Can we fix nuclear energy?

Annals ¢~l Nmlear Eneryy, Vol. 6~ pp. 473 to 482 0306-4549,'79q0014)473S02.(X),0 © Pergamon Press Ltd. 1979. Printed in Great Britain

CAN WE FIX NUCLEAR ENERGY?*

A. M. WEINBERG

Institute for Energy Analysis, Oak Ridge Associated Universities, Oak Ridge, Tennessee, U.S.A.

(Received 30 April 1979)

A 1 million kW pressurized water reactor contains 15 billion Ci radioactivity. This is about equal to the natural radioactivity in all the oceans; radioactivity that accompanies the decay of the 4 billion tons of uranium and its daughters dissolved in the seas. Nothing except time can turn radioactivity off. The radiation from an X-ray machine ceases when the technician turns the machine off. By contrast, the radioactivity in a reactor decays only .slowly after the reactor is shut off; it contributes about 200,000 kW heat while the reactor is running and, depending upon how long the reactor has been running before shut- down, is still generating 8500 kW a week later from the remaining 2 billion Ci. A large chain reactor must therefore be cooled even after the reaction has ceased or the fuel will melt. If it melts, radioactivity will excape from the fuel.

From the beginning in 1942, all of us at Arthur Compton's Metallurgical Laboratory in Chicago where the first chain reaction was established sensed that mankind was crossing a threshold when he learned howqo create at will radioactivity on an enor- mous scale. Up until then radioactivity was measured in pCi or mCi--I g radium, costing $50,000, was equal to I Ci. Enrico Fermi, the developer of the chain reactor and our scientific leader, on occasion would remind us of this. It was not only the bomb that changed things--it was also the creation of uni- maginably large amounts o.f radioactivity.

It is no wonder that from the very beginning safety was an obsession among the developers of nuclear energy. My first job at Chicago was to help Edward Teller estimate how much radioactive carbon would be spewed into the atmosphere from a small air- cooled reactor we were planning to build at Oak Ridge. And it was Edward Teller, more than anyone else, who insisted that the safety of reactors must pre- dominate. He was the founder, and first chairman in 1948, of what is now called the Advisory Committee on Reactor Safety (ACRS). No reactor could be built

* This article will also appear in the Wilson Quarterly.

unless the ACRS had, after sharp scrutiny, agreed 'that there is reasonable assurance that it can be operated without undue risk to the health and safety of the public'--to quote the officialese that ended every letter of approval from the ACRS to the Atomic Energy Commission.

The simplest way to assure that no member of the public would be hurt by the release of radioactivity from a malfunctioning reactor is to site the reactor remotely. To be sure, the first chain reaction took place on 2 December 1942 on 57th Street and Ellis Avenue in the heart of Chicago's South Side. Even this tiny venture into large scale radioactivity gave General Groves and Arthur Compton plenty of anxiety, but every day counted during the war. To await the completion of the site at Argonne Forest Preserve on the outskirts of Chicago would have lost too much time. The reactor was moved there a few months later, over the protests of many of us, since the total radioactivity was not much more than that of 15 kg or so of radium. But we took for granted at the time that the large plutonium-producing reactors, as well as a smaller pilot plant reactor, would be remotely sited--the latter at Oak Ridge in the hills of Tennessee (site X), the former on the huge Hanford Reservation in eastern Washington (site W). Most of the other reactors built by the Atomic Energy Commission were confined to these remote places and to three others--Savannah River, Idaho Falls and Los Alamos. All of these places were, at the time, far from people; and as the supporting towns such as Richland, Washington and Oak Ridge, Tennessee developed they were peopled by experts who had daily experience in the handling of large-scale radioactivity. They constituted a sort of new class who understood radioactivity, who knew how much was dangerous and how much was a tiny addition to the all-pervasive natural background.

Had all our power reactors been sited as remotely as were the original Hanford, Idaho Falls or Savan- nah River reactors, I believe the nuclear enterprise might have avoided much of the travail that it is now

473

474 A.M. WEINBERG

undergoing. We would have had by this time perhaps 25 sites, each remote, each surrounded by a large un- populated zone and bordered by supporting villages inhabited by people who worked at the plants and for whom large-scale radioactivity was part of life. I suppose, had the nuclear energy enterprise remained a government monopoly as it was under the Atomic Energy Act of 1946, then the siting and generation of nuclear electricity might have evolved along such lines. Generation would be in the hands of the Atomic Energy Commission or a successor agency; it would be confined to relatively few remote, large centers, each eventually having as many as l0 or 20 reactors (Hanford at one time had 9 large reactors), and the electricity so generated would be distributed by private and public utilities. In short, it might have developed rather like the Bonneville Authority or the Corps of Engineers, or even the Tennessee Valley Authority.

