Disturbing questions

Print edition : May 22, 1999

The heavy water leak in the Madras Atomic Power Station on March 26 has raised important questions about safety standards.

THE heavy water leak in the second unit of the Madras Atomic Power Station (MAPS) on March 26 (Frontline, April 23) raises disturbing questions. The most important among them are: how much of the spilled radioactive heavy water escaped into the environment, and what were the radiation doses that the 42 workers who "mopped up" the spill were subjected to? If one were to go by official statements, the impact of this accident would seem to be negligible. However, even a rough assessment of the magnitude of the dose and an analysis of past experience indicate otherwise.

There have been at least two similar accidents at MAPS. A heavy water leak forced the reactor to be shut down in September 1988. On March 5, 1991, 0.847 tonnes of heavy water escaped from the moderator system.

Heavy water is used both to slow down neutrons generated by the fission of uranium (that is, as a moderator) and to remove the resultant heat (that is, as a coolant). Over a period of time, the heavy water becomes radioactive because some of the heavy hydrogen absorbs neutrons to become tritium (hydrogen with two neutrons in an atom). Therefore, the process of cleaning up spills and recovering the heavy water or flushing it into the environment almost invariably leads to workers and the general public receiving radiation doses. During the 1997 strike by workers at MAPS, it was revealed that there were areas within the reactor which had routinely (that is, not related to spills or other accidents) high radiation levels and that employees were forced to work in these areas.

Some details of the 1991 accident were reported by four researchers from the Health Physics Division of the Bhabha Atomic Research Centre (BARC) in a paper presented at the National Symposium on Safety of Nuclear Power Plants and Other Facilities (March 11-13, 1992). Of the heavy water that had leaked, only 350 kg (barely 41 per cent) was recovered. Over 53 per cent of the spilled heavy water was released into the atmosphere through the stack and about 3 per cent was released into the sea. On March 5, 1991, the radioactivity concentration of the heavy water was 10 curies/litre and the radioactivity concentration of the air in the chamber was 3,225 DAC (Derived Air Activity). Derived Air Activity is defined as the radioactivity concentration level in the air such that if a worker were to work for a whole year (assumed to be 2,000 hours) in such an atmosphere, he or she would inhale the annual limit for tritiated water vapour set by the International Commission on Radiological Protection. In the case of tritiated water vapour, the DAC is 20 microcuries/cubic metre. Clearly, 3,225 DAC of tritium air activity is dangerously high. A worker will receive more than the annual limit if he or she were to work in such an atmosphere for even an hour. Since the volume of the spill was relatively small, the reactor containment is unlikely to have been saturated with heavy water vapour and the air concentrations lower than the theoretical limit. A larger spill would have led to even higher concentrations of tritium in the air.

The two units of the Madras Atomic Power Station (MAPS) at Kalpakkam near Chennai.-S. THANTHONI

Compared to the 1991 accident, the recent spill was much larger in magnitude. There have been different estimates about the amount of heavy water that leaked. According to Dr. K.S. Parthasarathy, Secretary, Atomic Energy Regulatory Board (AERB), it was less than four tonnes, whereas Dr. A. Gopalakrishnan, former Chairman of the AERB, felt that it could have been as high as 14 tonnes. According to a Press Trust of India report, it was about six tonnes. Despite the differences in estimates, it is clear that a large quantity of heavy water leaked. There is, however, no argument over the fact that the heavy water came from the coolant cycle.

If one were to assume that the heavy water from the recent spill was collected with the same efficiency as it was done in 1991, then somewhere between two and seven tonnes would have been released into the atmosphere through the stack. The radioactivity level of the heavy water depends on the area of its origin (whether it was from the coolant or the moderator), and the length of time it has been in the reactor. Typical values for coolant heavy water are in the range of 0.5 to 2 curies/kg. Thus, somewhere between 1,000 and 14,000 curies may have been released - several times the permitted 300 curies per day per reactor and perhaps even exceeding the discharge limit of 10 times the daily quota. Had the heavy water leaked from the moderator, it would have had 20-30 times more radioactivity, and thus the release of radioactivity into the atmosphere would have been that much greater.

According to the 1991-92 Department of Atomic Energy Annual Report, standards for MAPS are set in such a way that members of the public do not receive more than 1 mSv/year (or 0.1 rem/year) per person as recommended by the International Commission on Radiological Protection (ICRP). A release that is many times the prescribed limit would presumably lead to a potential dose which is that many times higher.

It is possible to estimate the radiation doses received by the workers involved in cleaning up the recent spill. The dosage depends chiefly on the radiation levels of the heavy water spilt and the length of time the workers are involved in the clean-up, and is independent of the extent of heavy water that is released into the atmosphere. As mentioned earlier, if the heavy water came from the coolant cycle, it would have a radioactivity concentration of about 1 curie/kg. While most of the water would remain in liquid form, part of it would evaporate and workers trying to clean it up would inhale the vapour. Since tritium emits relatively low-energy electrons, the primary dose received by workers would be internal - when the tritium enters the body. The amount of water that would evaporate is determined by the temperature. Just as clothes dry faster on a warm day, the level of vapourisation goes up with the temperature. The temperature recorded in Chennai on the morning of March 26 was 36o Celsius; at this temperature, the partial vapour pressure of water is about 40 mm of mercury. This means that the equilibrium fraction of heavy water content would be about 40/760 within the reactor chamber. At this concentration, the heavy water vapour would have a radioactivity of 0.053 Ci/cubic metre.

