Interdisciplinary Science Reviews 23:193-203;1998




Bernard L. Cohen

University of Pittsburgh





As an initial perspective, it is interesting to compare nuclear waste with the analogous waste from a single large coal-burning power plant. The largest component of the coal burning waste is carbon dioxide gas, produced at a rate of 500 pounds every second, 15 tons every minute. It is not a particularly dangerous gas, but it is the principal contributor to the "greenhouse effect" that is raising so much international concern and threatening the health of our world economy.. Among the other wastes from coal burning, probably foremost is sulfur dioxide, the principal cause of acid rain and perhaps the main source of air pollution's health effects, released at a rate of a ton every 5 minutes. Then there are nitrogen oxides, the second leading cause of acid rain and perhaps also of air pollution. Nitrogen oxides are best known as the principal pollutant from automobiles and are the reason why cars need expensive pollution control equipment which requires them to use lead-free gasoline; a single large coal-burning plant emits as much nitrogen oxide as 200,000 automobiles. The third major coal burning waste is particulates, including smoke, another important culprit in the negative health effects of air pollution. Particulates are released at a rate of several pounds per second. And next comes the ash, the solid material produced at a rate of 1,000 pounds per minute, which is left behind to cause serious environmental problems and long-term damage to our health.            Coal-burning plants also emit thousands of different organic compounds, many of which are known carcinogens. Each plant releases enough of these compounds to cause two for three cancer deaths per year. And then there are heavy metals like lead, cadmium, and many others that are known or suspected of causing cancer; plus a myriad of other health impacts. Finally there is uranium, thorium, and radium, radioactive wastes released from coal burning that serve as a source of radon gas. We will show that the impact of this radioactive radon gas from coal burning on the public's health far exceeds the effects of all the radioactive waste released from nuclear plants.

 The waste produced from a nuclear plant generating the same amount of electricity is different from coal-burning wastes in two very spectacular ways. The first is in the quantities involved: the nuclear waste is 5 million times smaller by weight and billions of times smaller by volume. The nuclear waste from 1 year of operation weighs about 1.5 tons and would occupy a volume of half a cubic yard, which means that it would fit under an ordinary card table with room to spare. Since the quantity is so small, it can be handled with a care and sophistication that is completely out of the question for the millions of tons of waste spewed out annually from our analogous coal-burning plant.

The second pronounced difference is that the nuclear wastes are radioactive, providing a health threat by the radiation they emit, whereas the principal dangers to health from coal wastes arise from their chemical activity. This does not mean that the nuclear wastes are more hazardous; on nearly any comparison basis the opposite is true.     For example, if all the air pollution emitted from a coal plant in one day were inhaled by people, 1.5 million people could die from it, which is 10 times the number that could be killed by ingesting or inhaling the waste produced in one day by a nuclear plant.

This is obviously an unrealistic comparison since there is no way in which all of either waste could get into people. A more realistic comparison might be on the basis of simple; cheap, and easy disposal techniques. For coal burning this would be to use no air pollution control measures and simply release the wastes without inhibition. This is not much worse than what we are doing now since the smoke abatement techniques, which are the principal pollution control on most coal-burning plants, contribute little to health protection. A typical estimate is that air pollution causes at least 25 deaths per year from each plant. Some other comparably serious consequences will be considered later.

For nuclear waste; a simple, quick, and easy disposal method would be to convert the waste into a glass--a technology that is well in hand--and simply drop it into the ocean at random locations. No one can claim that we don't know how to do that! With this disposal, the waste produced by one power plant in one year would eventually cause an average total of 0.5 fatalities, spread out over many millions of years, by contaminating seafood. Incidentally, this disposal technique would do no harm to ocean ecology. In fact, if all the world's electricity were produced by nuclear power and all the waste generated for the next hundred years were dumped in the ocean, the radiation dose to sea animals would never be increased by as much as one percent above its present level from natural radioactivity.

We thus see that if we compare the nuclear and coal wastes on the basis of cheap, simple, and easy disposal techniques, the coal wastes are (25 / 0.6 =) 40 times more harmful to human health than nuclear wastes. This treatment ignores some of the long-term health effects of coal burning to be discussed later.

Another, and very different, way of comparing the dangers of nuclear and coal waste is on the basis of how much they are changing our exposures to toxic agents. The typical level of sulfur dioxide in the air of American cities is 10 times higher than natural levels, and the same is true for the principal nitrogen oxides. For cancer ­causing chemicals the ratio is much higher. These are matters that might be of considerable concern in view of the fact that we really do not understand the health effects of these agents very well. For radioactive waste from a flourishing nuclear industry, on the other hand, radiation exposures would be increased by only a tiny fraction of 1 % above natural levels.

Another basis for comparison between the wastes from nuclear and coal burning is the "margin of safety," or how close average exposure is to the point where there is direct evidence for harm to human health. There have been several air pollution episodes in which there were hundreds of excess deaths, with sulfur dioxide levels about 1,000 micrograms per cubic meter, whereas levels around 100

micrograms per cubic meter, just 10 times less, are quite common in large cities. For radiation. on the other hand, exposures expected from hypothetical problems with nuclear waste are in the range of 1 millirem or less, tens of thousands of times below those for which there is direct evidence for harm to human health.

All of these comparisons have been intended as an introduction to the subject. Now, let us look into the problem in some detail.




Probably the most frequent question about nuclear power is, "What are we going to do with the radioactive waste?" The answer is very simple- we are going to convert it into rocks and put it in the natural habitat for rocks, deep underground. T he next question is usually, "How do you know that will be safe?" Answering that question is our next topic for consideration.

This discussion will be in terms of the high-level waste produced by one large power plant (1,000,000 KW)  in one year. The high-level waste from such a plant is contained inside the fuel.     If we are to have reprocessing. the 35 tons of spent fuel will be shipped to a chemical reprocessing plant where it will be dissolved in acid and put through chemical presses to remove 99.5% of the uranium and plutonium that are valuable as fuels for future use. The residue--1.5 tons of high-level waste--will then be incorporated into a glass in the form of perhaps 30 cylinders, each about 12 inches in diameter and 10 feet long. weighing about 1.000 pounds. There is a great deal of research on materials that may be superior to glass as a waste form, but far more is now known about glass, a material whose fabrication is simple and well developed. There is no evidence that it is not satisfactory. Many natural rocks are glasses and they are as durable as other rocks.