But, in retrospect, this could never have happened here. In the first place, the utilities, perhaps stung by their lack of response to the need for rural electrifica- tion, were anxious to forestall the specter of a government-operated nuclear TVA. And this was not a phantom: Senator Albert Gore, in the early 1950s, was calling for government construction of 6 large power reactors. In any event, the Atomic Energy Act of 1954 put nuclear energy into the private sector: it opened the way for utilities to own and operate nuclear power plants, and indeed, encouraged the private sector to design and develop reactors.

But beyond these institutional imperatives, generation of electricity in remote sites is rather awkward. Utility planners site conventional generating plants close to load centers, and near cooling water. The plant's impact on the environment has, until recently, been secondary. As for possible danger to the public, for a long time this was not an issue, even though emissions from fossil plants may exacerbate lung disease. It was therefore all but inevitable that nuclear generating plants should be sited similarly to conventional plants. After all, until Three Mile Island, nuclear plants were viewed by the public, no less than by the utilities, as being like other generating plants. Their siting tended to conform to the siting practices already established by each utility. Thus Consolidated Edison, which traditionally had sited its conventional plants close to New York City, proposed building an underground nuclear plant in the heart of the city; the proposal was withdrawn when it became clear that the Atomic Energy Commission would never license a plant in such a densely populated area. And even TVA, which will soon be the largest nuclear utility, sites its nuclear plants in a pattern, set by their

coal-fired counterparts, to maximize reliability of power supply at minimum cost.

How could nuclear engineers reconcile the intrinsic danger represented by 15 billion Ci in the core of an operating reactor with the necessity of placing the reactor fairly close to load centers--i.e, to populated centers? Several strategies envolved. First, reactors were not allowed t oo close to populated areas. The AEC and now the Nuclear Regulatory Commission require exclusion zones and controlled zones around reactors. No one lives in an exclusion zone, which extends typically about ½ mile from a reactor; the utility controls access to the controlled zone. Of the ninety-odd nuclear sites now in existence, only 13 have more than 25,000 people living within a 5 mile radius, only 10 have more than 100,000 within a 10 mile radius. To this extent the original approach to safety--remote siting--has prevailed, although only a handful of commercial power plants are sited as remotely as Hanford and Savannah River.

But this is clearly not enough. Elaborate engineering devices have been developed to interpose barriers between the 15 billion Ci inside the reactor and the environment. The AEC used to speak of 3 levels of safety: extremely careful design and quality assurance that would minimize the likelihood of a mishap; a variety of safety systems, such as shut-off rods and sensors that abort a mishap before it can get started; and back-up devices such as the emergency core cooling system that would cool the reactor and prevent it from melting should the regular cooling system fail. In addition, there are at least 3 physical barriers between the radioactivity and the environment: metallic zirconium that sheaths the fuel pellets in which the radioactivity is largely generated, the thick steel pressure vessel and the pipes that carry the primary cooling water at about 2000 lb/in 2, and the large steel dome designed to withstand a pressure of 50 lb/in 2 without leaking. In the event of a core meltdown, should any of these three barriers remain un- breached, little radioactivity would reach the public.

How well had these systems worked before Three Mile Island? Pretty much as planned in American reactors. There had been a partial meltdown at the Fermi fast reactor plant near Detroit; both primary system and contaitlment held. A serious nuclear excursion killed 3 operators in a small experimental reactor in Idaho; some radioactivity escaped because this reactor had no containment shell, being located so far from people. At Browns Ferry, Alabama, a fire disabled much of the emergency core cooling system, but enough remained to prevent a core meltdown and no radioactivity leaked to the atmosphere. Outside the United States the record, at least in the early days,

Can we fix nuclear energy? 475

was not so good. The worst incident was the Wind- scale fire of 1958. A plutonium-producing reactor made of graphite caught fire; since the reactor was not surrounded by a containment vessel some 20,000Ci radioactive iodine were released, several thousand times as much as was released into the environment in the Three Mile Island accident.

A properly operating reactor and its subsystems is a generally benign source of energy. It emits no carbon dioxide or sulfur dioxide or particulates. Its radioactive emissions during routine operation are rather less than those from a coal plant of the same output. (A 1 million kW coal plant burns 2.5 million tons of coal/yr. This coal may have some 10 tons of uranium in it, and this represents about 50Ci of radioactivity in the coal ash and in gases released in the air near the coal plant.) The main hazard imposed by a properly operating nuclear reactor comes from the 200 tons of uranium that is mined to keep it fueled each year. The mine tailings contain about 1000Ci--perhaps 20 times as much radioactivity as the coal plant emits/yr. The mine tailings are usually in remote places and much of the activity decays before it is dispersed. Covering the tailings with 1 ft of earth would reduce even these emissions.

