has already separated a significant amount of plutonium which is now hidden for a secret nuclear-weapon programme. The USSR provided North Korea with a number of hot cells which were used to separate some plutonium produced clandestinely in the 35– MW(e) reactor. Some also suspect that more plutonium has been produced in a pilot reprocessing plant which was not included in the inventory given to the IAEA. The announcement by North Korea that it was withdrawing from the NPT is not only a serious blow to international efforts to prevent the proliferation of nuclear weapons but it will also have serious repercussions throughout the Asia-Pacific region. While post-Cold War military budgets are decreasing in most Third World and industrialized countries, China, Malaysia, Taiwan, Thailand, Singapore and South Korea are all increasing their military budgets by more than 5 per cent in real terms. Japan is increasing its military budget by about 3 per cent per year. Any move which could lead to nuclear proliferation is thus bound to be alarming. The IAEA conducted six inspection trips in North Korea under NPT safeguards, but then ran into trouble by demanding special inspections of two suspected nuclear-waste sites which The Pyongyang government insisted were non-nuclear military facilities and deployed tanks and missiles to defend them. The Agency says that it has evidence that there is radioactive waste at the sites. By analysing the radioisotopes in the waste the Agency could find out if North Korea has produced more plutonium than the few grammes it has admitted to. According to the North Koreans, they separated plutonium only once, in 1990, but a sample already analysed by the Agency indicated that North Korea reprocessed spent fuel from the 5-MW(e) reactor three times—in 1989, 1990 and 1991. The North Koreans tried to explain the discrepancy by admitting that in 1975 they reprocessed a small amount of plutonium produced in the small research reactor supplied by the ex-Soviet Union in 1965. The Agency’s special inspections would have resolved the issue. The Pyongyang government was given a limited time to reconsider its position or face repercussions, including possible action by the United Nations Security Council. In reply to this ultimatum North Korea withdrew from the Treaty. It is likely that North and South Korea will unite in the next ten years. If, in the meantime, North Korea develops nuclear weapons, the united country could rapidly become a major nuclear-weapon power. This would put considerable pressure on other countries in the Asia-Pacific region and the situation could escalate into a full scale nuclear-arms race. As long as North Korea stayed in the non-proliferation regime, the nuclear-weapon programme that it is thought to have started, with Chinese assistance, lay dormant. Now that North Korea is isolated outside the regime, this programme may well be restarted. Perhaps the Agency should have let sleeping dogs lie.

14 ARGENTINA’S NUCLEAR PROGRAMME For the past forty years, the main goal of Argentine nuclear policy has been selfsufficiency in nuclear technology by acquiring a complete nuclear fuel cycle. In consequence, Argentina now has research reactors, power reactors, uranium mines and mills, uranium dioxide purification and uranium hexafluoride production plants, a uranium-enrichment facility, a reactor-fuel fabrication plant and a heavy-water production plant. A pilot reprocessing plant has been constructed but the project was suspended in 1990. Argentina’s nuclear programme is the responsibility of the National Atomic Energy Commission (CNEA). For many years, the top officials of the CNEA were military people—usually senior naval officers. The Commission essentially operated two ‘parallel’ programmes. One concentrated on the development of nuclear power for civilian use. The other was involved in acquiring a capability to produce fissile material suitable for the fabrication of nuclear weapons. However, the replacement in 1983 of the military government in Argentina with a civilian government led to the dismantlement of the military arm of the programme, which is now entirely peaceful. The change was formalized at a meeting in Foz do Iguacu, Brazil, on 28 November 1990. At the meeting, President Carlos Saul Menem of Argentina and President Fernando Collor of Brazil signed the Declaration on the Common Nuclear Policy of Brazil and Argentina which confirms the intention of both countries to use nuclear energy only for peaceful purposes. The Presidents have since carried out reciprocal visits of nuclear facilities. Nevertheless, both Argentina and Brazil are developing nuclear submarines and might even co-operate in this development.

ARGENTINA’S NUCLEAR FUEL CYCLE Argentina has significant uranium resources of about 44,000 tonnes; about 12,000 tonnes are mineable at economical prices, given the cur-rent cost of uranium on the metal market. Uranium mines are operating at La Estella and Sierra Pintada. Uranium mills are operating at La Estella and San Rafael. CNEA plans to start a mill at Sierra Pintada but is having difficulty financing it and has delayed the operation. Argentina is operating a UO2 purification plant at Córdoba, with a capacity of about 150 tonnes a year. The plant, supplied by West Germany, began full-scale operations in 1986. The status of a second plant, with a similar capacity and supplied and built by the Argentinians at Córdoba, is uncertain; it may not yet have been brought into operation.

Argentina's nuclear programme

88

A fuel fabrication plant has been in operation at Ezeiza since 1982. Its capacity is about 300 tonnes a year. A pilot plant for the production of enriched uranium fuel for research reactors at Constituyentes has been operating since 1976. Argentina produces heavy water for its nuclear-power reactors. A pilot plant with a capacity of about 2 tonnes a year has been operating since 1984 at Zarate (Atucha). Supplied from indigenous sources, the current status of the plant is uncertain. A production plant, built at Arroyito and supplied by Sulzer Brothers of Switzerland, has a capacity of about 250 tonnes a year. Producing heavy water by ammonia-hydrogen exchange, the plant came into operation in 1991. Argentina operates two nuclear-power reactors. Both are pressurized heavy-water reactors using natural uranium fuel. Atucha-1, supplied by the German firm Siemens, has been operating at Lima, near Buenos Aires, since 1974, generating 320 MWe. The fuel for the reactor comes from the Ezeiza plant. The heavy water was supplied by West Germany and the USA. The other power reactor has been operating at Embalse, Rio Tercero, Córdoba since 1984. Supplied by Atomic Energy of Canada, it generates 600 MWe. The fuel comes from the Ezeiza plant and the heavy water was supplied by Canada. A third reactor, Atucha-2, is under construction at Lima. Supplied by Kraftwerk Union of Germany, it will be a 700-MWe pressurized heavy-water reactor, fuelled with natural uranium, and is scheduled to come into operation in 1996. The fuel will be supplied by Argentina and the heavy water by Germany and Argentina. The construction of Atucha-2 has been delayed because of a shortage of money. The cost of the reactor has soared and it is estimated that the final cost will be more than $5,000 per installed kilowatt, which will make it one of the world’s most expensive reactors. Argentina has six research reactors. The oldest (called RA-1) came into operation in 1958 at a CNEA establishment at Constituyentes. Designed and built by CNEA, it is a tank reactor, fuelled with 10 kilogrammes of uranium enriched to a concentration of 20 per cent U-235, and has a power output of 70 kW(th). The second oldest (RA-O) came into operation in 1965 at the University of Córdoba. Designed and built by CNEA, it is a tank reactor, a critical assembly with virtually no power output. It is fuelled with about 15 kilogrammes of uranium enriched to a concentration of 20 per cent U-235. The next reactor (RA-2) came into operation in 1966 at Constituyentes. Designed and built by CNEA, it is a tank reactor, fuelled with 10 kilogrammes of uranium enriched to a concentration of 20 per cent U-235. It is a critical assembly with virtually no power output. The fourth (RA-3) came into operation in 1967 at Ezeiza. Designed and built by CNEA, it is a pool reactor, used for the production of radioactive isotopes for medical and industrial purposes. Fuelled with about 5 kilogrammes of uranium enriched to a concentration of 90 per cent U-235, it has a power output of 3.5 MW(th). The fifth research reactor (RA-4) came into operation in 1971 at the University of Rosario, Santa Fe. Supplied by Siemens, Germany, it is a homogeneous reactor fuelled with uranium oxide (U3O8), enriched to a concentration of 20 per cent U-235, homogeneously mixed with polythene, and has virtually no power output.

How nuclear weapons spread

89

The sixth research reactor (RA-6) came into operation in 1982 at the Bariloche Atomic Centre, San Carlos de Bariloche. Designed and built by CNEA, it is a pool reactor, used for materials testing. It is fuelled with uranium enriched to a concentration of 20 per cent U-235, and has a power output of 500 kW(th). Argentina was reportedly developing a heavy-water research reactor (RA-7), to be fuelled with natural uranium, with a power output of about 70 MW(th). The reactor could have produced about 15 kilogrammes of plutonium a year, enough to produce about three nuclear weapons a year. The project was apparently stopped in 1985. Argentina accumulated much knowledge about reprocessing plutonium by operating a small reprocessing facility, designed and supplied by CNEA, at Ezeiza between 1969 and 1973. The pilot plant, which has been dismantled, could reprocess up to 200 kilogrammes of spent fuel a year. Reportedly, a total of only a kilogramme or so of plutonium was separated during the facility’s lifetime. Apparently, only fuel from the RA-1 research reactor was reprocessed. A larger reprocessing facility, also designed by CNEA, is under construction at Ezeiza. It will be able to reprocess up to 5 tonnes of spent fuel a year, producing about 15 kilogrammes of plutonium a year. The plant has been delayed several times; when it will begin operating is not known. Nevertheless, the site of the reprocessing plant was visited in November 1988 by Brazilian President José Sarney as part of the programme of reciprocal Presidential visits to respective nuclear facilities. On 27 June 1990, President Menem reportedly dedicated a small hot cell at Ezeiza, capable of separating plutonium from spent reactor fuel. But the facility cannot separate militarily significant quantities of plutonium. CNEA has been experimenting with gaseous diffusion for the enrichment of uranium since 1983. It is surprising that Argentina chose to use gaseous diffusion for uranium enrichment rather than the gas centrifuge method. There is little doubt that Argentinian scientists and engineers could master centrifuge technology. By the early 1980s, gaseous diffusion was an obsolete technology. It takes far more electric power to produce a given amount of enriched uranium with gaseous diffusion than with gas centrifuges. Nevertheless, the technology was first tested in a pilot facility at Pilcaniyeu. And in 1988, a production plant, built by CNEA, began operating at Pilcaniyeu, producing uranium enriched to a concentration of 20 per cent U-235. The capacity of the plant was originally about 500 kilogrammes a year of uranium enriched to this concentration (equivalent to 23,000 SWUs a year) but the capacity is now somewhat less. CNEA is producing uranium hexafluoride at a facility at Pilcaniyeu. Its capacity is presumably sufficient to provide the feed material for the enrichment plant. It is reported, however, that the initial feedstock of hexafluoride for Pilcaniyeu was supplied by China. The enriched uranium produced by the Pilcaniyeu plant provides some of the fuel for Argentina’s research reactors and some of the fuel used by the Atucha-1 power reactor.