This 15 tons of waste glass, roughly one truckload, will then be shipped to a repository, where it will be permanently emplaced deep underground. The estimated cost of handling and storing our one truckload of waste is $5 million, which

corresponds to 0.07 cents per kilowatt-hour. about I% of the cost of electricity to the consumer.

We often I hear citizens saying "We don't want a waste dumping ground in our area." This conjures up a picture of dump trucks driving up and tilting back to allow their loads of waste to slide down into a hole in the ground. Clearly it doesn't cost $5 million for each truckload to do this, especially on a mass-handling basis. What is planned, rather, is a carefully researched and elaborately engineered emplacement deep underground in an elaborately designed mined cavity.

It may be of incidental interest to point out that the present $5 million price tag on storing one year's waste from one plant is several hundred times higher than the cost of the original plan as formulated in the 1960s. Those conversant with the earlier plan are still convinced that it provided adequate safety. although the added expenditures do, of course, contribute further safety. This tremendous escalation in the scope of the project is a tribute to the power of public hysteria in a democracy like ours, whether or not that hysteria is justified. It is also an application of Parkinson's Law--it

was clearly acceptable to devote 1% of the cost of electricity to disposal of the waste, so the money was “available”, and being available, the government has found ways to use it.

There have been many safety analyses of waste repositories, and alt of them agree that the principal hazard is that somehow the waste will be contacted by groundwater, dissolved, and carried by the groundwater into wells, rivers, and soil. This would cause contamination of drinking water supplies or, through pick-up by plant roots, contamination of food. It could therebv get into human stomachs.

The chance of this waste becoming suspended in air as a dust and inhaled by humans is very much smaller because groundwater rarely reaches surface soil; moreover, we inhale only about 0.001 grams of dust per day {and 95% of this is filtered out by hairs in the nose, pharynx, trachea, and bronchi and removed by mucous flow), whereas we eat over 1,000 grams of food per day. It is shown in Appendix A that an atom in the top inch of U.S. soil has 1 chance in 10 billion per year of being suspended as a dust and inhaled by a person, whereas an atom in a river, into which groundwater normally would bring the dissolved waste, has 1 chance in 10 thousand of entering a human stomach. Thus, an atom of waste. has a much better chance of becoming absorbed into the body with food or water than with inhaled dust.

External irradiation by radioactive materials in the ground is also a much lesser problem. Rock and soil are excellent shielding materials; while the waste remains buried. not a single particle of radiation from it can ever reach the surface., Compare this with the 15,000 particles of radiation from natural sources that strike each of us every second. If radioactivity is released by groundwater, this shielding is still effective as long as it remains underground or dissolved in river water. If the radioactivity does somehow become deposited on the ground surface, it will soon be washed away or into the ground by rain. The great majority of the radiation is absorbed by building materials, clothing, or even the air. Thus, there would be relatively little exposure to the human body due to radiation originating from materials on the ground.

Quantitative calculations confirm that food and water intake is the most

important mode of exposure from radioactive waste, a position agreed upon in all safety analyses. Hence we will limit our consideration to that pathway. In order to estimate the hazard, we therefore start by determining the cancer risk from eating or drinking the various radioactive materials in the waste.

We begin by estimating the number of cancer deaths that would be caused if all of the waste from one year of operation of a large nuclear plant were fed to people. The method for calculating this, outlined in Appendix B. is a straightforward

application of standard health physics methodology. Here we only discuss the results. If this were done shortly after burial; we would expect 50 million fatal cancers, whereas if it were done after 1,000 years there would be 300,000, and if after a million years there would be 2,000. Note that we did not specify how many people are involved because, as long as there are enough people involved, that does not matter, according to the linear no-threshold theory conventionally utilized in these calculations.

If nuclear power was used to the fullest practical extent in the United States, we

would need about 300 power plants of the type now in use. The waste produced each year would then be enough to kill (300 x 50 million = ) over 10 billion people. This may sound like an impressive number, but we produce enough chlorine gas each year to kill 400 trillion people, enough phosgene to kill 20 trillion, enough ammonia and hydrogen cyanide to kill 6 trillion with each, enough barium to kill 100 billion. and enough arsenic trioxide to kill 10 billion. All of these numbers are calculated, as for the radioactive waste, on the assumption that all of it gets into people. Hopefully, these comparisons dissolve the fear that, in generating nuclear electricity, we are producing unprecedented quantities of toxic materials.

Our calculation can also be used to estimate how much of the waste glass, converted into digestible form, would have a good chance of killing a person who eats it---this may be called a "lethal dose."` the results are as follows:

shortly after burial: 0.01 oz.

after 100 years: 0.1 oz.

after 600 years: 1 oz.

after 20,000 years: 1 lb.

Lethal doses for some common chemicals are as follows:

selenium compounds: 0.01 oz.

potassium cyanide: 0.02 oz.

arsenic trioxide: 0. I            oz.

copper: 0.7 oz.

Note that the nuclear waste becomes less toxic with time because radioactive materials decay, leaving a harmless residue, but the chemicals listed retain their toxicity forever.

By comparing these two lists we see hhat radioactive waste is not infinitely toxic, and in fact it is no more toxic than some chemicals in common use. Arsenic trioxide, for example, used as an herbicide and insecticide, is scattered on the ground in regions where food is grown, or sprayed on fruits and vegetables. It also occurs as a natural mineral in the ground, as do other poisonous chemicals for which a lethal dose is less than one ounce.