I would even put waste disposal in the category of lesser problems--largely because the high-level, potentially dangerous wastes occupy so little space (2 m3/reactor/yr) and because, after about 1000 yr, the wastes are no more hazardous than the original uranium from which the wastes were formed. And to sequester wastes for 1000 yr simply does not strike me as being beyond reason. In the Ekain caves near San Sebastian, Spain there are cave paintings that have survived very well, even though unattended for 10,000 yr. Is it too much to expect that we can sequester wastes safely for 1000yr, after which time they pose no more hazard than did the original uranium? And in Oklo, Gabon there are ancient natural chain reactors that operated for 500,000yr; many of the fission products and essentially all of the plutonium created in these remarkable phenomena remained in place for almost 2 billion yr! Despite the great public concern and political passion generated over nuclear wastes, I view them as a nuisance for which there are many solutions.

Nor do I consider proliferation, the issue on which so much discussion hinges, the main problem of nuclear energy. Most, if not all, of the world's bombs have come from reactors built to make bombs, not to produce power. Should a country want to make bombs badly enough, it can do so without troubling to build or buy a power reactor. No, I believe nuclear power is rather peripheral to the proliferation issue

and that our attempts to devise technical fixes tend to be 'allusive and sentimental' (to quote Robert Oppen- heimer) rather than 'substantive and functional.' We cannot prevent Pakistan from making bombs if it is intent on so doing, and if it places the manufacture of bombs above other national aims.

The 15 billion Ci in an operating reactor, and the possibility of its release, has long struck me as being the primary issue--the one on which nuclear energy will stand or fall. Since serious malfunction of a reactor is a matter of probability, the issue becomes more and more important as more reactors are built. If the probability of a serious malfunction, one in which significant amounts of radioactivity are released, is, say, 1 in 20,000/reactor/yr, then when there are 100 reactors operating one might expect one such accident every 200 yr; but if there are eventually 10,000 reactors throughout the world--as conceivably could happen were nuclear energy to become the world's primary energy source, then, unless the probability of an accident for each reactor is reduced, one could expect one such accident every 2 yr. Although the fre- quency of accident/reactor or/utility company remains very small, I do not believe the public would maintain confidence in an energy system that caused even relatively modest radioactive contamination every 2 yr. Nor does it make much difference where accidents happen: TV converts an accident anywhere into an accident everywhere. For nuclear energy to grow, the probability of such accidents will have to diminish, the public will have to be prepared to cope with the radiation risk such accidents might entail and the media will have to deal with them in the same way it deals with other industrial accidents that have comparable impact on health.

Bo Lindstrom, the Swedish aeronautical engineer, pointed out some 20yr ago that air travel faced exactly this dilemma. He argued that if air travel continued to expand, and the accident rate/passenger mile held constant, then by around 2000 there would be several serious accidents every day somewhere in the world. For the individual, air travel would remain as good a risk in 2000 as it was in 1960. But he argued that the public's confidence in air travel would collapse, and that this mode of transportation could not survive such a loss of confidence. Air transport has, or course, become much safer/passenger mile than it was at the time Lindstrom made his observations; and, I suppose, the public has become somewhat inured to air crashes. Both of these were necessary for air transport to survive.

What are the actual probabilities of malfunction in reactors? Before Three Mile Island, all of us in the nuclear enterprise were fairly comfortable with the

476 A.M. WEINBERG

estimates made by Professor Rasmussen in his famous study on probabilities and consequences of reactor accidents. He estimated that for a light water reactor the probability of a core meltdown that would release at least a few thousand Ci of radioactivity was 1 in 20,000/reactor/yr. Most of these incidents would not cause physical damage to the public. A few would, and a very few, estimated at 1 in 1 billion reactor yr, might be a major catastrophe--3300 immediate radiation deaths, 45,000 extra cancers and $17 × 109 in property damage. Rasmussen himself has set the uncertainty in his estimate of probability at about 10 either way (although the report puts the uncertainty at half this) and the estimate of consequences perhaps at 3; i.e. the probability of an accident causing significant property damage might be I in 2000/reactor/yr and the probability of the very worst accident 1 in 100,000,000 reactor yr. The recent Lewis committee report has set even greater uncertainties on the probabilities, though it generally praised Rasmussen's meth- odology.

Before Three Mile Island I was comfortable with the record up to that point. The world's light water reactors had amassed 500 reactor yr without a meltdown; if one added the nuclear navy's record, one could roughly triple this--no meltdowns in about 1500 reactor yr. Rasmussen's upper limit of meltdown--about 1 in 2000 reactor yr--was close to being vindicated.

Then came Three Mile Island. Though I write this before all the returns are in, I believe it is fair to say that Three Mile Island suggests that the probability of accidents that released a few thousand Ci may have been underestimated--not so much because the engineering was deficient, but because of human error. Shutting off the high pressure injection pumps in the emergency system, followed by various other malfunctions and operator errors, was not, as far as I can deduce, exactly contemplated in Rasmussen. Yet despite this and other deficiencies, the containment for the most part held. The iodine released in the incident was a few Ci; the maximum total whole body exposure to any member of the public is probably less than the 100 mrem we used to accept every day as the allowable occupational dose in the early days of the atomic energy project. No member of the public has suffered bodily harm from Three Mile Island.