INTERNATIONAL SAFEGUARDS Argentina is not a party to the NPT. It has signed but has not yet ratified the 1968 Treaty

Argentina's nuclear programme

90

for the prohibition of nuclear weapons in Latin America (Treaty of Tlatelolco). There are, therefore, no safeguards agreements with the IAEA in connection with these treaties. Now that both Argentina and Brazil have moved away from the production of nuclear weapons and because of improved relations with Brazil, it seems that Argentina may ratify the Treaty of Tlatelolco (Simpson 1992). If it does so, it will have to put all its nuclear facilities under safeguards and renounce nuclear weapons. In 1991, the Argentine-Brazilian Agency for Accounting and Control (ABAAC) was set up by Argentina and Brazil to inspect their nuclear facilities (Christian Democrat International 1992). Data from the inspections will be given to the IAEA under a safeguards agreement negotiated between the ABA AC and the IAEA. Argentina’s ratification of the Treaty will probably depend on an agreement that ABAAC will control the inspection process. The control body of the Treaty is, however, the Agency for the Prohibition of Nuclear Weapons in Latin America (OPANAL). The Treaty allows OPANAL to inspect nuclear sites in a Party at the request of other Parties. Argentina claims that OPANAL may not keep the results of its inspections secret and wants ABAAC to do the inspections. Argentina, like Brazil, has decided to move towards accepting international safeguards. Each country will open up some nuclear facilities for inspection by the other. And Argentina, like Brazil, has signed a full-scope safeguards agreement with the IAEA. Argentina is now committed to allowing all its nuclear facilities to be inspected by Brazil and the IAEA (Redick 1990). Whether or not the IAEA will regard ABAAC inspections to be strict enough to satisfy the IAEA that ABAAC can adequately safeguard the Treaty of Tlatelolco remains to be seen. But the signs are that it will and that Argentina will ratify the Treaty, presumably provided that it is satisfied that Brazil will also do so. Argentina already had non-NPT safeguards agreements with the IAEA covering the RA-1, RA-2, RA-3, RA-4, and RA-6 research reactors; the Atucha-1 and Embalse power reactors; the large UO2 purification plant at Córdoba; the heavy-water plant at Arroyito; and the fuel fabrication plants at Constituyentes, when safeguarded uranium is processed, and Ezeiza, when safeguarded uranium is processed. On 27 April 1990, President Menem signed a decree under which Argentina will export ‘materials, equipment, technology, technical assistance and/or services related to the processing and enrichment of uranium, reprocessing of nuclear fuel, production of heavy water and manufacture of plutonium’ only if these items are under IAEA safeguards. Argentina has nuclear ties with a number of countries. It is suspected of having given nuclear assistance to Iraq; has sold research reactors and the enriched uranium to fuel them to Algeria and Peru; and has bought nuclear materials from China without IAEA safeguards. Argentina has concluded an agreement with Turkey to build 25-MWe CAREM-15 nuclear-power reactors in both countries. A light-water reactor fuelled with uranium enriched to 20 per cent U-235, the CAREM-15 was developed in Argentina. The construction will be financed mainly by Turkey. Argentina hopes to export the reactor to Algeria, Peru, Egypt and other countries in the Middle East and Africa. Argentina is also developing a nuclear submarine. A prototype reactor is being

How nuclear weapons spread

91

developed at São Paulo in Brazil. The 50-MW(th) reactor will be fuelled with enriched uranium. The first nuclear submarine is expected to displace 2,800 tonnes and have a speed of 25 knots.

15 BRAZIL’S NUCLEAR PROGRAMME Brazil, like Argentina, wanted a self-sufficient nuclear programme. To this end, Brazil signed a far-reaching agreement, on 27 June 1975, with West Germany under which it would supply Brazil with a nuclear fuel cycle. Brazil was to receive eight pressurizedwater reactors over a fifteen-year period and a wide range of nuclear technologies, including uranium prospecting and treatment, uranium enrichment, and plutonium reprocessing. Joint Brazilian-West German companies would handle the business and Brazilians would be trained in nuclear technologies in West Germany. Under the deal, Brazil would have acquired an entire nuclear-fuel cycle and total nuclear independence, including the capability to fabricate a significant nuclear-weapon force. Although never officially admitted, a wish to acquire a capability to produce nuclear weapons was a powerful factor in Brazil’s early nuclear ambitions. Given that Brazil had a military government at the time, this is hardly surprising. In March 1985, a civilian government came into power when José Sarney became President. But the military’s nuclear-weapon programme was not stopped until Fernando Collor de Mello was elected President in December 1989. In October 1990, it was announced that the government had discovered and stopped a secret nuclear-weapon programme code-named ‘Solimoes’. The aim of the military was to develop a secret capability to produce fissile materials suitable for the fabrication of nuclear weapons. The National Atomic Energy Commission (CNEN) and the Brazilian Army, Air Force and Navy were all involved in the research, none of which was accountable to any civilian authority. CNEN’s secret research was mainly carried out at the IPEN Institute, near the São Paulo University. A programme to develop nuclear reactors for submarines was conducted here. Another CNEN-Navy project at IPEN was the development of gas centrifuges for the enrichment of uranium. Research was also done on the use of lasers for uranium enrichment. It appears that preparations were being made for an underground nuclear test at a remote air-base in the central Amazon. The government said that a nuclear weapon would probably have been designed in the Centro Tecnico Aerospacial, an Air Force establishment at São José dos Campos. After President Collor’s clamp-down, the Brazilian nuclear programme was restricted to civilian applications (Christian Democrat International 1992). Ten years ago, the Brazilians had very ambitious plans for civilian nuclear power. The President of Brazil’s electricity company forecast that by the year 2000 Brazil would have more than sixty nuclear plants, generating a total of 81,000 MWe. But Brazil will be lucky to have two nuclear-power reactors in operation by the year 2000, generating less than 2,000 MWe. Severe economic problems forced the Brazilian government to cut the

Brazil's nuclear programme

93

nuclear budget drastically. Brazil has large resources of uranium—estimated at 226,000 tonnes with 193,000 tonnes reasonably assured reserves. A uranium mine is operating at Pocos de Caldas. There is also a uranium mill operating at Pocos de Caldas, with a capacity of 450 tonnes a year. Currently, the Brazilians have one nuclear-power reactor operating at Itaorna, Rio de Janeiro, generating 626 MWe. This reactor was purchased in 1972 from the American firm Westinghouse, well before the 1975 deal with West Germany. A pressurized-water reactor, fuelled with low-enriched (about 3 per cent U-235) uranium, it came into commercial operation in 1982. Initially, the fuel for this reactor was supplied by the USA. Subsequently, fuel was supplied by URENCO, the German-British-Dutch enrichment company. More recently, Brazilian enriched uranium has been used. A second reactor, supplied by Kraftwerk Union of Germany, is under construction. It is a pressurized-water reactor, fuelled with low-enriched uranium, being built at Itaorna, Rio de Janeiro. It will generate 1,230 MWe. The current plan is that it will come into operation in 1997. But the construction of the reactor has already been delayed several times and more delays are likely. The first load of fuel for the reactor will come from URENCO. Subsequent loads may use some Brazilian enriched uranium. The construction of a third reactor has been started at Itaorna. Although about 40 per cent complete, no further work is being conducted on it and the government is considering cancelling the project. Germany has supplied Brazil with uranium-enrichment technology. An enrichment plant is in operation at Resende, using Becker jet-nozzle technology developed in Germany. The technology was tested in Brazil on a laboratory scale at Belo Horizonte, beginning in 1980. The technology has proved less cost-effective than gas-centrifuge technology, using significantly more energy for a given amount of enrichment. The Resende plant, supplied by Kraftwerk Union and using jet-nozzle technology, is designed to be expanded in several stages. The first stage, with a capacity of 6,000 SWUs a year, will produce very low enriched uranium, with a concentration of about 0.85 per cent U-235. The future of the plant is uncertain. This level of enrichment is too low for Brazil’s nuclear-power and research reactors. Several more cascades would have to be added to produce the degree of enrichment needed. It is unlikely, to say the least, that the plant will be extended to an adequate scale before the next century, if ever. A domestically built uranium-enrichment establishment is in operation at the Aramar Research Centre, Sorocaba in the state of São Paulo. This indigenous plant uses gas centrifuges to enrich uranium. As a very industrialized country, Brazil has proved capable of mastering centrifuge technology, using mainly its own scientists and engineers. The precise capacity of the Aramar plant is not publicly known but about 400 kilogrammes of uranium enriched to a concentration of 5 per cent in U-235 were reportedly produced in 1989. Small quantities have been enriched to 20 per cent at Aramar. This level of enrichment is suitable for use as fuel in reactors to propel nuclear submarines. According to David Albright (1989), there were about 300 centrifuges, made from maraging steel, operating at Aramar in 1988. Presumably, the plan is eventually to instal

How nuclear weapons spread

94

the large number of centrifuges needed to provide sufficient low-enriched uranium to fuel Brazil’s two nuclear-power reactors. Whether or not the government will provide the money for this remains to be seen. It is even more uncertain whether the government will provide enough money to expand the facility enough to produce fuel for naval reactors as well as civilian ones. Uranium hexafluoride for the uranium-enrichment facility is produced at a plant at IPEN, São Paulo, built with indigenous resources. The plant reportedly has a capacity of about 90 tonnes of UF6 a year and came into operation in 1984. Experience in the production of UF6 was obtained at a laboratory-scale facility supplied and built by the Brazilians. It came into operation in 1982, with a capacity of about 15 tonnes a year (Spector 1990). A larger plant has been built at Resende. The planned capacity is 500 tonnes a year to begin with, to be expanded eventually to 2,000 tonnes a year. It has been supplied by the French firms Péchiney Ugine Kuhlman and Comurhe. The start of production has been postponed, possibly indefinitely. Presumably, the future of the plant will depend on the decisions made about the amount of enriched uranium needed. The fuel elements for Brazil’s nuclear-power reactor are assembled at a facility at Resende, supplied by Kraftwerk Union. Components supplied by the German firm Reaktor-Brennelement Union are assembled. The planned capacity of the plant is about 100 tonnes a year of processed uranium. But with only one power reactor operating, only about 20 tonnes a year are needed today. When the second reactor starts up a total of about 60 tonnes a year will be needed. The 1975 Brazilian-West German agreement included the supply of a plutonium reprocessing plant capable of separating about 25 kilogrammes of plutonium a year. But the order was subsequently cancelled for financial reasons. The Brazilians have done some research into reprocessing and appear to have built a laboratory-scale facility at IPEN. It is doubtful if any reprocessing was done, however, because all the spent reactor fuel in Brazil is under safeguards. Brazil operates four research reactors. The oldest, the IEAR-1, went into operation in 1957 at São Paulo. Supplied by the American firm Babcock & Wilcox, it is a pool reactor with a power output of 5 MW(th). It is fuelled by 4 kilogrammes of uranium enriched to a concentration of 93 per cent U-235. The Triga-UMG reactor came into operation in 1960 at Belo Horizonte. It is a tank reactor, fuelled with 13 kilogrammes of uranium enriched to 20 per cent U-235, supplied by the American firm General Electric, with a power output of 100 kW(th). The RIEN-1 reactor came into operation in 1965 at Rio de Janeiro. A tank reactor with a power output of 10 kW(th), it is fuelled with 11 kilogrammes of uranium enriched to 20 per cent U-235, and was domestically supplied. The fourth research reactor is the IPEN light-water reactor fuelled with uranium enriched to 4 per cent U-235. It came into operation in 1988 with a power output of only 100 watts(th).