Since the waste loses 99% of its toxicity after 600 years, it is often said that our principal concern should be limited to the short term, the first few hundred years. Some people panic over the requirement of security even for hundreds of years. They point

out that very few of the structures we build can be counted on to last that long, and that our political, economic, and social structures may be completely revolutionized within that time period. The fallacy in that reasoning is that it refers to our environment here on the surface of the earth. where it is certainly true that most things don't last for hundreds of years. However, if you were a rock 2,000 feet below the surface, you would find the environment to be very different. If all the rocks under the United States more than 1.000 feet deep were to have a newspaper, it couldn't come out more than once in a million years, because there would be no news to report. Rocks at that depth typically last many hundreds of millions of years without anything eventful occurring. They may on rare occasions be shaken around or even cracked by earthquakes or other diastrophic events, but this doesn't change their position or the chemical interaction with their surroundings.

 One way to comprehend the very long term toxicity in the waste is to compare it with the natural radioactivity in the ground. The ground is, and always has been, full of naturally radioactive materials--principally potassium, uranium, and thorium. On a long-term basis (thousands of years or more), burying our radioactive waste would increase the total radioactivity in the top 2.000 feet of U.S. rock and soil only by 1 part in 10 million. Of course, the radioactivity is more concentrated in a waste repository, but that doesn't matter. The number of cancers depends only on the total number of radioactive atoms that get into people, and there is no reason why this should be larger if the waste is concentrated than if it is spread out all over the country. In fact, quite the opposite is the case -- being concentrated in a carefully selected site deep underground with the benefit of several engineered safeguards and subject to regular surveillance provides a given atom far less chance of getting into a human stomach than if it is randomly located in the ground. Incidentally, it can be shown that the natural radioactivity deep underground is doing virtually no damage to human health, so adding a tiny amount to it is truly innocuous.

When it is pointed out that concentration of the waste in one place does not increase the health hazard, one common complaint is that it does mean that the risks will be concentrated on the relatively small number of people who live in that area. That, of course, is true, but it is also true for just about any other environmental problem. We in Pittsburgh suffer from the air pollution generated in making steel for the whole country; citizens of Houston and a few other cities bear the brunt of the considerable health hazards from oil refineries that make our gasoline, and there are any number of similar examples. The health burden from 'these inequities is thousands of times larger than those from living near a nuclear waste repository will ever be.

Our calculation shows that the toxicity of the uranium ore originally mined out of the ground to produce the fuel from which the waste was generated is larger than the toxicity of the waste after 15,000 years; after that the hazard from the buried waste is

less than that from the ore if it had never been mined. After 100,000 years, there is more radioactive toxicity in the ground due to the natural radioactivity in the rock.directly above the repository (i.e., between the buried waste and the surface) than in the repository itself.

The design of the waste package to be buried includes several features to mitigate the hazards. For the first few hundred years, when the toxicity is rather high, there is a great deal of protection through various time delays before the waste can escape through groundwater. First and foremost, the waste will be buried in a rock formation in which 'there is little or no groundwater flow, and in which geologists are as certain as possible that there will be little groundwater flow for a very long time. If the geologists are wrong and an appreciable amount of water flow does develop in the rock formation in which the waste is buried, it would first have to dissolve away large fractions of that rock -- roughly half of the rock would have to be dissolved before half of the waste is dissolved. This factor might seem to offer minimal protection if the waste is buried in salt, since salt is readily dissolved in water. However, in the New Mexico area, where a repository for defense waste has been constructed, if all the water now flowing through the ground were diverted to flow through the salt, ii would take a million years to dissolve the salt enclosing the repository. The quantity of salt is enormous, while the water flow is not a stream but more like a dampness, slowly progressing through the ground.

The next laver of protection is the backfill material surrounding the waste package. This will be a clay, which tends to swell when wet to form a tight seal keeping water flow away from the package.

If water should penetrate the backfill, before it can reach the waste it must get past the metal casing in which the waste glass is to be sealed. Materials for this casing have been developed that give very impressive resistance to corrosion. One favorite is a titanium alloy. which was tested in a very hot (480 deg-F, versus a maximum expected repository temperature of 250 deg-F) and abnormally corrosive solution, and even under these extreme conditions, corrosion rates were such that water penetration would be delayed for a thousand years. Under more normal groundwater conditions, these casings would retain their integrity for hundreds of thousands of years. Thus, the casings alone provide a rather complete protection system even if everything else fails.

The next layer of protection is the waste form, probably glass, which is not readily dissolved. Glass artifacts from ancient Babylonia have been found in river beds, where they have been washed over by flowing river water--not just by the slowly seeping dampness which better describes groundwater conditions--for 3.000 years without dissolving away. A Canadian experiment with waste glass buried in soil permeated by groundwater indicates that it will last for a hundred million years--i.e., only about one 100 millionth dissolves each year.

But suppose that somehow some of the waste did become dissolved in groundwater. Groundwater moves very slowly, typically at less than 1 foot per day--in the region of the New Mexico site it moves only about 1 inch per day and at the Yucca Mountain, Nevada site which is under investigation it moves only 0.03 inches per day. Furthermore, groundwater deep underground does not ordinarily travel vertically upward toward the surface; it rather follows the rock layers, which tend to be essentially horizontal, and hence it typically must travel about 50 miles before reaching the surface. In the New Mexico area it must travel over a hundred miles and at the Nevada site at least 30 miles. Anyone can easily calculate that to travel 50 miles at 1 foot per day takes about 1,000 years, so this again gives a very substantial protection for the few hundred years that most concern us here. (Long-term effects will be considered later.)

But the radioactive material does not move with the velocity of the groundwater. It is constantly filtered out by adsorption on the rock material, and as a result it travels hundreds or thousands of times more slowly than the groundwater. If the groundwater takes a thousand years to reach the surface, the bulk of the radioactive materials would take hundreds of thousands or millions of years.

We thus have seven layers of protection preventing the waste from getting out during the first few hundred years when it is highly toxic; (1) the absence of groundwater, (2) the insolubility of the surrounding rock, (3) the sealing action of the

Backfill material; (4) the corrosion resistance of the metal casing, (5) the insolubility of the waste glass itself, (6) the long time required for ground water to reach the surface, ­

and (7) the filtering action of the rock. But even if all of these protections should fail, the increased radioactivity in the water would be easily detected by routine monitoring, and the water would not be used for drinking or irrigation of food crops; thus there would still be no damage to human health.