Chauncey Starr, Vice-President of the Electric Power Research Institute, has deduced from the Ras- mussen study that an incident like Three Mile Island had a 50:50 chance of occurring in 400 reactor yr. Can nuclear energy survive if Three Mile Island incidents have a 50~ probability of occurring every 400 reactor yr? I do not think it can. It is not that people

will actually be hurt; it is that people will be scared out of their wits and will lose confidence in nuclear energy. The drama of the hydrogen bubble in the pressure vessel has rarely been matched on television. It was like seeing a new episode of the China Syn- drome every night for a week without having to pay the admission charge. And with 200 reactors on line by the 1990s, one might expect a new, gripping serial once every few years.

I should think that long before this the utilities themselves would have done with nuclear energy. The public often forgets that in an incident like Three Mile Island, it is the utility operating the reactor that suffers the most. Who will pay to restore the reactor to working order, who will pay the many claims that arise out of the incident? To be sure, the Price-Ander- son Act limits public liability to $560 million; beyond that all reactor operators are retroactively assessed to help pay for claims arising from an accident. But the general hassle this entails, now that much of the fuel has been destroyed and the reactor building conta- minated, is something that won't go away easily.

Can the nuclear enterprise be redesigned so as to make it acceptable? Can the probability of an incident like Three Mile Island be reduced very significantly; and is it likely that the public reaction to future Three Mile Island will be more commensurate with the actual public damage rather than with its perception of the potential hazard? It is my belief that such changes are needed if nuclear energ~ is to survive as a long-term energy system.

Before offering suggestions for a redesigned enterprise, I must remark that Three Mile Island could be the salvation of nuclear energy. Before Three Mile Island the possibility of a core meltdown and its potential consequences were, by and large, the private knowledge of the nuclear and the anti-nuclear com- munities. Today, every reader of the newspaper or watcher of the television knows about cooling systems and their malfunction. Best of all, the utilities that operate nuclear plants are acutely aware that nuclear reactors are not simple replacements for coal boilers --that the responsibility entailed in operating a nuclear power plant is far greater than that entailed in operating a fossil-fueled plant. (I recall asking, in 1970, the executive vice-president of a utility about to build a nuclear power plant whether his reactors could possibly melt down. 'No' he said, Upon my asking why he was so sure he answered, 'Because my board of directors would not have approved a plant if there were any possibility of its melting down.')

I can identify six characteristics of what I would consider to be an acceptable nuclear energy system: physical isolation, technical fixes, separation of gene-


ration and distribution, professionalization of the nuclear cadre, heightened security and, perhaps most difficult, public education about the hazards of radiation. I discuss them in turn.

Physical isolation

It is too late to return to the original siting policy that confined nuclear activities to very remote places like Hanford and Idaho. But I believe we can achieve a good deal by confining the enterprise, forever, to those of the existing sites that have few people near them. Of the nearly 100 sites now operating or being built, only 13 have as many as 25,000 people living within a 5mile radius, only 10 with as many as I00,000 within a 10 mile radius. We therefore already have 80 or so sites with so few people nearby that evacuation in an emergency ought to be relatively easy. Perhaps more important, everyone living within 5 miles could be educated about radiation--perhaps each household should be equipped with a radiation detector, much like smoke detectors. Moreover, the size of the population even in the future within 5 miles ought to be limited. This could be accomplished if the area (75 miles 2) around each site were zoned to prevent further urbanization. If we eventually have 100 nuclear sites in all, this would amount to committing 7500 miles 2 to the enterprise into perpetuity; but only a small part of this, perhaps 200 miles 2, would be exclusively reserved for nuclear operations.

Since the number of sites is limited, their generating capacity will increase if the nuclear enterprise grows. Eventually, each site may have as many as I0 or so reactors. To my mind, such clustering ought to bring in its wake other improvements. Most important, large sites are likely to have more able people than small sites. There will develop an organizational memory--small mishaps on Unit 2 5 yr ago are not likely to be repeated on Unit 4 today because the operators of Unit 2 ride in the same car pool as those on Unit 4. I speak of this from my experience at Oak Ridge--a large, powerful nuclear center where, to be sure, accidents have happened, but which has always had the logistic strength and organizational memory to contain the damage. I believe the same can be said for Hanford and Savannah River.

The sites ought also to be invested with an impu- tation of permanence--possibly in the same sense that dams are so regarded. For if one concedes that the sites are permanent, then one can simply leave the voluminous low-level radioactive wastes and old reactors in place until their radioactivity has largely decayed. We estimate that after 100 yr or so the bulk of the low-level wastes as well as the old reactors have become innocuous. Dismantling old reactors after

that should be relatively easy. During the decay per- iod the old reactor buildings might be used to store the other wastes. In any case, their surveillance should pose little difficulty since they are part of an ongoing, active nuclear complex.