Brazil's nuclear programme

95

INTERNATIONAL SAFEGUARDS The nuclear facilities in Brazil under IAEA safeguards or containing safeguarded materials are the Angra-1 nuclear-power reactor; the IAER-1, RIEN-1, Triga-UMG, and IPEN research reactors; the fuel-fabrication plant at Resende; the uranium hexafluoride conversion plant at Resende; the uranium-enrichment facility at Belo Horizonte; and the uranium-enrichment plant at Resende. Brazil and Argentina have mutually decided to move towards accepting international safeguards on all their nuclear facilities—full-scope safeguards. They have each opened up some nuclear facilities for inspection by the other and established a bilateral verification agency, ABAAC (Simpson 1992). Brazil, with Argentina, has signed a fullscope safeguards agreement with the IAEA. This agreement effectively opens all nuclear facilities in Brazil to inspection by Argentina and to international inspection.

NUCLEAR SUBMARINES The indigenous facility for enriching uranium at Aramar is related to Brazil’s effort to develop a nuclear submarine (Redick 1990). The reactor to propel the submarine would probably be fuelled with uranium enriched to 20 per cent in U-235. Admiral Mario Cesar Flores, appointed Secretary of War by then President Collor, has supported Brazil’s development of nuclear submarines. The President himself stated that, although the development of nuclear weapons by Brazil was not desirable, the development of nuclear submarines should be the focus of the military nuclear programme. The Brazilian Navy’s research on nuclear propulsion for submarines is undertaken at the Submarine Research Establishment at Rio de Janeiro. According to a statement made by the Brazilian Navy Minister in 1986, the first nuclear submarine is expected to be launched in 1995. Brazil’s nuclear-submarine programme is not a violation of its safeguards commitments. The normal safeguards agreement between the IAEA and a non-nuclearweapon state allows the state to withdraw nuclear material from safeguards if it is used for a ‘non-proscribed military activity’ such as its use as fuel for the reactor of a nuclear submarine.

16 SOUTH AFRICA’S NUCLEAR PROGRAMME Until 1990 many observers believed that South Africa was developing nuclear weapons; some even believed that South Africa had already manufactured some. There is no doubt that South African nuclear scientists and engineers are sufficiently competent to design and produce fission (first generation) nuclear weapons. They would be sufficiently confident that their weapons would work not to insist on the need to test them. In 1990, Foreign Minister Pik Botha announced that his government ‘is now prepared to accede to the NPT in the context of an equal commitment by other states in the South African region’. South Africa acceded to the NPT on 9 July 1991. By doing so, South Africa had stepped back from the nuclear brink. In 1991, Mozambique, Tanzania and Zambia, the so-called front-line states, also acceded to the NPT, raising the prospect of the creation of a nuclear-weapon-free zone in Southern Africa. Botha also said that South Africa had the capability to fabricate nuclear weapons but did not need to do so because ‘the threat of a conventional military conflict in the South African region involving superpower rivalry had diminished substantially’. According to Botha, South Africa had never tested a nuclear weapon but he refused to say whether or not it had produced nuclear weapons. However, on 24 March 1993 President de Klerk announced that South Africa had produced six nuclear weapons but had dismantled them. The IAEA has inspected facilities in South Africa to confirm that the weapons have been dismantled.

SOUTH AFRICA’S NUCLEAR FUEL CYCLE South Africa has a large amount of uranium, with reasonably assured reserves of about 400,000 tonnes. The production of uranium dates back to the late 1940s, when it mined uranium for export to Britain and the United States. These countries used much of the uranium in their nuclear-weapon programmes. Uranium production has, however, considerably decreased over the years because of the decrease in demand. In 1981, for example, South Africa had fourteen uranium mines operating; in 1991, there were only five. Over this period, South Africa’s share of the world uranium market fell from about 14 per cent to about 8 per cent. The main South African uranium production centre is at Witwatersrand Basin at Palabora. Uranium is mined as a by-product of gold and copper production (Uranium Institute 1991). In 1965, a research reactor, SAFARI-I, became operational at Pelindaba, near the border with Botswana. The reactor, a light-water tank-type reactor fuelled with highly-

South africa's nuclear programme

97

enriched uranium, has a power output of 20 MW(th). It was supplied by the USA. Until 1981, the fuel for the reactor was also supplied by the USA; after 1981, South Africa supplied the fuel itself. The original fuel was about 4.5 kilogrammes of uranium enriched to 90 per cent in U-235. After the USA refused to supply more fuel the power of the reactor was reduced to 5 MW(th) and the enrichment reduced to 45 per cent in U-235. South Africa subsequently designed and constructed its own research reactor— SAFARI-II—cooled and moderated with heavy water and fuelled with low-enriched uranium. SAFARI-II produced zero power and was decommissioned in the early 1980s. The fuel and heavy water were supplied by the USA. South Africa is operating two nuclear-power reactors, Koeberg 1 and Koeberg 2, at Melkbosstrand in the Cape. Supplied by the French firm Framatome, they are both pressurized-water reactors, each generating 920 MW(e). Koeberg 1 started up in 1984 and Koeberg 2 in 1985. Until 1988, the fuel for the reactors (uranium enriched to about 3 per cent in U-235) was initially obtained from a number of countries (reportedly including Belgium, China, France, Switzerland and West Germany); after 1988, South Africa supplied the fuel itself. South Africa has a long history of uranium enrichment. In 1970, then Prime Minister John Vorster announced that South African scientists had developed the jet-nozzle technique for enriching uranium, although they had help from West German experts in the technique (Christian Democrat International 1992). Using the jet-nozzle system, a small pilot plant for uranium enrichment began operating in 1978 with a capacity of about 50 kilogrammes of highly-enriched uranium a year. The plant, indigenously constructed at Valindaba near the border with Botswana, was shut down in 1990. Fuel for the SAFARI-I research reactor comes from this plant. In 1988, a larger enrichment plant began operating at Valindaba, also based on the jetnozzle system. The indigenously constructed plant has a capacity of about 50 metric tonnes of low-enriched uranium a year. The uranium hexafluoride needed to feed the plant is produced, at the rate of about 700 tonnes a year, at Valindaba. The hexafluoride plant began operating in 1986. Fuel elements for the SAFARI-I and Koeberg reactors are manufactured at the Elprod plant at Pelindaba. South Africa has also built a hot-cell complex at Pelindaba which may be used for the separation of small amounts of plutonium from spent reactor fuel elements from the SAFARI-I and Koeberg reactors. If South Africa used fissile material for nuclear weapons it would have been highlyenriched uranium produced by the pilot plant at Valindaba. If the plant worked at full capacity during the 1980s, it could have produced about 500 kilogrammes of weapongrade uranium, enough for about twenty-five nuclear weapons (Albright et al. 1992). It is not likely that the plant produced this amount of enriched uranium. A more reasonable estimate is about 300 kilogrammes, enough for about fifteen nuclear weapons.

INTERNATIONAL SAFEGUARDS The SAFARI-I research reactor was put under safeguards in July 1967 in an agreement

How nuclear weapons spread

98

between South Africa and the USA. The two nuclear-power reactors at Koeberg were put under safeguards in January 1977 in an agreement between South Africa and France. Safeguards on these facilities were later transferred to the IAEA. Now that South Africa has joined the NPT all its nuclear facilities are under IAEA safeguards. Previously unsafeguarded facilities now under IAEA safeguards are the uranium-hexafluoride production plant at Valindaba, the uranium-enrichment plant at Valindaba, the fuel-fabrication plant at Pelindaba, and the hot-cell complex at Pelindaba, where safeguarded spent reactor fuel elements are processed. In January 1984, South Africa announced that it will not: sell uranium to non-nuclearweapon countries; make available sensitive technology to any other country; and sell enriched uranium or nuclear equipment without IAEA or EURATOM safeguards.

17 IRAN’S NUCLEAR PROGRAMME Under the regime of Shah Mohammed Reza Pahlavi, Iran had a large nuclear programme. Soon after India’s nuclear test in May 1974, the Shah stated that Iran should have the option of producing nuclear weapons. In 1974, contracts were signed with France, the USA, and West Germany for nuclear-power reactors. France was to supply five reactors, West Germany two and the USA six. Also in 1974, arrangements were made to guarantee supplies of the low-enriched uranium needed to fuel the nuclear power reactors. Iran bought a 10 per cent share in Eurodif, a company set up in 1973 for the construction of a large uranium-enrichment plant, using gas diffusion, in Tricastin. The capital for Eurodif came from Belgian, French, Italian and Spanish sources (Christian Democrat International 1992). The Iranians contracted Eurodif to supply about 270 tonnes of uranium enriched to 3 per cent in U-235. This would have been suitable for use as a power-reactor fuel. Iran also negotiated for a financial stake in another large enrichment plant, with the same consortium. In addition, long-term contracts for the supply of uranium fuel were signed with France, West Germany and the USA. Bilateral nuclear-cooperation agreements were signed with Belgium, Canada, France, Italy, the USA and West Germany. The Shah sent thousands of people to these countries to be trained in a variety of nuclear specialities. In 1976, the Shah negotiated with a number of West German companies for the acquisition of uranium-enrichment and plutonium-reprocessing technologies. Iran also acquired, from the USA in the 1960s, some hot cells for the separation of plutonium from spent reactor fuel elements. The Shah wanted Iran to have a complete fuel cycle on its territory. In 1967 the Iranians started operating a small research reactor at the Tehran Nuclear Centre. The light-water reactor, fuelled with highly-enriched uranium, was supplied by the USA. It has a power output of 5 MW(th). In May 1987, Iran made a contract with Argentina to supply a new core for the research reactor at Tehran which is fuelled with uranium enriched to 20 per cent of U-235, and to supply the fuel for the reactor. The construction of two large nuclear-power reactors, supplied by the West German firm Kraftwerk Union AG, began in the mid-1970s at a location about 60 kilometres from Bushehr. Each reactor was to have a power output of 1,300 MW(e). The construction of another nuclear-power reactor, to be supplied by the French firm Framatome, was also started, at Darkhouin, near Ahwaz. It was to have a power output of 930 MW(e). Each of these three reactors was to be a light-water reactor, fuelled with low-enriched (3 per cent of U-235) uranium. According to the Shah’s plan, these reactors were to usher in an ambitious nuclear programme that would provide Iran with twenty or so nuclear-power reactors by the year 2000.