Our next task is to develop a quantitative estimate of the hazard from buried radioactive waste, extending indefinitely into the future. Our treatment is based on an analogy between buried radioactive waste and average rock. We will justify this analogy later, but it is very useful because we know a great deal about the behavior of average rock and its interaction with groundwater. Using this information in a calculation outlined in Appendix C demonstrates that an atom of average rock 2000 feet below the surface has one chance in a trillion each year of being dissolved out of the rock, eventualIy being carried into a river or well, and ending up in a human stomach.

In the spirit of our analogy between this average rock and buried radioactive waste, we assume that this probability also applies to the latter; that is, an atom of buried radioactive waste has one chance in a trillion of reaching a human stomach each year. Since we know from our calculation outlined in Appendix B the consequences of all of the waste reaching a human stomach, all we need do is multiply those results by one-trillionth to obtain the number of fatal cancers expected each year. For example, if all of the waste were to reach human stomachs a thousand years after burial, there would be 300,000 deaths, but since only one-trillionth of it reaches human stomachs each year, there would be (300,000 / 1 trillion =) one chance in 3 million of a single death each year.

What we really want to know is the total number of people who will eventually die from the waste. Since we know how many will die each year, we simply must add these up. Totalling them over millions of years gives 0.014 eventual deaths. One

might ask how far into the future we should carry this addition, and therein lies a complication that requires some explanation.

From the measured rate at which rivers carry dissolved and suspended material into the ocean each year, it is straightforward to calculate that the surfaces of the continents are eroding away at an average rate of 1 meter (3.3 feet) of depth every 22,000 years. (All the while the continents are being uplifted such that on an average the surface remains at about the same elevation.) Nearly all of this erosion comes from river beds and the minor streams and runoff that feed them. However, rivers change their courses frequently; changing climates develop new rivers and eliminate old ones (e.g., 10,000 years ago the Arizona desert was a rain forest): and land areas rise and fall under geological pressures to change drainage patterns. Therefore, averaged over very long time periods, it is reasonable to assume that most land areas erode away at roughly this average rate, 1 meter of depth every 22,000 years. This suggests that the

ground surface will eventually erode down to the level of the buried radioactive waste.   Since the waste will be buried at a depth of about 600 meters, this may be expected to occur after about (600 x 22,000 =) 13 million years. When it happens, the remaining radioactivity will be released into rivers, and according to estimates developed in Appendix A, one part in 10,000 will get into human stomachs.

Therefore, the process of adding up the number of deaths each year should be discontinued after 13 million years; that was done in calculating the number of deaths given above as 0.014. To this we must add the number of deaths caused by release of the remaining radioactivity into rivers. The calculation in Appendix C shows that this leads to 0.004 deaths. Adding that number to the 0.014 deaths during the first 13 million years gives 0.018 as the total number of eventual deaths from the waste produced by one power plant in one year. This is the -final result of this section, 0.018 eventual fatalities. Recall now that air pollution, one of the wastes from coal burning, kills about 25 people in generating the same amount of electricity. We see that this one waste from coal burning is a thousand times more injurious to human health than are the high-level nuclear wastes.

Before closing this section, readers with a scientific bent may be interested in t1 he justification for the basic assumption used in arriving at our estimate; the analogy between buried radioactive waste and average rock. Actually, there are three ways in which the waste may be less secure:

1. The radioactivity in the waste continues to generate heat for some time after burial, making the buried waste considerably hotter than average rock. There has been some fear that this heat may crack the surrounding rock, thereby introducing easy pathways for groundwater to reach the waste and/or to carry off the dissolved radioactivity. This question has been extensively studied since the late 1960s, and the conclusion now is that rock cracking is not a problem unless the temperature gets up to about 650°F. The actual temperature of the waste can be controlled by the extent to which it is diluted with inert glass and by the spacings chosen for burial of the waste packages. More dilution and larger spacings result in lower temperatures. In the repositories now envisioned, these factors will be adjusted to keep the temperature below 250°F, which leaves a large margin of safety relative to the 650°F danger point. If it is decided that this margin should be larger, the easiest procedure would be to delay the burial, since the heat evolving from the waste decreases tenfold after a hundred years, and a hundredfold after 200 years. Above ground temporary storage facilities for waste packages have been designed, and there would be no great difficulty or expense in building them.

2. The waste glass will be a foreign material in its environment and hence will not be in chemical equilibrium with the surrounding rock and groundwater. However, chemical equilibrium is a surface phenomenon. Since rocks are largely composed of silica, groundwater reaching the waste package would be saturated with silica. As the waste glass is largely silica, when groundwater reaches it, the silica of its surface can only be dissolved by precipitating out the most insoluble silica compounds which then coat the surface. As this coating builds up in thickness, it shelters the material beneath from further interaction with the groundwater. The dissolution of the waste glass is therefore effectively stopped after only a minute quantity of it is dissolved.

3. Shafts must be dug down from the surface in order to emplace the waste packages. This might provide access routes for intrusion of water and subsequent escape of dissolved waste that are not available to ordinary rock. The seriousness of this problem depends on how securely the shafts can be sealed. That is a question for experts in sealing, and the general opinion among the experts is that the shafts can be sealed to make them at least as secure as the original undisturbed rock.


Since we have discussed the three ways in which buried waste may be less secure than the average rock used in our analogy. it is appropriate to point out ways in which it is more secure. In the first place, burial will be in an area free of groundwater flow, whereas the rock we considered was already submerged in flowing groundwater. Second, it will be in a carefully chosen rock environment, whereas the rock in our analogy is in an average environment; it might, for instance, be in an area of extensive rock fracturing. Third, the waste package will be in a special casing which provides extreme resistance to water intrusion, a protection not available to average rock; this casing should provide adequate protection even if everything else goes wrong. And finally, if any radioactive waste should escape and get into food and water supplies, it would easily be detected by routine monitoring programs in plenty of time to avert any appreciable health consequences.