Technical fixes

As Three Mile Island has shown, the nuclear enterprise is still learning. Is it learning fast enough? Are the incremental backfittings prompted by malfunctions sufficient to reduce the probability of failure faster than the number of reactors grow? Surely Three Mile Island will lead to corrections of faults on existing pressurized water reactors that were revealed by the incident. It will also lead to even tougher regula- tions than are now in force.

Beyond this is a much more fundamental question: are there reactor types, entirely different from the pressurized water system (PWR), that are inherently safer than the PWR? After all, the PWR was conceived as a compact reactor capable of being stuffed into a cramped submarine; its blossoming into huge central power plants is still a source of wonderment to the original designers of the compact submarine reactors. The British, without quite saying so, suggest that their large graphite reactors cooled with gas are less prone to the China Syndrome than is the PWR.

The Russians continue to build large graphite moderated water cooled reactors, as well as reactors like Three Mile Island; the Canadians build heavy water moderated systems. I myself was a long-stand- ing exponent of a completely different reactor type which used fuel that was already in the molten state-- the molten salt reactor. Is it impossible to return to square one and try to design a reactor that is more immune from the China Syndrome? Does the technical community have the nerve, and do the other actors--utilities, government and manufacturers-- have the money and will to design and commercialize a completely new system, especially since we do not know whether a totally 'safe' reactor can be designed? To develop a complete reactor system costs about $10 billion; yet I see one possibility for launching such an enterprise--to develop, if possible, an inherently safe reactor that is also a breeder. (A breeder is a reactor that uses only 1/60th as much uranium as do the current types. The breeder is therefore an all but inexhaustible energy source--like fusion or the sun. It has always been a sort of holy grail for nuclear energy. The molten salt breeder may be a contender here-- but work on this reactor was discontinued several years ago. My personal view is that this was a gross error in our nuclear development. Perhaps in this reexamination the molten salt breeder will be looked

478 A.M. WEINBERG

at much more seriously than it has been. Perhaps other breeder systems, such as those based on gas cooling, will be conceived that are inherently incap- able of melting down. But these are questions that can hardly be raised seriously as yet. I mention 'them as a possibility, since the Administration is committed to an advanced breeder development. Could this com- mitment be to a breeder that is practically immune from China Syndromes?)

Separation of nuclear generation and distribution

The nuclear system requires a powerful organiza- tion if it is to be operated properly. We have in this country about 200 generating utilities. Nuclear electricity does not lend itself very well to fragmented, small operators. A 1 million kW power plant often represents a large fraction of the total output of a smaller utility. If a utility has but a single reactor it seems to me less likely that it can maintain the organizational memory or even the financial strength necessary to operate a device that conceivably imposes such heavy demands as does a nuclear reactor.

If the siting policy that I espouse evolves, it seems natural that the large, clustered sites will be operated by powerful organizations whose main job is operating nuclear facilities. Presumably most of the sites would serve more than a single utility; the entity that operates a site might therefore, in most cases, be a consortium of utilities. Thus, one could envisage the gradual separation of generation from distribution of nuclear electricity--the former being in the hands of powerful organizations who do nothing but generate nuclear electricity, the latter consisting of the existing utilities. (In the early 1950s Charles Thomas, Presi- dent of Monsanto Chemical Company, suggested that the generation of steam by nuclear energy was more properly done by the technically sophisticated, research-minded chemical industry than by the utilities. He proposed to sell nuclear steam to a utility which would then generate electricity with it. The idea foundered largely because it was so incongruent with the existing structure of the electrical utility business.)

It is a delicate, and not very clear, question as to whether the operating consortia should be public or private---whether one gets better operation, and safer operation, from private organizations policed by the Nuclear Regulatory Commission or by a public Nuc- lear Energy Authority. There is an inherent tension between safe as possible and cheap as possible. The conflict manifests itself when a pilot decides to cancel a flight because the weather is bad, even though this costs his company money, or when an auto designer decides how much protection the gas tank requires. And I suppose many would argue that a public opera-

tor is more likely to weight his decisions on the side of safety. But public authorities seem to me to be harder to regulate than are private ones.

It is therefore not clear to me that public operation is the right answer. After Three Mile Island, every nuclear utility must realize that its very existence may depend on avoiding incidents of this sort. This must be powerful medicine for clearing one's brain of possible confusion as to which comes first--safety or con- tinuity of electricity supply. I would therefore hope that the utilities themselves seek to set up the powerful generating entities I deem so necessary.

Professionalization of the nuclear cadre

The pilot of a transatlantic 747 is paid about $100,000/yr, perhaps 50~ of what the president of his company gets. The superintendent of a nuclear plant gets about 20~ of what his president gets. In both cases, I would claim that the pilot and the operator bear a heavier burden of direct responsibility than do their respective presidents. In the case of the pilots, this is more or less recognized in their pay; in the case of the plant superintendent, it is not. Why?