Iran's nuclear programme

100

The construction of the nuclear-power reactors was suspended when Ayatollah Ruhollah Khomeini came to power in 1979. But the Bushehr site was bombed by the Iraqis repeatedly during the Iran/Iraq war, and the partially built reactors badly damaged. Only a small amount of site preparation had been done on the Darkhouin site when work was suspended. When the Shah was deposed in 1979, there was an active group of nuclear scientists and engineers working at Tehran Nuclear Research Centre. But many nuclear scientists fled abroad when the Ayatollah came to power. Nevertheless, a new research centre was started up at Isfahan, 130 kilometres south of Tehran (Timmerman 1992). The Ayatollah’s regime attempted to restart work on the unfinished reactors. There were talks with France and West Germany to this end but they did not get far. Other efforts, involving, for example, the Argentinian firm, Enace, were nullified by the bombing of the reactor sites. There are also reports of negotiations with China for technical assistance to complete the reactors at Darkhouin and Bushehr. These apparently occurred when Chinese President Yang Shang Kun visited Iran in October 1991. But the status of these plans is unclear. And so is that of Iran’s plan to purchase two 440-MW(e) nuclear-power reactors from Russia, perhaps to be located near Gorgan on the Caspian Sea (Albright and Hibbs 1992b). Iran intends to exploit its uranium reserves and has begun to mine uranium at Saghand in the Yazd province, where exploration suggests that there are 5,000 tonnes of highgrade uranium ore. A large uranium processing plant began operating in 1989. But whether or not Iran will complete any or all of the three nuclear-power reactors is unclear. There are indications, however, that Iran does intend to reinvigorate its nuclear activities, including a nuclear-weapon programme, initiated by the Shah in the 1970s. On 23 October 1991, for example, Iranian Vice-President Mohajerani, in charge of parliamentary and legal affairs, stated that ‘because the enemy has nuclear facilities, the Muslim states should be equipped with the same capacity’. He wanted Iran, with its Muslim partners, to build an ‘Islamic bomb’ to counter Israel’s nuclear arsenal. The Islamic bomb was, he said, ‘not the business of the United Nations or of the Security Council’. Similar statements have been made by other Iranian officials since 1985. Reportedly, Iran has an agreement with China which includes the supply of a very small ‘mini’ reactor, with a power output of only 27 kW(th), and a calutron (Perabo 1992). The reactor is too small to have any military significance and will probably be used to produce radioisotopes for medical purposes. The calutron is too small to produce significant amounts of enriched uranium. But there is some suspicion that it may be used as a model to build larger versions which could produce significant amounts of weapongrade uranium. There have also been reports that Iran has acquired two or three nuclear weapons from Kazakhstan. These are tactical nuclear weapons from the arsenal of the former Soviet Union. It is also reported that nuclear-weapon experts from the former Soviet Union have been recruited by Iran. Although these reports come from several sources it is not possible to confirm them. In November 1991, Iran negotiated with India for the purchase of a 10-MW(th)

How nuclear weapons spread

101

research reactor to be installed at Moallem Kalayeh. After strenuous objections from the USA, India temporarily cancelled the sale. But by March 1992 the deal was apparently on again. Suspicions about Iran’s nuclear-weapon ambitions are fuelled by its clandestine efforts to acquire sensitive technologies from the industrialized countries, reminiscent of similar activities by Iraq before it invaded Kuwait. For example, the former Czechoslovakia has reportedly sold Iran quantities of the conventional explosive HMX, suitable for use in the explosive lenses to implode the sphere of fissile material in the core of a nuclear weapon. Many of the nuclear scientists who left Iran after the Shah was deposed have been returning to the country. Iranian students have been educated in universities in Australia, Germany, the UK, the USA and so on. Among the subjects studied are chemistry, electronics and metallurgy, as well as nuclear physics and engineering. These skills are those needed in a nuclear-weapon programme. Moreover, Iran has been pressing Eurodif to supply it with enriched uranium under the 1974 contract. Up to 1990, Iran refused to accept or pay for the enriched uranium. Even though a legal case has left Iran with an indirect share in Eurodif, the company will not supply any enriched uranium to Iran. This may be because the French government has promised the American government that it will not allow enriched uranium to be transferred to Iran. Iran ratified the NPT in 1970. Under its safeguards agreement with the IAEA, undeclared nuclear facilities can be inspected. In February 1992, an IAEA team visited a number of sites in Iran, including the nuclear centres in Bushehr, Isfahan and Tehran. The inspectors also visited facilities in Karaj and Moallem Kalayeh at which unconfirmed reports suggested that Iran may be developing gas centrifuges for uranium enrichment (Albright and Hibbs 1992b). The suspicion that Iran is developing gas centrifuges for uranium enrichment is related to reports of Iranian nuclear collaboration with Pakistan. In 1987, for example, Pakistan signed a secret agreement with Iran for the training of Iranian nuclear scientists and engineers in Pakistani nuclear institutes. Some also suspect that China is assisting Iran in developing gas centrifuge technology. The IAEA inspectors announced that the activities they had seen ‘were found to be consistent with the peaceful applications of nuclear energy’. The IAEA inspection of Iran has, however, been criticized as inadequate. The IAEA presented a list of sites it wanted to inspect to the Iranians in advance of the visit. The Iranians then selected the sites they felt were appropriate. It is said that the Moallem Kalayeh site visited by the team is not the one at which the Iranians are suspected of pursuing a clandestine uranium-enrichment programme. The Iranians reportedly took the inspectors to a village of the same name in another part of Iran. There is insufficient evidence to say whether or not Iran has a significant programme to develop nuclear weapons. In the words of David Albright and Mark Hibbs (ibid.): US officials say they have clear indications that Iran wants nuclear weapons. But so far, the US government has failed to identify any clandestine facilities in Iran that might be part of a secret nuclear weapons program. Despite numerous

Iran's nuclear programme

102

press reports alleging that Iran has secret installations to enrich uranium and develop nuclear weapons, one US official said: ‘We have not established that there are any secret nuclear facilities in Iran’

18 NUCLEAR PROLIFERATION TO SUBNATIONAL GROUPS The proliferation of nuclear weapons to countries that do not yet have them has received far more attention than the risk that sub-national groups may acquire nuclear explosives. But it is becoming increasingly realized that while the risk that some governments may take the political decision to acquire nuclear weapons may be decreasing, the risk that sub-national groups, such as terrorist groups, may acquire nuclear devices may be increasing. Until recently, the majority of commentators argued that the most likely way in which a sub-national group would acquire a nuclear explosive was by stealing a nuclear weapon from a nuclear-weapon stockpile or hijacking a nuclear warhead when it was being transported. But, as plutonium becomes more available, it is increasingly likely that a sub-national group will steal, or otherwise illegally acquire, plutonium and construct its own nuclear explosive device to detonate it or threaten to detonate it. Although a sub-national group could choose to use either plutonium or highly-enriched uranium as the fissionable material for nuclear explosives, plutonium is increasingly the more likely option. Most civilian highly-enriched uranium is currently used as fuel in research reactors. The development of new low-enriched uranium fuels for use in research reactors will sharply reduce the amount of civilian highly-enriched uranium in circulation. It should be emphasized, however, that large amounts of highly-enriched uranium will be come into the civilian sector as nuclear weapons are dismantled. As J.Carson Mark et al. point out (1987), it may be easier for terrorists to make a nuclear explosive from highly-enriched uranium than from even weapon-grade plutonium. This is because the neutron source from spontaneous fission in such material is smaller than that in even the best grades of plutonium by a factor of more than a thousand. In the relatively slow-moving gun-type device one might wish to assemble a couple of critical masses, or so, which would imply bringing together something like 50 kilogrammes of 94 per cent U-235, since the critical mass with a reflector can be about half the bare critical mass of 52 kilogrammes….Luis Alvarez, a scientist with the Manhattan Project during its war years, has said, ‘With modern weapons-grade uranium the background neutron rate is so low that terrorists, if they had such material, would have a good chance of setting off a high-yield explosion simply by dropping one half of the material on the other half’

Nuclear proliferation to sub-natioanl groups

104

The security of highly-enriched uranium in nuclear weapons is almost certainly much greater while the material is in military hands than when it is in civilian ones. This is why the prospect of having large amounts of highly enriched uranium from dismantled nuclear weapons under civilian control is so worrying. This material is the ideal fissile material for a sub-national group wanting to make a nuclear explosive, if it can get hold of enough of it. The amount of plutonium available will, as has been described, rapidly increase and it will become available in a number of countries. As the amount of plutonium produced worldwide in civilian nuclear-power reactors and separated from spent reactor fuel elements in commercial reprocessing plants increases, it will become easier to obtain plutonium illegally. The disintegration of the Soviet Union and the instability of Eastern Europe may make the theft of plutonium produced in Soviet-supplied reactors more likely. There is a very serious danger that no authority in these countries will protect fissile material properly, and that doing so will have a low priority, given the seriousness of all their other problems. In addition to civilian plutonium, there is the military plutonium removed from nuclear weapons. Some of this plutonium may be stolen or sold on the black market. Of particular concern is the risk that plutonium from ex-Soviet nuclear weapons may be illegally acquired. These weapons may be relatively secure while they are in the hands of the military, although the risk that a few of these weapons may get into the wrong hands is significant. But the weapons are handed over to civilian control for dismantling. The security of fissile material while it is under civilian control is likely to be much less than the security of it while it is under military control. The people handling the fissile material have such meagre wages that the temptation to steal material which has a high value on the black market may be virtually irresistible. The increasing availability of plutonium is not the only reason why the risk of nuclear terrorism is increasing. Other factors which increase the risk of nuclear terrorism include the relatively small amount of plutonium needed to fabricate a nuclear explosive, the availability in the open literature of the technical information needed to fabricate a nuclear device, and the small number of competent people needed to fabricate a primitive nuclear device. After reprocessing, reactor-grade plutonium is normally stored as plutonium oxide rather than plutonium metal. The critical mass of reactor-grade plutonium in the form of plutonium-oxide crystals is about 35 kilogrammes, if in spherical shape. The radius of this sphere of plutonium oxide would be about 9 centimetres. Reactor-grade plutonium oxide in uncompacted powder form has a critical mass of about 875 kilogrammes if in a sphere; the radius of the sphere would be about 45 cm. If a sub-national group acquires plutonium oxide it may convert it to plutonium metal, which can be done using straightforward chemical methods. The critical mass of reactorgrade plutonium in metal form is about 15 kilogrammes. The main problem with using reactor-grade plutonium as a nuclear weapon is that the spontaneous fission rate of plutonium-240 is much greater than that of plutonium-239. In