In balancing the ways in which buried waste is more secure and less secure than the average rock submerged in groundwater with which we compared it, it seems reasonable to conclude that our analogy is a fair one.            If anything, it overestimates the dangers from the waste.

It should be pointed out that there are other methods than the one presented here for quantifying the probability for an atom of material in the ground to enter a human stomach. For example, we know the amounts of uranium and radium in the human body from measurements on cadavers, and we know the amounts in the ground. From the ratios of these, we can estimate the probability per year for an atom of these elements in the ground to enter a human body. This gives a result in good agreement with that of the calculation we presented.

One sometimes hears a complaint that the demonstration we have given is invalid because we don't know how to bury waste. How anyone can possibly be so naive as to believe that we don't know how to bury a simple package is truly a challenge to the imagination. Many animals bury things, and people have been burying things for a million years. Any normal person could easily suggest several different ways of burying a waste package. Why, then, are we devoting so much time and effort to developing a waste burial technology? The problem here is to decide on the best way of burying the waste to maximize its security. Whether it is worthwhile to go to such pains to optimize the solution is highly questionable, especially since we have shown that even the haphazard burial of average rock by natural processes would provide adequate safety. But the public's irrational worries about the dangers

have forced our government to spend large sums of money on the problem . Actually, we are paying for it in our electricity bills. Fortunately, this increases these bills by only

one percent.




In order to appreciate the meaning of the result obtained in the last section­ --018 eventual cancer deaths from each plant-year of operation -- it is useful to put it in perspective by comparing it with the very long-term cancer risks from the buried

wastes produced by other methods of generating electricity. Here we consider certain types of chemicals that cause cancer.

The only reason buried radioactive waste has a calculable health effect is because we are using a linear dose-response relationship, assuming, for example, that if 1 million mrem causes a given risk, 1 mrem will cause a millionth of that risk. This linear relationship does not apply to most toxic chemicals, like carbon monoxide poisoning, for which there is a "threshold" exposure required before there can be any harm. However, it is now widely believed that a linear, no-threshold relationship is as valid for chemical carcinogens (i.e., chemicals that can induce cancer)  as for radiation. Basically. the reason for this thinking in both cases is that cancer starts from injury to a single DNA molecule on a single chromosome in the nucleus of a single cell, often induced by a single particle of radiation or by a single carcinogenic molecule. The probability for this simple process is not dependent on whether or not other carcinogenic molecules are causing injury elsewhere. Thus the probability of cancer is simply proportional to the number of carcinogenic molecules taken into the body, as described by a linear relationship. This linear, no-threshold, dose-response relationship for chemical carcinogens is officially accepted by all U.S. Government agencies charged with responsibilities in protecting public health, such as the Environmental Protection Agency, the Occupational. Safety and Health Administration, and the Food and Druq Administration.

The particular carcinogens of interest in our discussion are the chemical elements cadmium, arsenic, beryllium, nickel, and chromium. There is a considerable body of information indicating that all of these cause cancer and enough quantitative information to estimate the risk per gram of intake. All of them are present in small amounts in coal. Therefore, when coal is burned, they are released and eventually, by one route or another, end up in the ground. They can then be taken up by plant roots and get into food.   _            .            _ .

As in the case of radioactive waste, the principal problem in evaluating the hazard is to estimate the probability for an atom of these elements in the ground eventually to reach a human stomach. However, the problem is simpler here because the toxicity in these carcinogenic elements does not decay with time, as does that of radioactive waste. Since some quantity of these elements is ingested by humans every year, if an atom remains in the ground indefinitely, eventually after a long enough time, it has a good chance of reaching a human stomach. However, there are other processes for removing materials from the ground. The most important of these

is erosion of soil, with the materials being carried by rivers into the oceans. The probability for an atom in the ground to enter a human stomach is therefore the probability for this to happen before it is washed into the oceans. This is just the ratio of the rate for entering human stomachs to the rate of removal by erosion.

The rates for both of these are known. From chemical analyses of food and the quantity consumed each year; we can calculate the rate at which cadmium, for example, enters human stomachs in the United States. From the rate at which American soil is washed into oceans and the measured abundance of cadmium in soil, we can calculate the rate at which cadmium is carried into oceans. We thus can determine the probability for an atom of cadmium in the soil to enter a human stomach before it is washed into the ocean. It turns out to be surprisingly high, about 1.3%. it is 5 times smaller for arsenic, 20 times lower for beryllium and nickel, and 60 times lower for chromium (cadmium atoms are more easily picked up by plant roots because of their chemical properties).

Since we know the quantities of each of these carcinogenic elements deposited into the top layers of the ground as a result of coal burning, the probability for them to be transferred from the ground into humans, and their cancer risk once ingested; it is straightforward to estimate the number of deaths caused by coal burning. The result is about 50 deaths due to one year's releases from one large plant. These deaths are spread over the time period during which the top few meters of ground will be eroded away, about 100,000 years.

If one takes a much longer-term viewpoint, these deaths would have eventually occurred even if the coal had not been mined and burned, because after several millions of years, erosion would have brought this coal near the surface, where its cadmium, arsenic, beryllium, chromium, and nickel could have been picked up by plant roots and gotten into food.

However, the fact that the coal was mined means that its turn near the surface is taken by other rock which, we assume, has the average amount of these carcinogenic elements in it. This carcinogenic material represents a net additional health risk to humankind due to the use of coal. Taking this into account, the final result is that there would be a net excess of 70 deaths. Note that this is thousands of times larger than the 0.018 deaths from high-level radioactive waste.

The chemical carcinogens also have an impact on the long term health consequences of electricity produced by solar energy. It is estimated that producing the materials for deployment of a solar array requires about 3% as much coal burning as producing the same amount of electricity by direct coal burning. The quantity of chemical carcinogens released in the former process is therefore 3% of those released in the latter, so in the very long term we may expect about 2 eventual deaths (3% of 70) from an amount of solar electricity equivalent to that produced by a large nuclear or coal-fired power plant in one year. This is a small number, but it is still a hundred times the 0.018 deaths from nuclear plant waste.