I believe the answer is to be found again in the mistaken belief that a nuclear plant was just another generating station: the pay scales and the whole conception of training and expertise tended to be strongly influenced by this perception. Moreover, a utility would find it awkward to pay operators of one kind of plant much more than they pay operators of another.

Yet it is clear that the responsibility borne by the nuclear operator is so great that he and his staff must be extra special--like airline pilots, air controllers or the coast guard. They must constitute a cadre with tradition, with competence, with confidence, with training that goes beyond what we already have. Is it possible to get people of such quality for jobs that are essentially very boring--99.9~o or more of the time a nuclear reactor operates perfectly, and there is little for the operators or supervisors to do. But this is the same problem faced, say, by the pilots on the Eastern Airline shuttle, or even by the anesthesiologist in a routine operation. We deal with this ennui with money, with status in the community and with shorter work schedules; this is the very least we can do with nuclear operators.

Perhaps this is the primary reason I have so strongly urged center siting and separate generating entities: both would be more conducive to the creation of the professionalized cadre necessary to keep the nuclear enterprise out of trouble. At the center, there will be many people to choose from when vacan- cies arise, and there will be a general ambience of


expertise. A generating entity can pay its employees salaries that are not bound by the tradition set by the earlier history of coal-fired utilities.

Heavy security

The nuclear enterprise will always demand far greater security than the fossil-fueled enterprise. Ter- rorism and sabotage can incapacitate a fossil plant; they can, though with some difficulty, lead to serious accidents in nuclear plants. So this is another reason why cluster siting is important. It is easier to guard 10 reactors in a single site than 10 separate sites each with a single reactor.

Does this represent an unwarranted infrigement of our freedom, or make nuclear energy incompatible with democracy? Does the security at SAC or nuclear submarine bases constitute a threat to our freedom? If one concedes that we now have both freedom for the society and security at a relatively few military bases, then I see no reason why imposition of such stringent security at a few large reactor centers should be im- compatible with a free society.

Public education about the hazard of radiation

But none of these measures will ensure the survival of nuclear energy unless the public and the media come to accept the risk of radiation as no different than the risk from other noxious agents that are products of our technology. Why, after all, did Three Mile Island create such extreme concern, especially since not one member of the public has been harmed by it, or, for that matter, by the operation of any other nuclear reactor? It is ironic that the very same newspapers that devoted many columns to Three Mile Island that first day devoted one paragraph to an air crash in Quebec that killed 17 people. Why was Three Mile Island the biggest story of the year when the collapse of the Grand Teton Dam, or even the collision of the jumbo jets in Tenerife, vanished from the front pages in a few days?

I see several reasons for this seeming double standard. First, the potential for a disaster was there, though exactly how close we were will probably have to await the outcome of the current investigations. Since its dimensions could not be gauged and the whole situation was so completely novel, there were the ingredients of high media drama. The fear of possible radiation-induced death goes deep. Radiation is mysterious: it cannot be sensed, yet it can kill you. It is as though the widespread use of electricity had taken place in this age of nightly television news rather than a century ago: would not our hypochon- dria over electricity be far greater than it now is?

But secondly, the estimate of the hazard of radiation is clouded by bitter scientific dispute. In particu- lar, there is the strongest kind of disagreement among scientists as to the effect of very low levels of radiation, even levels as low as our natural radiation background. Most of the estimated delayed cancer deaths associated with so-called hypothetical accidents are supposed to be caused by exposures well below the occupational limits. If one assttmes that any extra radiation, however small, causes cancer, then if mil- lions of people are exposed some extra cancers will result. Governor Thornburgh's decision to evacuate pregnant women was, I presume, based on the retros- pective study of Stewart and Kneale who claimed to have found a significant correlation between child- hood cancer and fetal X-irradiation. But this finding is contradicted by the experience at Nagasaki: slight irradiation of fetuses there did not increase the risk of cancer. The matter is, to say the least, moot. I would hope that eventually the risk of low level radiation will be sorted out: if, as I believe, low level radiation is nowhere near as dangerous as, for example, Tom Brokaw of NBC seems to think, then the public's and, perhaps more important, the media's reaction to the possibility of such irradiation may be far more res- trained.

Indeed, the whole question of low level radiation hazard is so critical to the acceptance of nuclear energy that I would judge this to be a leading, if not the leading, scientific issue underlying the nuclear controversy. Unfortunately, because the effects (if any) in any case are so rare because the exposures are so small, the issue may be beyond the proficiency of science to decipher--it may lie in the domain of 'trans-science,' not science. Fortunately, we do have a standard--natural background--to compare addi- tional exposures with. At Three Mile Island the total dose to the population was about 1% of natural background--a level where no effects can be seen.