We may assume conservatively that the concentration was only 0.01 Ci/cubic metre to allow for a lower humidity level, some admixture with ordinary water vapour, and the possibility of lower radioactivity levels in the heavy water. This is much smaller than the measured tritium air activity during the 1991 spill. Another useful point of comparison is the tritium concentration measured in experiments conducted by the TLD laboratory at MAPS (reported in the Bulletin of Radiation Protection, Vol. 10, 1987). In these experiments, a radioactivity concentration of 0.016 Ci/cubic metre was found to result from tritiated heavy water with a radioactivity level of 0.8 Ci/litre. Scaling this to the assumed 1 Ci/litre in the case of the heavy water spill in MAPS would result in a radioactivity concentration of 0.02 Ci/cubic metre, twice the value we have assumed.

Radiation safety standards in India and elsewhere are set by assuming that the average worker engaged in "light activity" breathes in a little over one cubic metre of air each hour and that the average worker weighs 70 kg. If workers weigh less or breathe more air, their dose would increase. Under these assumptions, assuming a concentration of 0.01 Ci/cubic metre and using standard methods of dose calculation, the radioactive dose to a single worker is calculated at about 6-8 mSv for each hour of work. This estimate should be compared with the 1991 recommendation by the ICRP, which set a limit of 20 mSv per year per worker averaged over five years, the effective dose not exceeding 50 mSv in each year. Thus, even at the lower limit an employee working for less than four hours would receive a dose in excess of the ICRP recommendation. Additional evidence that the actual dose the workers received is in this range is provided by the fact that representatives of the workers' union have said that seven of the workers who helped plug the leak have been placed in the "removal category" (The Hindu, April 9, 1999). The Nuclear Power Corporation (NPC) subsequently confirmed that "some of the persons involved in leakage rectification tasks received tritium uptake in excess of investigation level" and that these employees have been "restricted to work in the reactor as per the station procedures" (The Hindu, April 21, 1999).

If the dose of radiation received by each worker during each hour of work in the chamber where the spill occurred was 6 mSv, the collective exposure to 42 workers could be as high as 0.252 Sieverts for each hour they worked. Given the magnitude of the spill, it is likely that the "mopping up" operation would have taken several hours, if not days. (The 1991 spill was cleaned up over a period of four days.) Thus, the collective exposure would almost certainly exceed the 0.25 person Sieverts, the AERB Secretary has claimed.

THERE are additional components to radiation exposure. Some of the tritium gets bound to organic molecules and is not eliminated from the body for long periods of time. If workers come in contact with liquid tritiated water during mop-up operations or when the heavy water is upgraded for re-use, a dose from the tritiated water would be absorbed through the skin. Wearing plastic suits can reduce this risk, but it is unlikely to be completely eliminated. While both these paths of radiation doses are well known, standard methods do not incorporate them adequately because they are hard to quantify. In addition, if there are any small holes in the fuel element cladding, there could be tiny amounts of fission products in the heavy water, which would add to the radioactive dose. These additional components would only increase the size of the dose.

Over the last few decades, epidemiological and microbiological research has increasingly revealed that even low levels of radiation are hazardous. There is no scientific evidence about any threshold below which radiation doses may be considered "safe". Broadly speaking, two kinds of effects are predominant: an increase in a variety of cancers and mutations to genes. In addition, exposure of in utero foetuses to radiation could affect its physical and mental development.

Tritium could cause these effects because the body easily absorbs it in the form of tritiated water. Any tritiated water vapour that is inhaled, absorbed through the skin, or ingested will result in complete absorption of all the radioactivity. The absorbed tritiated water is rapidly distributed throughout the body via blood, which in turn equilibriates with extracellular fluid in about 12 minutes. Since tritiated water can pass through the placenta, it could lead to mental retardation and other developmental effects of babies when ingested by pregnant women. Further, some experiments suggest that tritium is more effective in causing many of these effects than, say, high-energy gamma rays resulting from nuclear explosions, because the low-energy electrons that result from the decay of tritium are more efficient in ionising cellular material.

The dose estimates made here are, of course, based on assumptions and are admittedly approximate. But it does provide a genuine basis for concern. For the authorities at MAPS, the AERB and the NPC to say merely that the releases and doses are within limits is insufficient and unsatisfactory. The only way to know for sure is that the authorities release all the relevant raw data so that their claims can be verified independently. It is important that we know if the radioactivity levels in the ambient air at the point of the spill were measured, what instruments were used to measure them, when and how these instruments were calibrated, and their measurements over a period of time - both during reactor operation and shutdown. All the workers who were in the plant at the time of the spill or clean-up operation should be monitored for internal exposure by analysing their urine samples. These data should be made public, although the identities of the workers need not necessarily be revealed. More importantly, adequate steps should be undertaken to ensure that such accidents do not occur again.

M.V. Ramana is Research Associate, Centre for Energy and Environmental Studies, Princeton University.

A letter from the Editor

Dear reader,

The COVID-19-induced lockdown and the absolute necessity for human beings to maintain a physical distance from one another in order to contain the pandemic has changed our lives in unimaginable ways. The print medium all over the world is no exception.

As the distribution of printed copies is unlikely to resume any time soon, Frontline will come to you only through the digital platform until the return of normality. The resources needed to keep up the good work that Frontline has been doing for the past 35 years and more are immense. It is a long journey indeed. Readers who have been part of this journey are our source of strength.

Subscribing to the online edition, I am confident, will make it mutually beneficial.


R. Vijaya Sankar

Editor, Frontline

Support Quality Journalism
This article is closed for comments.
Please Email the Editor