But there may be much more important problems. A prime candidate material for future solar cells is cadmium sulfide. When these cells deteriorate, that material will probably be disposed of in one way or another into the ground. The cadmium

introduced into the ground in this way is estimated to eventually cause 80 deaths for the quantity of electricity produced by a large nuclear plant in one year. Most of the cadmium used in the U.S. is imported. Thus. in contrast to the situation for coal, it would not have eventually reached the surface of U.S. soil without this technology. Again we encounter consequences thousands of times higher than the 0.018 deaths from radioactive waste.




Many may be bothered by the idea of adding up effects over millions of years as we have been doing here. Personally, I agree with them. The idea of considering effects over such long time periods was introduced by opponents of nuclear power, who often insist that it be done. The analyses presented above show that such a procedure turns out very favorably for nuclear power. However, there are many good reasons for confining our attention to the "foreseeable future." which is generally interpreted to be 100-1,000 years. Several groups, including most government agencies, have adopted 500 years as a reasonable time period for consideration, so we will use it for this discussion.

The reason for ignoring lives that may be lost in the far future is not because those lives are less valuable than our own or those of our closer progeny. There surely can be no moral basis for any such claim. But there are at least three reasonable and moral bases for ignoring the far distant future.

One such basis is the improvement in cancer-cure rates. In recent years. they have been improving by about 0.5% per year.            It is unlikely that cure rates will continue to improve at this rate. More probably they will stagnate before long, or on the other hand there may be a breakthrough leading to a dramatic improvement. But most knowledgeable people agree that there is a good chance that cancer will be a largely curable disease in 500 years. The deaths we calculate to occur after that time will therefore probably never materialize.

The second basis for ignoring deaths calculated to occur in the distant future is what I call the "trust fund" approach. There are now many ways -- cancer screening programs and highway safety measures are examples -- in which money can be used to save lives at a rate of about one life saved per $100,000 spent. But even suppose that so many improvements are initiated that it will cost $1 million to save a life in the distant future (all our discussions discount inflation). There is a continuous record extending back 5,000 years of money always being able to draw at least 3% real (i.e., discounting inflation) interest. Each dollar invested now at 3% interest is worth $2.5 million after 500 years and will therefore be capable of saving more than one life. It is therefore much more effective in saving far future lives to set up a trust fund to be spent in their time for that purpose rather than for us to spend money now to protect, them from our waste. One might wonder about the mechanics and practicality of setting up a trust fund to be transmitted to future generations, but this is no problem. By overspending, we are now building up a public debt that they will have to pay interest on, so rather than actually sat up a trust fund, we need only reduce our spending and thereby leave future generations with more money to use for lifesaving. By not spending profligately on waste management, we do just that.

A third basis for ignoring the distant future in our considerations is the "biomedical research" approach. if we don't feel comfortable with the "trust fund" approach of putting aside money to be spent in the future for saving lives, we can save future lives by spending money today on biomedical research. One study concludes that the $68 billion spent on such research between 1930 and 1975 is now saving 100,000 lives per year in the United States, one life per year for every $680,000 spent. Another study estimates that the $20 billion spent in 1955-1965 is now saving 100;000 lives per year, or one life per year for every $200,000 spent. Age-specific mortality rates in the United States have been declining steadily in recent years: if one-fourth of this decline is credited to biomedical research, the latter is now saving one life per year for every $1 million spent. As the easier medical problems are solved and the more difficult ones are attacked, the cost goes up. Let us therefore say that for the near future, it will cost $5 million to save one life per year in the United States, not to mention that it will save lives in other countries. Over the next 500 years, saving one life per year will avert a total of 500 deaths, a cost of $10,000 per life saved. If we extend our considerations beyond 500 years, the price drops proportionally. Since we are now spending many millions of dollars per future life saved in handling our radioactive waste, it would clearly be much more beneficial to future generations if we put this money into biomedical research.

The straightforward way to implement this alternative would be for the nuclear industry to contribute money to biomedical research rather than to improved long-term security of nuclear waste, but even that may not be necessary. The nuclear industry pays money to the government in taxes, and the government spends roughly an equal sum of money to support biomedical research. All that is necessary to implement an equitable program from the viewpoint of future citizens is that 0.1 % of this cash flow from the nuclear industry to government to biomedical research be redefined as support of biomedical research by the nuclear industry. If it would make people happier, the government tax on the nuclear industry could actually be raised by 0.1% and this money could be directly added to the government's biomedical research budget. No one would notice the difference since it is much less than a typical annual change in these taxes and budgets.

As a result of any one of the above three rationales, or a combination of them, many people, including myself, feel that we should worry only about effects over the next 500 years. For their benefit we now consider the health effects expected during that time period, from one year's operation of a large power plant

For the high-level radioactive waste, the calculation we have previously described added up over the first 500 years gives 0.003 deaths, most of them in the first 100 years. However, this ignores all the time delays and special protections during this early time period that were outlined above: geologists feel confident that no groundwater will intrude into the waste repository for at least a thousand years; the corrosion-resistant casing is virtually impregnable for several hundred years; it ordinarily takes groundwater a thousand years to get to the surface; and so on.

Moreover, since it is quite certain that some sort of surveillance would be maintained during this period, any escaping radioactivity would be detected in plenty of time to

avert health problems. These factors reduce the dancers at least 30-fold -- probably

much more -- to no more than 0.0001 deaths. This is close to a million times less then the number of deaths due to air pollution from coal burning.

The chemical carcinogens released into the ground in coal burning do about 1 percent of their damage during the first 500 years; we may consequently expect 0.5 deaths in that period. (With a 500-year perspective, there is no need to consider the much later effects of the coal reaching the surface by erosion if it had not been mined.) We see that the effects of chemical carcinogens from coal burning are 5,000 times worse than those of nuclear waste (0.5 / 0.0001).

By similar reasoning, the chemical carcinogens from solar electricity will cause 0.015 deaths due to the coal used in making materials for it; if cadmium sulfide solar cells are used. they will cause 0.8 deaths. Again these are far greater than the 0.0001

deaths due to the waste from nuclear power.