How nuclear weapons spread

105

reactor-grade plutonium the average time between spontaneous fissions is less than a micro-second (a millionth of a second). This means that very fast implosion techniques would be necessary in a nuclear device made from reactor-grade plutonium to prevent pre-detonation, which leads to uncertain explosive yields. Spontaneous fission produces a neutron background in a weapon-grade plutonium core of about one neutron every two or three microseconds. With a mean time of a few microseconds between neutrons—a very much longer time than the duration of the fission chain reaction—radial compression rates of a few millimetres per microsecond will prevent pre-detonation. Implosion techniques can achieve this without much difficulty. But for reactor-grade plutonium the mean time between neutrons is a small fraction of a micro-second. Extremely fast assembly would be needed to achieve super-criticality. Implosions technologies to provide the very high shock velocities and compression needed to prevent pre-detonation are available but probably not to a sub-national group, at least in the foreseeable future. The ease with which a sub-national group could construct a nuclear weapon is discussed in detail by J.Carson Mark, Theodore Taylor, Eugene Eyster, William Maraman, and Jacob Wechsler (Mark et al. 1987). They say that, so far as crude nuclear devices (devices guaran-teed to work without the need for extensive theoretical or experimental demonstration) are concerned: 1 Such a device could be constructed by a group not previously engaged in designing or building nuclear weapons, providing a number of requirements are adequately met. 2 Successful execution would require the efforts of a team having knowledge and skills additional to those usually associated with a group engaged in hijacking a transport or conducting a raid on a plant. 3 To achieve rapid turn-round (that is, to make the device ready within a day or so of obtaining the material), careful preparations extending over a considerable period would have to be carried out, and the materials acquired would have to be in the form prepared for. 4 The amounts of fissile material necessary would tend to be large—certainly several times the minimum quantity required by expert and experienced nuclear-weapon designers. 5 The weight of the complete device would also be large—not as large as the first atomic weapons (about 4.5 tonnes), since these required aerodynamic cases to enable them to be handled as bombs, but probably more than a tonne. 6 The conceivable option of using oxide powder (whether of uranium or plutonium) directly, with no post-acquisition processing or fabrication, would seem to be the simplest and most rapid way to make a bomb. However, the amount of material required would be considerably greater than if metal were used. 7 There are a number of obvious potential hazards in any such operation, among them those arising in the handling of a high explosive; the possibility of inadvertently inducing a critical configuration of the fissile material at some stage in the procedure; and the chemical toxicity or radiological hazards inherent in the materials used. Failure to foresee all the needs on these points could bring the

Nuclear proliferation to sub-natioanl groups

106

operation to a close; however, all the problems posed can be dealt with successfully provided appropriate provisions have been made. The nuclear devices considered by J.Carson Mark et al. (1987) are of types similar to those dropped on Nagasaki and Hiroshima. But much cruder designs that will still give a powerful nuclear explosion are possible. These could produce nuclear explosions equivalent to the explosion of between 100 and 1,000 tonnes of TNT. They might yield several thousand tonnes, but are unlikely to yield 10,000 tonnes. The crudest design could be very easily constructed by a team of technicians (or a competent technician working alone) from, say, a sub-critical mass of plutonium. The plutonium need not be in metal form; plutonium oxide, for example, is more convenient and safer to handle. The plutonium oxide could be contained in a spherical vessel placed in the centre of a large mass of conventional high explosive, such as TNT. When detonated remotely by an electronic signal the shock wave from the conventional explosive could compress the plutonium enough to produce some nuclear fission. In a primitive device, no effort would be made to focus the shock wave and so the high explosive would be simply stacked around the plutonium, probably in the form of a cube. A few detonators would probably be used, arranged to go off simultaneously. The device would easily fit into a medium-sized van. Even if the explosion from such a crude device was equivalent to the explosion of only a few tens of tonnes of TNT, it would completely destroy the centre of a relatively large city. The device might, however, explode with a considerably larger explosive power, equivalent to a few hundreds of tonnes of TNT. Even a thousand tonnes is not impossible. An explosion equivalent to that of 100 tonnes of TNT exploded on the surface would produce a crater about 30 metres across. In the words of Mason Willrich and Theodore Taylor: Under conceivable circumstances, a few people, possibly even one person working alone, who possessed about 10 kilogrammes of plutonium oxide and a substantial amount of chemical high explosive could, within several weeks, design and build a ‘crude fission bomb’. By a ‘crude fission bomb’ we mean one that would have an excellent chance of exploding with the power of at least 100 tonnes of chemical high explosive. This could be done using materials and equipment that could be purchased at a hardware store and from commercial suppliers of scientific equipment for student laboratories. (Willrich and Taylor 1974) A similar conclusion is drawn by the experts of the Office of Technology Assessment (OTA) of the US Congress. In its publication Nuclear Proliferation and Safeguards (1977), the OTA states that: A small group of people, none of whom have ever had access to the classified literature, could possibly design and build a crude nuclear explosive device.

How nuclear weapons spread

107

They would not necessarily require a great deal of technological equipment or have to undertake any experiments. Only modest machine-shop facilities that could be contracted for without arousing suspicion would be required. The financial resources for the acquisition of necessary equipment on open markets need not exceed a fraction of a million dollars. The group would have to include, at a minimum, a person capable of researching and understanding the literature in several fields and a jack-of-all-trades technician…. There is a clear possibility that a clever and competent group could design and construct a device which would produce a significant nuclear yield (i.e. a yield much greater than the yield of an equal mass of high explosive). A primitive nuclear explosive manufactured by a terrorist group, perhaps contained in a vehicle such as a van, could be positioned so that, even if it did not produce any nuclear fission, the explosion of the chemical high explosive would widely disperse the plutonium. Dispersal would be even more widespread if the explosion caused a fire. The plutonium would be scattered in the form of small particles, capable of being inhaled into the lung. Inhaled particles can become embedded in the lung and seriously irradiate surrounding lung tissue. Irradiation by the alpha-particles, given off when plutonium nuclei undergo radioactive decay, can cause lung cancer. Plutonium in particulate form is, therefore, extremely toxic. Reactor-grade plutonium is actually several times more radioactive than weapon-grade plutonium. This makes it a greater hazard to populations if dispersed. The threat of dispersion is perhaps the most likely danger that would follow the illegal acquisition of fissionable material by a sub-national group. The dispersal of some kilogrammes of the material would make a significant area, of a city, for example, uninhabitable until it had been decontaminated, a process that could take a long time. The very possession by a sub-national group of significant amounts of nuclear material is, therefore, a threat in itself. A government being blackmailed by a group known to have fissile material would not need to be convinced that the group had the expertise to construct an effective nuclear explosive. The authorities would know that if the device failed to produce a significant nuclear explosion it would almost certainly scatter nuclear material over a large area. And this would be threat enough.

19 PREVENTING THE SPREAD OF NUCLEAR WEAPONS The main international instrument to prevent the spread of nuclear weapons is the 1970 NPT. Although the vast majority of the world’s countries have joined the Treaty many believe it to be fragile. Some countries, Israel and India among them, sincerely doubt whether any international non-proliferation measure can protect their national security. They are particularly worried about the effectiveness of IAEA safeguards. To justify this concern the Director General of the IAEA, Dr Hans Blix, is quoted. On 11 December 1981, he said: The safeguards do not, of course, reveal what future intention the State may have. It may change its mind on the question of nuclear weapons and wish to produce them despite possible adherence to the NPT. Neither such adherence nor full-scope safeguards are full guarantees that the State will not one day make nuclear weapons. Ten years later Iraq proved his point. The abrogation clause, Article X of the NPT, is a major concern to some countries. Under it a party can at any time declare its withdrawal from the treaty with three months’ notice if it decides that ‘extraordinary events’ have occurred which ‘it regards as having jeopardized its supreme interests’. Under the NPT a country can legally manufacture the components of a nuclear weapon, notify the IAEA and the UN Security Council that it is withdrawing from the Treaty, and then assemble its nuclear weapons.

THE NPT Article I of the NPT commits the nuclear-weapon parties (China, France, Russia, the USA, the former USSR and the UK) not to transfer nuclear weapons and not to assist in their manufacture by the non-nuclear-weapon states. Article II commits the non-nuclearweapon states not to receive nuclear weapons or control over such weapons, and not to receive any assistance in the manufacture of nuclear weapons. Article III obligates the non-nuclear-weapon parties to sign agreements with the IAEA submitting all their nuclear activities to Agency safeguards. Article IV promises co-operation and assistance to non-nuclear-weapon states in their peaceful nuclear programmes. Article VI commits

Preventing the spread of nuclear wepons

109

the nuclear-weapon parties to take significant steps towards halting and reversing the nuclear arms race and towards nuclear disarmament. In 1995, a conference of the parties to the NPT will be held to decide for how much longer the Treaty should continue. Under particular scrutiny will be the effectiveness of the international safeguards system. It would be a serious blow to efforts to restrain the spread of nuclear weapons if the parties decided at the 1995 extension conference that the NPT is not worth maintaining.

IAEA SAFEGUARDS One of the main tasks of the 35-year-old International Atomic Energy Agency (IAEA) is to ‘administer safeguards designed to ensure that special fissionable and other materials, services, equipment, and information made available by the Agency or at its request or under its supervision or control are not used in such a way as to further any military purpose’ (Article III.5 of the IAEA Statute). The objective of safeguards is the timely detection (rather than the prevention) of the diversion of significant quantities of nuclear material from peaceful nuclear activities to the manufacture of nuclear weapons or other nuclear explosive devices and the deterrence of such diversion by the risk of early detection. The IAEA safeguards system includes the application of measures for materials accountancy, supplemented by containment and surveillance. IAEA safeguards begin to operate when an agreement is signed between the IAEA and the country owning the nuclear material under safeguards which gives the Agency the right to make ad hoc inspections, routine inspections and special inspections. Inspectors are sent to the country to verify information that the country must give to the Agency about the location, identity, quantity and composition of nuclear material subject to safeguards. Many exporters rely on the IAEA to safeguard nuclear material produced in exported nuclear facilities. As of the end of 1990, the IAEA had 177 safeguards agreements in force with 104 states. It had 515 nuclear installations under safeguards, including 185 power reactors, 170 reactors, 5 reprocessing plants and 7 enrichment plants (IAEA, 1990). The other main role of the IAEA is to promote the use of peaceful nuclear technology, as defined in Article II of its Statute: The Agency shall seek to accelerate and enlarge the contribution of atomic energy to peace, health and prosperity throughout the world. It shall ensure, so far as it is able, that assistance provided by it or at its request or under its supervision or control is not used in such a way as to further any military purpose. The problem is that military and peaceful nuclear programmes are, for the most part, virtually identical. In fact, the initial research and development of the nuclear fuel cycle

How nuclear weapons spread

110

was funded from military budgets. And, even today, the evolution of peaceful nuclear programmes depends, to a large extent, on the continuing interest in many countries in acquiring the capability to fabricate nuclear weapons. Given the relationship between military and civilian nuclear programmes, can a single agency effectively promote and safeguard nuclear technology at the same time? Many observers argue that it cannot. The effectiveness of the IAEA is, they say, seriously— perhaps fatally—jeopardized by its attempt to combine the promotion and the control of nuclear technology.