Is all this a 'cynical capstone' to my nuclear advo- cacy, as Robert Jungk has described my claim that nuclear energy will have easier sledding because people will come to accept the hazard of radioactivity in the same spirit as they have accepted the hazard of electricity or of flying at 600mph? It surely is the capstone: without such acceptance nuclear energy cannot survive. I would describe it as realistic rather than cynical, but others can judge this better than I.

CAN WE GO FROM HERE TO THERE'?

If one accepts this design for an acceptable nuclear future, are there credible paths from our present system to the proposed system? The first step, it seems to

A N . I . 6 ¢ ] - I0 I~

480 A.M. WEINBERO

me, is to halt the proliferation of nuclear sites. Insofar as possible, confine the entire enterprise to those of the existing sites that have few people near them. We cannot in this way return to the original conception of safety by remoteness, as at Hanford (some 80 of the 100 largest cities in the United States are within 50 miles of a nuclear power plant), but we can refrain from starting new construction of reactors on sites that, say, have more than 25,000 or even 10,000 people with 5 miles, more than 50,000 or 100,000 within 10miles. We can greatly strengthen the emergency training we give to people in the vicinity; until Three Mile Island, emergency training was deemed to be too unsettling, now we can do no less than accept it as one of the costs of nuclear energy.

(I cannot refrain from comparing the ostrich-like posture we had with respect to nuclear accidents with our posture with respect to civil defense against nuclear attack. The probability of a nuclear attack can hardly be less than the probability of a serious meltdown. Is it too outlandish to suggest that more serious civil defense be part of the U.S. SALT II pack- age? This could give us the opportunity to strengthen civil defense against uncontained meltdowns. Or con- versely, could emergency plans to deal with meltdowns, which the public would now concede to be sensible, be the nucleus for a serious civil defense system against nuclear weapons?)

Will the utilities regroup into a relatively few nuclear generating consortia and a much larger number of distributing companies, the latter continuing to generate by conventional means? I would hope this would happen more or less naturally as a by-product of the restricted siting policy. But such regrouping will require imaginative leadership as well as a sense, on the part of the utilities themselves, that structural changes are needed.

Finally, we come to the public's attitude. None of this can happen, nor can the enterprise survive, unless the public (and the public opinion-molders) can regain confidence in nuclear energy. Unless this happens, the Vietnamization of nuclear power will continue; the nuclear age will grind to a halt as the present reactors run their course and we shall have to revert to the energy strategies that were available to us before fission was discovered.

THE ALTERNATIVES

What are the alternatives to fission? Aside from conservation, there are only four--geothermal, fusion, fossil and the various forms of solar energy. Geother- mal energy is relatively small-- the total geothermal energy produced in the earth is about 1/20,000 the

energy the sun sends to us. This is not far from the total energy man now uses. Thus, if we are to contemplate a world that has many more people and then uses, say, three times as much energy as we now use, geothermal can hardly help.

As for fusion, despite the optimism that prevails in the fusion community, it seems to me that the possibility still remains just tha t - -a possibility. The fuel (lithium in present schemes, heavy hydrogen in more advanced schemes) is all but inexhaustible. But the engineering remains formidable, even after scientific feasibility has been demonstrated. Moreover, fusion is not devoid of radioactivity. To be sure, there is 100 times less in a fusion reactor than in a fission reactor (100million Ci of radioactive tritium), but, as Three Mile Island suggests, if fusion is to be acceptable it too will require a public that understands relative hazards of radioactivity. Thus, my view about fusion is agnostic--let 's work on it, but let's not count on it.

Fossil fuel is, of course, what we shall turn to in increasing amounts whether or not we have fission; but if we had a moratorium on new fission plants beginning in 1985 we might, in the United States, have to burn about 1 billion tons more coal by 2000 than if we had no such moratorium. And as for oil, the political pressures might become quite intolerable should our share of the world's oil supply increase drastically.

Nor is fossil fuel a benign source of energy. Even a properly operating coal plant emits fumes that are believed to be noxious. The dangers from burning fossil fuels are distributed and undramatic--black lung disease among miners, bronchial troubles down- wind of a coal plant. By contrast, the dangers of the nuclear plant are localized and dramatic--the China Syndrome---even though, as Three Mile Island has shown, a nuclear plant can suffer an extraordinary amount of damage without anyone being hurt. But fossil fuels do pose a large-scale threat that, in being worldwide, is comparable to the threat of proliferation. I refer to the accumulation of carbon dioxide in the atmosphere at a rate of about ½~o/yr of what is already there. Most climatologists (though not all) believe that doubling the CO2 may increase the aver- age surface temperature by about 2°C, and would diminish the equatoridl-polar temperature gradient (which drives the wind system)by about 10°C. Not all the ensuing changes need be bad, but the doubling of CO2 in the atmosphere, which could happen by, say, 2050, represents an unprecedented climatological experiment by man. It might change agricultural pat- terns, it might cause the sea to rise, it might turn deserts into grasslands and grasslands into deserts. If this is a real possibility (and I must confess that the


matter is still under debate), then is the continued burning of fossil fuel a responsible course, even if fossil fuel were inexhaustible (which it is not)?