EPA estimates that exposure to radon in homes is now causing about 14,000 fatal lung cancers each year in the United States, more fatalities than are caused by all other natural radiation sources combined. This radon is generated from the decay of natural uranium in. the ground.

When uranium is mined out of the around to make nuclear fuel, it is no longer there as a source of radon emission. Since the great majority of uranium mined  comes from depths well below 1 meter, the radon emanating from it is often viewed as harmless. The fallacy of this reasoning is that it ignores erosion. As the ground erodes away at a rate of 1 meter every 22,000 years, any uranium in it will eventually approach the surface, spending its 22,000 years in the top meter, where it will presumably do great damage. The magnitude of this damage is calculated in Appendix D, where it is shown that mining uranium to fuel one large nuclear power plant for one year will eventually save about 600 lives! This completely overshadows all other health impacts of the nuclear industry, making it one of the greatest lifesaving enterprises of all time if one adopts a very long-term viewpoint.

Before we can count these lives as permanently saved, we must specify what is to be done with the uranium, for only a tiny fraction of it is burned in today's nuclear reactors. There are two dispositions that would be completely satisfactory from the lifesaving viewpoint. The preferable one would be to burn it in breeder reactors and thereby derive energy from it. An easy alternative, however, is to dump it into the ocean. Uranium remains in the ocean only for about a million years before settling permanently into the bottom sediments. All uranium in the ground is destined eventually to be carried by rivers into the ocean and spend its million years therein. From a long-range viewpoint it makes little difference to the health of humans or of sea animals if it spends that time in the ocean now or a million years in the future. But by preventing it from having its 22,000 year interlude within 1 meter of the ground surface, we are saving numerous human lives from being lost due to radon.

If one adopts the position that only effects over the next 500 years are relevant, there is still an important effect from uranium mining because about half of all uranium is surface mined. Approximately 1 % of this comes from within 1 meter of the ground surface where it is now serving as a source of radon exposure. Calculations indicate that eliminating this source of exposure by mining will save 0.07 lives over the next 500 years.

We still have one more radon problem to discuss, namely, the radon released in coal burning. Coal contains small quantities of uranium; when the coal is burned, by one route or another, , this uranium ends up somewhere in the ground. Again, the

problem is complicated by the fact that, if the coal had not been mined, erosion would eventually have brought the coal with its uranium to the surface anyhow. The final result is that the extra radon emissions caused by the burning of coal in one large power plant for one year will eventually cause 30 fatal lung cancers. This toll, like so many others we have encountered here; is thousands of times larger than the 0.018 deaths caused by the high-level waste produced in generating the same amount of electricity from nuclear fuel.

As discussed previously, solar electricity burns 3% as much coal as would be needed to produce the same amount of electricity by direct coal burning. Solar electricity should therefore be charged with (3% of 30 =) one death per year from radon. This again, is far larger than the O.018 deaths per year from high-level radioactive waste.



APPENDIX A: Probabilities for Entering a Human Body


We inhale about 20 cubic meters of air per day, or 7,000  m3 per year. Dust levels in air from materials on the ground becoming suspended are about 35 E-6 g/m3. Thus we inhale (20 x 35 E-6 =) 0.7 E-3 grams per day of material from the ground, or 0.25 grams per year.

The area of the United States is about 1 E13 m2, so the volume of the top inch (0.025 m) of soil is (.025 x 1E13 =) 2.5 E11 m3.            Since the density of soil is 2 E6 g/m3, this soil weighs (2 E6 x 2.5 E11 =) 5  E17 grams. Since each person inhales 0.25 g/yr of this soil, the quantity inhaled by the U.S. population (250 E6) is (250 E6 x 0.25 =) 6 E7 g/yr. The probability for any one atom in the top inch of U.S. soil to be inhaled by a human in one year is therefore (6 x 107 / 5 E17 =) 1.2 E-10, a little more than 1 chance in 10 billion.

The probability for an atom in a river to enter a human is very much larger. The total annual water flow in U.S. rivers is 1.5 E15 liters, whereas the total amount ingested by humans is (2.2 liters/person per day x 365 days/year x 250 E6 =) 1.8 E11 liters per year. Thus the probability for an atom in a river to be ingested by a human is (1.8 E11 /  1.5 E15 =) 1.2 E-4, or a little more than 1 chance in 10, 000 per year.



APPENDIX B: Calculation of death s caused if all of the waste enters human stomachs


For illustrative purposes, let us calculate the number of liver cancers expected from people eating 1 millicurie (3.7 E7 radioactive decays per second) of plutonium-239 (239Pu) that is present in the waste. Since the radiation emitted by 239Pu, called alpha particles, does not go very far--It can  barely get through a thin sheet of paper -- ­this material can cause liver cancer only if it gets into the liver. Experiments indicate. that 0.01% of ingested 239Pu gets through the walls of the gastrointestinal tract into the bloodstream, and of this, 45% is deposited in the liver; thus (3.7 E7 x 0.0001 x0.45 =)   1,700 alpha particles strike the liver each second. Since 239Pu remains in the liver for an average of 40 years (1.2 E9 seconds), the total number of alpha particles that eventually strike the liver is (1,700 x 1.2 x E9 =)  2 E12. This is multiplied by the energy of the alpha particle to give the energy deposited, 1.5 joules, which is then divided by the mass of the liver, 1.8 kg, and multiplied by a conversion factor to give the dose in millirem, 1.6 E6.

The risk of liver cancer per millirem of alpha bombardment is estimated from studies of patients exposed for medical purposes to be 0.015 E-6 per millirem. The number of liver cancers expected from eating 1 millicurie of 239Pu is therefore (1.6 E6  x :015 x E-6 =) 0.024. Since the waste produced in one year by one plant contains 6 E4  millicuries of 239Pu, the number of liver cancers expected if this were fed to people would be (6 E4 x 0.024 =) 1,400.

But once 239Pu gets into the bloodstream, it can also get into the bone--45% accumulates there, and it stays for the remainder of life; it therefore can cause bone cancer. A calculation like that outlined above indicates that 700 cases are expected. When other body organs are treated similarly, the total number of cancers expected totals 2,300 if all of the plutonium in the waste from one year of operation of one nuclear power plant were fed to people.