IMPROVING THE CREDIBILITY OF INTERNATIONAL NUCLEAR SAFEGUARDS Although 114 countries are Members of the IAEA, its budget for 1992 is only about $207 million, supporting a staff of some 2,000 professional and support staff. Currently, the Agency has about 930 nuclear facilities under safeguards in about sixty countries, requiring about 10,500 person-days of inspection a year. By the year 2000, this is likely to increase to over 15,000 person-days of inspection a year. The safeguards budget is, however, only about $65 million a year, just over a third of the Agency’s total budget. Moreover, the IAEA is planning a 13 per cent cut in its programmes, including safeguards, because the countries of the Commonwealth of Independent States are not making their contributions (the Soviet Union accounted for about 10 per cent of the IAEA budget). It is hard to see how the IAEA safeguards system can be made credible enough to strengthen the NPT, unless the Agency’s safeguards budget is increased. The discovery of the extent of Iraq’s nuclear-weapon programme was a blow to the IAEA safeguards system and to the NPT. Iraq has been a party to the Treaty since 1969 and IAEA inspectors have been visiting Iraq’s nuclear research centre at Tuwaitha about twice a year for many years. Nevertheless, they totally failed to discover any sign of Iraq’s nuclear-weapon programme, even though uranium-enrichment and plutoniumreprocessing research was going on under their noses. In fact, the calutron programme, for example, only came to light when an Iraqi defector blew the whistle to the Americans in May 1991. The case of Iraq shows that a country intent on doing so can illegally establish a nuclear-weapon programme while a party to the NPT, taking advantage of its membership of the Treaty to obtain assistance in acquiring nuclear technology from numerous foreign suppliers. That the IAEA failed to detect any of Iraq’s nuclear-weapon activities, all conducted while Iraq was apparently fulfilling its obligations under the NPT, has inevitably raised questions about the effectiveness of the NPT verification regime, and, therefore, about the worth of the Treaty itself. Some experts argue that a consequence of Iraq’s violation of its safeguards agreement with the IAEA and of its NPT obligations by not declaring its indigenous production of enriched uranium and plutonium could be the strengthening of the international safeguards system and, therefore, of the NPT. IAEA safeguards should be made more stringent by covering nuclear facilities as well as fissile materials. Also, the IAEA should

Preventing the spread of nuclear wepons

111

have the power to make ‘special inspections’ on facilities that NPT parties have excluded from safeguards. Incidentally, some lawyers argue that the IAEA already has the power to inspect these facilities if it suspects that fissile material is being illegally produced at them. There is no doubt that UN Security Council Resolutions, such as Resolution 687, passed after the defeat of Iraq, suggest that the Council is prepared to act against states that violate their NPT obligations. But it should be noted that it is one thing to coerce a defeated Iraq by preventing it continuing with its nuclear activities, and quite another to penalize in peacetime other potential proliferators, even if they are suspected of engaging in a clandestine nuclear-weapon programme. It should also be noted that Resolution 687 denies Iraq’s rights, under Article IV of the NPT, ‘to develop research, production, and use of nuclear energy for peaceful purposes without discrimination’. Will the UN be prepared in the future to intervene to such an extent in the internal affairs of sovereign states, even to the extent of taking military action, to prevent proliferation? Will the Special Commission set up to deal with Iraq be given powers to deal with other proliferators? Perhaps more intrusive IAEA inspections will be evolved. But it is difficult to be optimistic about truly coercive international actions, other than diplomatic ones. It will, for one thing, be difficult to persuade Third World countries that any threat to their sovereignty is not a new imperialism. Whatever happens, the actions against Iraq, including the bombing of its research reactors and the post-war dismantling of its nuclear-weapon programme, may be seen by other developing countries as making the international non-proliferation regime even more discriminatory than many Third World countries already believe it to be. The suspicion will be enhanced that the main aim of the declared nuclear-weapon powers (i.e. the permanent members of the Security Council) is to maintain their monopoly on nuclear weapons. It will be hard to dispel these suspicions if these powers fail, for example, to persuade Israel to give up its nuclear weapons.

DO WE NEED A NEW INTERNATIONAL NUCLEAR CONTROL AGENCY? If the clandestine production of nuclear weapons in other countries is to be prevented, there are good arguments for a new international agency with the authority and resources to conduct surprise inspections of undeclared nuclear facilities in any country, even though such actions may be seen as an intrusion into national sovereignty. To be effective, such an agency would need access to military intelligence information about the existence and location of undeclared nuclear facilities. Other tasks for such an agency will emerge. For example, new safeguards responsibilities will arise as the amount of civilian and military plutonium continues to grow. With so much fissile material available and governments and sub-national groups willing to pay large sums of money for it, a flourishing nuclear black market is a very real possibility. When there is a surplus, the price will fall and sub-national groups will be

How nuclear weapons spread

112

more able to afford it. Given these dangers, there is a clear need for an international agency to deal with nuclear smuggling. Such an agency should have severe powers—a sort of nuclear Interpol. The crucial question is: Is the IAEA, with its dual nuclear promotional and control interests, able to evolve into a credible investigative and policing agency, able to operate an intrusive safeguards system, and plutonium and highly-enriched uranium stores? Or should a new international agency be established for all these purposes, reporting directly to the United Nations Security Council and responsible only for the control of nuclear energy?

20 THE PROLIFERATION OF NUCLEARWEAPON DELIVERY SYSTEMS Third World countries which acquire nuclear weapons are most likely to deliver them by combat aircraft or ballistic missiles. An ordinary nuclear fission weapon (of the ‘first generation’) developed today is likely to weigh no more than 200 or 300 kilogrammes and have a diameter of less than 60 centimetres. This is well within the payload of typical combat aircraft and ballistic missiles. A variety of nuclear-capable aircraft has been supplied, through the global arms trade, to a large number of air forces. Examples of western nuclear-capable tactical aircraft transferred to other countries include the F-104G/S Starfighter, F-4E/F Phantom, F-16 Fighting Falcon, Mirage IVP, Mirage 2000N, Jaguar A, Tornado IDS, Super Etendard, Sea Harrier and Buccaneer. Russian nuclear-capable tactical aircraft sold abroad include the MiG-27 Flogger D/J, Su-17 Fitter D/H/K and Su-24 Fencer. Chinese nuclear-capable tactical aircraft sold abroad include the J-6. Western nuclear-capable tactical surface-to-surface ballistic missiles include the US Lance; Russian nuclear-capable tactical surface-to-surface ballistic missiles include the FROG, SCUD and SS-21. The Third World countries which have developed, or are suspected of developing, nuclear weapons have some of these weapon systems in their arsenals. They could be modified to carry nuclear weapons relatively easily and quickly. Nuclear weapons could, of course, also be delivered in relatively primitive ways. They are small enough to be carried even in a small ship. A ship carrying a nuclear weapon could be sailed into a harbour and the weapon detonated. A terrorist group could build a nuclear explosive in, for example, a motor vehicle, such as a van. The vehicle could be parked in the centre of a city and the weapon detonated remotely by, for example, a radio signal.

BALLISTIC MISSILES The 1991 Gulf War focused attention on the global spread of nuclear capable weapon systems. But there is most concern about the proliferation of ballistic missiles, brought on by Iraqi attacks on Israel and coalition forces in Saudi Arabia, using SCUD surface-tosurface missiles. Surface-to-surface ballistic missiles are very effective systems for delivering nuclear weapons, particularly over long distances. These missiles are, however, expensive. Given

The proliferation of nuclear-weapon delivery systems

114

their cost, and the relatively small payloads they carry, their use to deliver conventional warheads is not cost-effective. More and more countries are acquiring ballistic missiles through the global arms trade. In addition, several Third World countries are able to design and manufacture them indigenously (Karp 1988). The more producers there are, the easier it is to acquire missiles, either on the global arms trade or illegally. Third World countries are often willing suppliers of weapons to other Third World countries. Currently, twenty Third World countries either possess ballistic missiles or are trying to develop them. These are Algeria, Argentina, Brazil, Egypt, Greece, India, Iran, Iraq, Israel, Libya, North Korea, South Korea, Kuwait, Pakistan, Saudi Arabia, South Africa, Syria, Taiwan, Turkey and Yemen (Karp 1991). Sub-national groups may, in the future, acquire and use ballistic missiles as delivery systems for nuclear weapons. This risk is obviously enhanced if significant numbers of ballistic missiles are deployed by many countries. Details of the proliferation of ballistic missiles in the Third World are shown in tables 1 to 4 inclusive, extracted from SIPRI data (SIPRI 1991). The extent to which ballisticmissile production has already spread into Third World countries is shown in table 1. The Third World countries producing short-range missiles are shown in table 2. Imported ballistic missiles in service are shown in table 3. Third World countries developing and producing ballistic missiles have usually received technical assistance from industrialized countries. Suppliers of ballistic-missile technology to Middle East countries are shown in table 4. Before the Gulf War, Iraq was making rapid progress in the development of several types of ballistic missiles. North Korea has shown significant expertise in ballistic-missile development. Not only is it manufacturing Soviet SCUD missiles but it is developing the SCUD-PIP (SCUD Product Improvement Programme) having a maximum range of about 500 kilometres. This development is noteworthy because it would give North Korea, suspected of having nuclear-weapon ambitions, a good nuclear delivery system. India and Israel are making steady progress with their missile programmes. India has two major ballistic-missile projects, as well as two space-launch projects. The Prithvi 240-kilometre range missile, comparable with the US Lance surface-to-surface missile, will soon be deployed by the Indian Army. It could carry a nuclear warhead. The Agni missile has a range of about 2,400 kilometres. Now under development, the missile could give India a nuclear delivery system with an intermediate range ballisticmissile capability. India is also developing the Augmented Satellite Launch Vehicle which could be the basis of an intercontinental ballistic missile, if India decided it needed one. Israel’s Jericho-II missile was probably developed to carry nuclear warheads. The sophistication of Israel’s missile activities is shown by its development of the Arrow antitactical ballistic-missile interceptor. A country capable of developing this system is clearly able to produce long-range and very accurate surface-to-surface ballistic missiles for the delivery of nuclear warheads. Although Pakistan has produced Hatf missiles they are probably not effective delivery systems for nuclear warheads. Pakistan’s efforts to develop a suitable missile have been

How nuclear weapons spread

115

frustrated by the difficulty it has had in acquiring missile technology because of international controls on the transfer of this technology. Some Third World countries are developing sounding rockets, for use for researching the properties of the upper atmosphere, and space-launch vehicles, to put satellites into orbit. Both of these technologies help countries to develop long-range ballistic missiles. The rockets used in these civilian activities are, in fact, basically similar to military missiles. Brazil, India and Pakistan have each developed two types of sounding rocket; and Indonesia is developing one type. Israel has developed a space-launch vehicle and has actually launched two satellites into orbit. As mentioned above, India is developing two types of space-launchers. Brazil and Taiwan are also trying to develop these vehicles. And South Korea and Pakistan plan to do so.