This then leaves solar energy (including hydro, wind, waves, biomass, ocean thermal gradients as well as direct solar)--immense, environmentally benign, the darling of the people (no one is against solar energy}---but intermittent and, insofar as one can tell, expensive. If what we contemplate is an all solar wor ld--not one in which small household solar water heaters are backed up by electricity from the friendly uti l i ty--then we must come to terms with the intermittency of solar energy. Either we adjust our lives to an intermittent sun or we arrange for storage--perhaps using auxiliary engines operated on alcohol, electric batteries or perhaps hydrogen generated photoelectrically. To overcome intermittency seems to me to be expensive, though how expensive I cannot say. What seems clear is that an all solar society is almost surely a low energy society, and one in which energy will be a good deal more costly than it is from the sources we now use.

To my mind, the only ultimate alternative, or perhaps adjunct, to an all solar society is the one based on fission, at least if one concedes that fossil fuels are limited or that CO2 must be taken seriously, and that fusion will forever evade us. But such a fission future might involve several thousand reactors, and one must then contemplate the Bo Lindstrom problem I alluded to earlier: even though the probability/reactor of a serious accident is small, when the system becomes large the number of accidents may become too frequent for the public to tolerate. Thus, if a solar society could be made to work by all means let us work hard to achieve it. I therefore favor pushing solar technology as hard as we can. But let us not mislead ourselves. Solar cannot be ready to take over very much of the load for a long time; and despite Amory Lovins' shrill cries to the contrary, a solar society will not be the Utopia he perceives it to be--unless some very major improvements in energy storage and photoelectric conversion are achieved.

But suppose we do not achieve these technological breakthroughs--can we put a price on solar energy at which we would prefer it to nuclear energy because nuclear is faulted by its radioactivity and its plutonium? There are many who would say abolish nuclear in favor of solar, whatever the cost. This I cannot view as a rational response. But neither can I say how much extra one should pay for solar to avoid the disadvantages of nuclear. And it is not impossible that the price one must pay for an acceptable nuclear sys- t e m - w i t h its better technology, its higher-paid cadre

and its tighter security---conceivably could price nuclear out of the market.

THE FAUSTIAN BARGAIN

About 8 yr have passed since I referred to nuclear energy as a Faustian bargain. I have since been cor- rected both by nuclear advocates (who prefer Pro- metheus to Faust) and by exegesists (who say that Goethe's Faust made a wager, not a bargain). Never- theless, what I meant was clear: nuclear energy, that miraculous and quite unsuspected source of energy, demands expertise of man, attention to detail and social stability that is unprecedented. In return, man has, in the breeder, an inexhaustible energy source.

Three Mile Island has undoubtedly turned many away from nuclear energy, has reinforced their belief that nuclear energy is simply too hazardous. Three Mile Island for me has a rather different significance. I have often said that Goethe's Faust was redeemed:

'Who e'er aspiring, struggles on, For him there is salvation,'

and that man in his striving will finally master this complex and unforgiving technology.

My anti-nuclear colleagues retort that this is foolish technological optimism--that man is imperfect and that anything that can go wrong will go wrong. But man is also ingenious, and the history of the two worst American reactor accidents--Browns Ferry and Three Mile I s l a n d ~ e m o n s t r a t e this. In both cases I think it is fair to say that the accident was precipi- tated by human error-- the lighted candle at Browns Ferry and the shut off pumps at Three Mile Island. In both cases the operators used their ingenuity to get out of a nasty s i tuat ion--and in neither case was anyone harmed by excess radioactivity. It is this denial of human ingenuity while accepting human fal- libility which is the main weakness of the more stri- dent nuclear opponents--those who wish to close down the entire enterprise.

But I am not an uncritical advocate of nuclear energy. I believe the enterprise needs fixing if it is to survive; my own prescriptions for such fixes I have outlined above. Other prescriptions will surely come in the wake of Three Mile Island. Nor do I think the nuclear enterprise can wait too long before it demon- strates to the public that it recognizes changes must be made if the public's confidence is to be maintained. With the Presidential campaign already started, I should think the future of nuclear energy will be a prime political issue. I hope those who believe that nuclear energy is too important to lightly cast aside

482 A.M. WEINBERG

will do more than simply restate their faith in the technology. They must come up with positive, new and convincing initiatives that prove to the public that the lessons of Three Mile Island have been

learned. To do less will commit us to energy options whose difficulties, though not as stark as those of nuclear energy, could in the long run be even more serious.

Documents

Can we fix nuclear energy?