The quantity of 239Pu in the waste does not stay constant, for two reasons. Every time a particle of radiation is emitted, a 239'Pu atom is destroyed, causing the quantity to decrease. But 239Pu is the residue formed when another radioactive atom,

americium- 243, emits radiation, which adds to the quantity. These two effects cause the toxicity of the plutonium to vary with time in a manner that can easily be accurately calculated.

There are many other radioactive species besides 239Pu in the waste; similar calculations are done for them and the results are added to give the total number of deaths expected if all the waste produced by one power plant in one year were fed to people in digestible form.



APPENDIX C: Probability for an Atom of Rock to Enter a Human Stomach


From the measured rate at which rivers carry dissolved material into oceans, it is straightforward to calculate that an average of 1.4 E-5 meters of depth is eroded away each year. Hydrologists estimate that 26% of this erosion is from dissolution of rock by groundwater; the rest is from surface water. Thus (0.26 x 1.4 E-5 =)  3.6 E-6 meters of depth are dissolved annually by groundwater. The fraction of this derived from 1 meter  of depth at 600 meters below the surface may be estimated from our knowledge of how groundwater flow varies with depth; it is about 2.6 x E-4. The total amount of rock derived from our 1 meter of depth in 1 year is then (3.6 E-6 x 2.6 E-4 =)  1 E-9 meters per year. If 1 E-9 meters is removed loved from 1 meter of depth each year, the probability for any one atom to be removed is 1 E-9, one chance in a billion per year. This result applies to all rock, averaged over the continent.

We now provide an alternative derivation for an atom of rock which is submerged in groundwater. Consider a flow of groundwater, called an aquifer, along a path through average rock and eventually into a river. There is a great deal of information available on aquifers, like their paths through the rocks, the amount of water they carry into rivers each year, and the amounts of various materials dissolved in them. From the latter two pieces of information, we can calculate the quantity of each chemical element carried into the river each year by an average aquifer--i.e., how much iron, how much uranium; how much aluminum, etc.

Where did this iron, uranium, and aluminum in the groundwater come from? Clearly, this material was dissolved out of the rock. From our knowledge of the path of the aquifer through the rock and the chemical composition of rock, we know the

quantity of each of the chemical elements that is contained in the rock traversed by the aquifer. We can therefore calculate the fraction of each element in the rock that is dissolved out and carried into the river each year. For example, a particular aquifer may carry 0.003 pounds of uranium into a river each year that it dissolved out of a 50­mile-long path through 200 million tons of rock that contains 1;000,000 pounds of uranium as an impurity (this is typical of the amount of uranium in ordinary rock). The fraction of the uranium removed each year is then 0.003/1 ,000,000, or 3 parts per billion. Similar calculations give 0.3 parts per billion for iron, 20 parts per billion for calcium, 7 parts per billion for potassium, and 10 parts per billion for magnesium. To simplify our discussion, let us say that 10 parts per billion of everything is removed each year--this is faster than the actual removal rates for most elements. This means that the probability for any atom to be removed is 10 chances in a billion each year. This is 10 times larger than our first estimate, 1 chance in a billion. This is partly explained by the fact that most rock is not submerged in groundwater, as we have assumed here. But the important point is that we have given two entirely independent derivations and have arrived at roughly the same result. To be conservative, we will, use the higher probability, 10 chances in a billion per year. Incidentally, this result implies that for ordinary rock submerged in groundwater, only 1 % is removed per million years [(10/1 billion) x 1 million = 0.01], so it will typically last for 100 million years.

It was shown in Appendix A that an average molecule of water in a river has 1 chance in 10,000 of entering a human stomach before flowing into the Oceans. For materials dissolved in the water, the probability is somewhat smaller because some of it is removed in drinking water purification processes, but the probability is increased by the fact that some material in rivers finds its way into food and enters human stomachs by that route. These two effects roughly compensate one another. We therefore estimate that material dissolved in rivers has 1 chance in 10,000 of getting into a human stomach.

Since an atom of the rock has 10 chances in a billion of reaching a river each

year, and once in ! a river has 1 chance in 10,000 of reaching a human stomach, the overall probability for an atom of the rock to reach a human stomach is the product of these numbers (10 / 1 billion) x (1 / 10,000) =) 1 chance in a trillion per year. That is the value used in our discussion. In more elaborate treatments, use of well water and irrigation water is included, but the final result is essentially the same.

In considering effects of erosion; we assumed that all the waste would be released  into rivers after 13 million years. According to the calculation in Appendix B, if all of the

waste remaining at that time were to get into human stomachs, about 40 deaths would be expected. But we have just shown that if the material is released into rivers, only 1 atom in 10,000 reaches a human stomach; we thus expect (40/10,000 =) 0.004 deaths. That is the result used in our discussion



APPENDIX D: Radon-caused deaths averted by mining uranium to fuel nuclear plants


The radon to which we are now exposed comes from the uranium and its decay products in the top 1 meter of U.S. soil, since anything percolating up from deeper regions will decay before reaching the surface. From the quantity of uranium in soil (2.7 parts per million) and the land area of the United States (contiguous 48 states), it is straightforward to calculate that there are 66 million tons of uranium in the top meter of U.S. soil. This is now causing something like 14,000 deaths per year from radon, and will continue to do so for about 22,000 years, the time before it erodes away. This is a total of (14,000 x 22;000 =) 300 million deaths caused by 66 million tons of uranium, or 4.5 deaths per ton. As erosion continues, all uranium in the ground will eventually have its 22,000 years in the top meter of U.S. soil and will hence cause 4.5 deaths per ton.

In obtaining fuel for one nuclear power plant to operate for 1 year, 180 tons of uranium is mined out of the ground. This action may therefore be expected to avert (180 x 4.5=) 800 deaths. This must be corrected for the effects of radon emitted from uranium mill tailings, not considered here, which reduces the net effect to 600 deaths averted.