COMBAT AIRCRAFT All the Third World countries having, or suspected of developing, nuclear weapons operate aircraft which could be used to deliver nuclear weapons. Examples of suitable aircraft deployed by these countries are: F-4 aircraft, which have a weapon load of up to about 6,000 kilogrammes; F-16 aircraft, which have weapon loads of up to about 5,500 kilogrammes; Jaguar aircraft, which can deliver up to about 4,700 kilogrammes; MiG-27 aircraft which can carry a load of up to about 4,500 kilogrammes; and J-6 aircraft, which can carry a weapon load of about 500 kilogrammes. The Indian air force operates Jaguar and MiG-27 Flogger aircraft. The Iranian air force operates F-4 Phantom and Su-24 Fencer aircraft and has recently bought some Tu-22M Backfire long-range bombers from Russia. The Iraqi air force has MiG-27 Flogger, Mirage F1EQ5 and Mirage 200 aircraft. The Israeli air force operates F-4 Phantom, F-16 Fighting Falcon and Kfir aircraft. The Pakistani air force operates Mirage and F-16 Fighting Falcon aircraft. The North Korean air force operates J-6 and Su-25 Frogfoot aircraft. All these strike aircraft can deliver nuclear weapons effectively.

Table 1 Ballistic missiles under development or being produced in the Third World Range 40–150 km 150–600 km over 600 km Brazil EE-150 SS-300 SS-60 X-40 Egypt Saqr-80 Badr-2000 SCUD 100 India Prithvi Agni

The proliferation of nuclear-weapon delivery systems

Indonesia Iran Iraq

Israel Libya North Korea Pakistan Saudi Arabia South Africa Taiwan Thailand

RX-250 Oghab Shanin-2 Nazeat Laith Ababil 50 Sijeel 60 Ababil 100 MAR-350 SCUD B Hatf-2 SS-60

116

al-Hussein Fahd

al-Abbas Condor 2 Tamuz-1

Jericho I

Jericho II Ittisalt

SCUD PIP Hatf-1 Jericho II

Ching Feng Thanu Fan

Table 2 Short-range missiles produced in Third World countries Designation Missile type Argentina Martin Pescador anti-tank Mathago anti-tank Brazil Pirhana air-to-air Carcara air-to-surface Israel Barak surface-to-air Gabriel-1/2/3 ship-to-ship Gabriel-3A/S air-to-ship Picket anti-tank air-to-air Python-3 air-to-air Shafrir-2 Taiwan Kun Wu anti-tank

How nuclear weapons spread

117

Table 3 Other surface-to-surface ballistic missiles in service in Third World countries Designation Range(km) Supplier Afghanistan SCUD B 280 the former USSR Algeria FROG-7 70 the former USSR Cuba FROG-4 50 the former USSR FROG-7 70 the former USSR SCUD B 280 the former USSR Egypt FROG-5 50 the former USSR FROG-7 70 the former USSR Iran SCUD B 280 the former USSR North Korea SCUD-PIP 480 Iraq SCUD B 280 the former USSR FROG-7 70 the former USSR SS-60 60 Brazil Israel Lance 120 USA North Korea FROG-5 50 the former USSR FROG-7 70 the former USSR South Korea Honest John 37 USA Nike Hercules 240 USA Kuwait FROG-7 70 the former USSR Libya SCUD B 280 the former USSR FROG-7 70 the former USSR M-9 600 China Saudi Arabia CSS-2 2700 China SS-60 60 Brazil SCUD B 280 the former USSR Syria FROG-7 70 the former USSR SS-21 Scarab 120 the former USSR M-9 600 China Taiwan Honest John 37 USA Honest John 37 USA Turkey Yemen SCUD B 280 the former USSR FROG-7 70 the former USSR SS-21 Scarab 120 the former USSR

The proliferation of nuclear-weapon delivery systems

118

Table 4 Suppliers of ballistic-missile technology to Middle East countries Recipient Supplier Technology supplied Algeria the former FROG launchers and missiles, training USSR Egypt France Saqr design and technical assistance the former FROG and SCUD launchers and missiles; USSR SCUD 100 missiles North Korea SCUD B production assistance; SCUD 100 design and technical assistance Iran China Oghab design and production assistance North Korea Oghab production assistance; SCUD missiles SCUD missiles and launchers Libya SCUD missiles Syria Iraq Brazil SS-60 launchers and missiles, training; al Hussein and al Abbas training al Hussein and al Abbas technical assistance Egypt the former FROG and SCUD launchers and missiles, USSR training Yugoslavia design and technical assistance in missile development France design and technical assistance in missile development Israel USA Lance launchers and missiles, training France Jericho I design and production assistance Kuwait the former FROG launchers and missiles, training USSR Libya the former FROG and SCUD launchers and missiles, USSR training Otrag and Ittisalt design and technical West assistance Germany Saudi Brazil SS-60 launchers and missiles, training Arabia China CSS-2 launchers and missiles, training Syria the former FROG, SS-21 Scarab, SCUD launchers and

How nuclear weapons spread

Yemen

USSR the former USSR

119

missiles, training SS-21 Scarab launchers and missiles, FROG, SCUD launchers and missiles, training

APPENDIX TEXT OF THE TREATY ON THE NON-PROLIFERATION OF NUCLEAR WEAPONS1 1. The States concluding this Treaty: hereinafter referred to as the ‘Parties to the Treaty’, 2. Considering the devastation that would be visited upon all mankind by a nuclear war and the consequent need to make every effort to avert the danger of such a war and to take measures to safeguard the security of peoples, 3. Believing that the proliferation of nuclear weapons would seriously enhance the danger of nuclear war, 4. In conformity with resolutions of the United Nations General Assembly calling for the conclusion of an agreement on the prevention of wider dissemination of nuclear weapons, 5. Undertaking to co-operate in facilitating the application of International Atomic Energy Agency safeguards on peaceful nuclear activities, 6. Expressing their support for research, development and other efforts to further the application, within the framework of the International Atomic Energy Agency safeguards system, of the principle of safeguarding effectively the flow of source and special fissionable materials by use of instruments and other techniques at certain strategic points, 7. Affirming the principle that the benefits of peaceful applications of nuclear technology, including any technological by-products which may be derived by nuclear-weapon States from the development of nuclear explosive devices, should be available for peaceful purposes to all Parties to the Treaty, whether nuclearweapon or non-nuclear-weapon States, 8. Convinced that, in furtherance of this principle, all Parties to the 1

The text is taken from UN document A/RES/2373(XXII).

Treaty are entitled to participate in the fullest possible exchange of scientific information for, and to contribute alone or in co-operation with other States to, the further development of the application of atomic energy for peaceful purposes, 9. Declaring their intention to achieve at the earliest possible date the cessation of the nuclear arms race and to undertake effective measures in the direction of nuclear disarmament, 10. Urging the co-operation of all States in the attainment of this objective, 11. Recalling the determination expressed by the Parties to the 1963 Treaty banning nuclear weapon tests in the atmosphere, in outer space and under water in its

Appendix

121

Preamble to seek to achieve the discontinuance of all test explosions of nuclear weapons for all time and to continue negotiations to this end, 12. Desiring to further the easing of international tension and the strengthening of trust between States in order to facilitate the cessation of the manufacture of nuclear weapons, the liquidation of all their existing stockpiles, and the elimination from national arsenals of nuclear weapons and the means of their delivery pursuant to a treaty on general and complete disarmament under strict and effective international control, 13. Recalling that, in accordance with the Charter of the United Nations, States must refrain in their international relations from the threat or use of force against the territorial integrity or political independence of any State, or in any other manner inconsistent with the Purposes of the United Nations, and that the establishment and maintenance of international peace and security are to be promoted with the least diversion for armaments of the world’s human and economic resources, Have agreed as follows: Article I Each nuclear-weapon State Party to the Treaty undertakes not to transfer to any recipient whatsoever nuclear weapons or other nuclear explosive devices or control over such weapons or explosive devices directly, or indirectly; and not in any way to assist, encourage, or induce any non-nuclear-weapon State to manufacture or otherwise acquire nuclear weapons or other nuclear explosive devices, or control over such weapons or explosive devices. Article II Each non-nuclear-weapon State Party to the Treaty undertakes not to receive the transfer from any transferor whatsoever of nuclear weapons or other nuclear explosive devices or of control over such weapons or explosive devices directly or indirectly; not to manufacture or otherwise acquire nuclear weapons or other nuclear explosive devices; and not to seek or receive any assistance in the manufacture of nuclear weapons or other nuclear explosive devices. Article III 1. Each non-nuclear-weapon State Party to the Treaty undertakes to accept safeguards, as set forth in an agreement to be negotiated and concluded with the International Atomic Energy Agency in accordance with the Statute of the International Atomic Energy Agency and the Agency’s safeguards system, for the exclusive purpose of verification of the fulfilment of its obligations assumed under this Treaty with a view to preventing diversion of nuclear energy from peaceful uses to nuclear weapons or other nuclear explosive devices. Procedures for the safeguards required by this article shall be followed with respect to source or special fissionable material whether it is being produced, processed or used in any principal nuclear facility or is outside any such facility. The safeguards required by this article shall

Appendix

122

be applied on all source or special fissionable material in all peaceful nuclear activities within the territory of such State, under its jurisdiction, or carried out under its control anywhere. 2. Each State Party to the Treaty undertakes not to provide: (a) source or special fissionable material, or (b) equipment or material especially designed or prepared for the processing, use or production of special fissionable material, to any nonnuclear-weapon State for peaceful purposes, unless the source or special fissionable material shall be subject to the safeguards required by this article. 3. The safeguards required by this article shall be implemented in a manner designed to comply with article IV of this Treaty, and to avoid hampering the economic or technological development of the Parties or international co-operation in the field of peaceful nuclear activities, including the international exchange of nuclear material and equipment for the processing, use or production of nuclear material for peaceful purposes in accordance with the provision of this article and the principle of safeguarding set forth in the Preamble of the Treaty. 4. Non-nuclear-weapon States Party to the Treaty shall conclude agreements with the International Atomic Energy Agency to meet the requirements of this article either individually or together with other States in accordance with the Statute of the International Atomic Energy Agency. Negotiation of such agreements shall commence within 180 days from the original entry into force of this Treaty. For States depositing their instruments of ratification or accession after the 180-day period, negotiation of such agreements shall commence not later than the date of such deposit. Such agreements shall enter into force not later than eighteen months after the date of initiation of negotiations. Article IV 1. Nothing in this Treaty shall be interpreted as affecting the inalienable right of all the Parties to the Treaty to develop research, production and use of nuclear energy for peaceful purposes without discrimination and in conformity with articles I and II of this Treaty. 2. All the Parties to the Treaty undertake to facilitate, and have the right to participate in, the fullest possible exchange of equipment, materials and scientific and technological information for the peaceful uses of nuclear energy. Parties to the Treaty in a position to do so shall also co-operate in contributing alone or together with other States or international organizations to the further development of the applications of nuclear energy for peaceful purposes, especially in the territories of non-nuclear-weapon States Party to the Treaty, with due consideration for the needs of the developing areas of the world. Article V Each Party to the Treaty undertakes to take appropriate measures to ensure that, in accordance with this Treaty, under appropriate international observation and through appropriate international procedures, potential benefits from any peaceful applications of nuclear explosions will be made available to non-nuclear-weapon States Party to the

Appendix

123

Treaty on a non-discriminatory basis and that the charge to such Parties for the explosive devices used will be as low as possible and exclude any charge for research and development. Non-nuclear-weapon States Party to the Treaty shall be able to obtain such benefits, pursuant to a special international agreement or agreements, through an appropriate international body with adequate representation of non-nuclear-weapon States. Negotiations on this subject shall commence as soon as possible after the Treaty enters into force. Non-nuclear-weapon States Party to the Treaty so desiring may also obtain such benefits pursuant to bilateral agreements. Article VI Each of the Parties to the Treaty undertakes to pursue negotiations in good faith on effective measures relating to cessation of the nuclear arms race at an early date and to nuclear disarmament, and on a treaty on general and complete disarmament under strict and effective international control. Article VII Nothing in this Treaty affects the right of any group of States to conclude regional treaties in order to assure the total absence of nuclear weapons in their respective territories. Article VIII 1. Any Party to the Treaty may propose amendments to this Treaty. The text of any proposed amendment shall be submitted to the Depositary Governments which shall circulate it to all Parties to the Treaty. Thereupon, if requested to do so by one third or more of the Parties to the Treaty, the Depositary Governments shall convene a conference, to which they shall invite all the Parties to the Treaty, to consider such an amendment. 2. Any amendment to this Treaty must be approved by a majority of the votes of all the Parties to the Treaty, including the votes of all nuclear-weapon States Party to the Treaty and all other Parties which, on the date the amendment is circulated, are members of the Board of Governors of the International Atomic Energy Agency. The amendment shall enter into force for each Party that deposits its instrument of ratification of the amendment upon the deposit of such instruments of ratification by a majority of all the Parties, including the instruments of all nuclear-weapon States Party to the Treaty and all other Parties which, on the date the amendment is circulated, are members of the Board of Governors of the International Atomic Energy Agency. Thereafter, it shall enter into force for any other Party upon the deposit of its instrument of ratification of the amendment. 3. Five years after the entry into force of this Treaty, a conference of Parties to the Treaty shall be held in Geneva, Switzerland, in order to review the operation of this Treaty with a view to assuring that the purposes of the Preamble and the provisions of the Treaty are being realized. At intervals of five years thereafter, a majority of the Parties to the Treaty may obtain, by submitting a proposal to this effect to the Depositary Governments, the convening of further conferences with the same

Appendix

124

objective of reviewing the operation of the Treaty. Article IX 1. This Treaty shall be open to all States for signature. Any State which does not sign the Treaty before its entry into force in accordance with paragraph 3 of this article may accede to it at any time. 2. This Treaty shall be subject to ratification by signatory States. Instruments of ratification and instruments of accession shall be deposited with the Government of the Union of Soviet Socialist Republics, the United Kingdom of Great Britain and Northern Ireland and the United States of America, which are hereby designated the Depositary Governments. 3. This Treaty shall enter into force after its ratification by the States, the Governments of which are designated Depositaries of the Treaty, and forty other States signatory to this Treaty and the deposit of their instruments of ratification. For the purposes of this Treaty, a nuclear-weapon State is one which has manufactured and exploded a nuclear weapon or other nuclear explosive device prior to 1 January 1967. 4. For States whose instruments of ratification or accession are deposited subsequent to the entry into force of this Treaty, it shall enter into force on the date of the deposit of their instruments of ratification or accession. 5. The Depositary Governments shall promptly inform all signatory and acceding States of the date of each signature, the date of deposit of each instrument of ratification or of accession, the date of the entry into force of this Treaty, and the date of receipt of any requests for convening a conference or other notices. 6. This Treaty shall be registered by the Depositary Governments pursuant to article 102 of the Charter of the United Nations. Article X 1. Each Party shall in exercising its national sovereignty have the right to withdraw from the Treaty if it decides that extraordinary events, related to the subject matter of this Treaty, have jeopardized the supreme interests of its country. It shall give notice of such withdrawal to all other Parties to the Treaty and to the United Nations Security Council three months in advance. Such notice shall include a statement of the extraordinary events it regards as having jeopardized its supreme interests. 2. Twenty-five years after the entry into force of the Treaty, a conference shall be convened to decide whether the Treaty shall continue in force indefinitely, or shall be extended for an additional fixed period or periods. This decision shall be taken by a majority of the Parties to the Treaty.

REFERENCES Albright, D. (1992), ‘Bomb potential for South America’, The Bulletin of the Atomic Scientists 45, 4 (May), 16–20. Albright, D, and Feiveson, H.A. (1992), ‘Plutonium recycling and the problem of nuclear proliferation’, in F. Barnaby (ed.), Plutonium and Security: The Military Aspects of the Plutonium Economy, London: Macmillan. Albright, D. and Hibbs, M. (1992a), ‘Iraq’s bomb: blueprints and artifacts’, The Bulletin of the Atomic Scientists 48, 1 (January/February), 8–10. ——(1992b), ‘uSpot-light shifts to Iran’, The Bulletin of the Atomic Scientists 48, 2 (March), 9–11. ——(1992c), ‘Iraq’s “shop till you drop atomic programme” ’, The Bulletin of the Atomic Scientists 48, 3 (April), 26–37. ——(1992d), ‘It’s all over at Al Atheer’, The Bulletin of the Atomic Scientists 48, 5 (June) 8–10. ——(1992e), ‘Pakistan’s bomb: out of the closet’, The Bulletin of the Atomic Scientists 48, 6 (July/August), 38–43. ——(1992f), ‘India’s silent bomb’, The Bulletin of the Atomic Scientists 48, 7 (September), 27–31. ——(1992g), ‘North Korea’s plutonium puzzle’, The Bulletin of the Atomic Scientists 48, 9 (November), 36–40. Albright, D., Berkhout, F., and Walker, W. (1993), World Inventory of Plutonium and Highly Enriched Uranium, 1992, Oxford: Oxford University Press. Barnaby, F. (1989), The Invisible Bomb: The Nuclear Arms Race in the Middle East, London: I.B.Tauris. Bebbington, W. (1976), ‘The reprocessing of nuclear fuels’, Scientific American (December), 30–41. Berkhout, F., Diakov, A., Feiveson, H., Hunt, H., Miller, M. and von Hippel, F. (1992), Disposition of Separated Plutonium, Princeton: Center for Energy and Environmental Studies. Bermudez, J.S. (1991), ‘North Korea’s nuclear programme’, Jane’s Intelligence Review (September), 404–11. Christian Democrat International (1992), Nuclear War: An Ever-present Possibility, Brussels: DC Info Documents, 44 (July), 6. Dirks, W. (1992), ‘World status report: plutonium’, Energy Economist (May), 127. Factfile (1992), Arms Control Today, 22, 1 (January/February), 53. Fetter, S. (1988), Towards a Comprehensive Test Ban, Cambridge, Mass.: Ballinger. Fischer, D. and Szasz, P. (1985), Safeguarding The Atom, London: Taylor & Francis. Garwin, R.L. (1992), ‘Post-Soviet nuclear command and security’, Arms Control Today,

References

126

22, 1 (January/February), 18–23. Gowing, M. (1974), Independence and Deterrence, London: Macmillan. Greenpeace, (1992) The World Nuclear Industry. Status Report: 1992, London: Greenpeace. Hansen, C. (1988), US Nuclear Weapons: The Secret History, New York: Orion Books. Hersh, S.M. (1991), The Samson Option: Israel America, and The Bomb, London: Faber & Faber. International Atomic Energy Agency (IAEA) (1990), The Agency’s Programme and Budget for 1991 and 1992, Vienna: IAEA. ——(1992a), ‘Nuclear power status around the world’, IAEA Bulletin, 34, 2, 58. ——(1992b), IAEA Inspections and Iraq’s Nuclear Capabilities, Vienna: IAEA. International Institute for Strategic Studies (IISS) (1992), The Military Balance, 1992– 1993, London: IISS. Karp, A. (1988), ‘The frantic third world quest for ballistic missiles’, The Bulletin of the Atomic Scientists 44 (June), 5. ——(1991), ‘Ballistic missile proliferation’, World Armaments and Disarmament, SIPRI Yearbook 1991, Oxford: Oxford University Press. Lovins, A.B. (1980), ‘Nuclear weapons and power-reactor plutonium’, Nature, 28 February, and typographical corrections 13 March. Mark, Carson J., Taylor, T., Eyster, E., Maraman, W., and Weschler, J. (1987), ‘Can terrorists build nuclear weapons?’, in P. Leventhal and Y. Alexander (eds), Preventing Nuclear Terrorism, Lexington, Mass.: Lexington Books. Mochavez, Z. (1991), Nuclear Weapons Proliferation in the Indian Subcontinent, London: Macmillan. Nuclear Fuel (1992), 17, 7 (30 March), 5. Nuclear Proliferation and Safeguards (1977), US Congress, Office of Technology Assessment (OTA), Washington DC: OTA. Pean, P. (1982), Les deux bombes, Paris: Fayard. Perabo, B. (1992), A Chronology of Iran’s Nuclear Programming, Monterey: Monterey Institute of International Studies. Redick, J.R. (1990), Argentina and Brazil: An Evolving Nuclear Relationship, Southampton: Programme for Promoting Nuclear Non-Proliferation, Occasional Paper 7 (July). Rotblat, J. (1981), Nuclear Radiation in Warfare, London: Taylor & Francis. Scheinman, L. and Fischer, D.A.V. (1992), ‘Managing the coming glut of nuclear weapon materials’, Arms Control Today 22, 2 (March), 7–12. Simpson, J. (1992), Nuclear Non-Proliferation: The Evolving Context, Paper given to the Uranium Institute Annual Symposium 1992, London: Uranium Institute. Spector, L.S. (1990), Nuclear Ambitions, Boulder: Westview Press. Stockholm International Peace Research Institute (SIPRI) (1991), World Armaments and Disarmament, SIPRI Yearbook 1992, Oxford: Oxford University Press. Timmerman, K.R. (1992), Weapons of Mass Destruction: The Cases of Iran, Syria, and Libya, A Simon Wiesenthal Report, California: Simon Wiesenthal Center. The Uranium Institute (1991), Uranium and Nuclear Energy: 1991, London: Uranium

References

127

Institute. Willrich, M., and Taylor, T. (1974), Nuclear Theft: Risks and Safeguards, Cambridge, Mass.: Ballinger.