India's dangerous love affair with nuclear power

For decades, India’s nuclear establishment has been pushing for the expansion of nuclear energy to meet the country's growing energy needs without any regard to the fact that the country does not even have a proper independent regulatory authority to evaluate safety. Its efforts have found a ready response in the current administration, which, in its drive for nuclear expansion, is legislating to limit the liability of nuclear energy investors in the event of accidents.

Praful Bidwai

NUCLEAR power has been advocated for five decades as the ideal solution to India's energy crisis and as a cost-effective, reasonably safe option for electricity generation in keeping with the country's needs. But nuclear power has turned out to be inappropriate to India's need for decentralised and distributed energy generation, and far more expensive than electricity from renewable sources, such as wind turbines, as well as from fossil-fuel burning. Nuclear power poses major issues of safety, some of which are generic to that technology, but some of which have a special character defined by the Indian context, where there is no independent regulatory authority or safety audit, and where the operator, planner, licensor, builder and manager of all nuclear projects are all a single agency, the Department of Atomic Energy (DAE), which also acts as the supervisory and regulatory agency.

So poor is the DAE's project planning, so irresponsible is its management, and so persistent is its failure in learning from experience, that the Indian nuclear power programme would probably have lost steam and become totally marginal years ago had the DAE also not been responsible for India's nuclear weapons pursuit, on account of which it commanded political and financial clout. The nuclear tests of 1998 gave the DAE a boost and a bigger budget, but did not lead to a large nuclear capacity addition. That is now being planned through imports of uranium fuel, and of nuclear technology and reactors.

Ironically, India is banking on installing specific reactor types although they have not been approved on their home ground - e.g. the Westinghouse AP-1000 reactor or the Evolutionary Pressurised Water Reactor in Europe. More important, India has no agency that can independently evaluate the safety of these designs or set standards for the complex parameters for a sound, efficient and safe reactor. This means India is putting the cart before the horse. This is obnoxiously true of Jaitapur in Maharashtra where land has been acquired under emergency provisions although the plant has not received environmental clearance.

There are other problems too. India has limited reserves of uranium ore, only enough to run a water-cooled reactor-based programme with 10,000MW of capacity. The existing uranium mines are running out of ore. They cannot produce enough to feed the 17 operating reactors, some of which are being run at less than their full power rating. A confrontation has broken over the government's plan to open a new uranium mine in Meghalaya in the Northeast, which is strongly opposed by the local people. Therefore, India is also planning to reach agreements to import uranium from countries like Kazakhstan and Niger.

However, strong sentiment within India's nuclear establishment favours reliance on indigenous resources rather than imports, in view of past experience with uncertain fuel supplies to two US-built reactors at Tarapur, after India conducted its first nuclear explosion in 1974. But India's own uranium reserves cannot support an ambitious nuclear power programme, which would go much beyond 3 or 4% of the total power generation capacity projected for 2020/2025.

Aware of this, India's nuclear planners way back in the 1950s made a virtue out of necessity by proposing a three-stage nuclear programme for India involving fast-breeder reactors, which would make use of thorium, a material which India has in abundance. (Naturally occurring thorium-232 can be converted through fast neutron bombardment to uranium-233, which can sustain a chain reaction.)

The first stage of the programme would be based on pressurised heavy water reactors (PHWRs), from which plutonium would be extracted. The second stage is to be based on fast-breeder reactors using this plutonium, which theoretically produce more fissile material than they burn. The thorium would be placed in the 'blanket' surrounding their core. This would convert it to uranium-233. If depleted or natural uranium is placed in the blanket, plutonium would be produced. In the third stage, uranium-233 would be burnt in the core and thorium would be placed in the blanket.

This programme appears neat, but is based on two crucial assumptions. First, fast-breeders - which are a complicated, high-risk and accident-prone technology using hundreds of critical masses of plutonium or highly-enriched uranium - can be made reasonably safe, technologically robust, amenable to smooth, uninterrupted operation, and commercially affordable.

The second assumption is that thorium can be converted in large quantities into uranium-233 and used in special reactors on an industrial scale. The thorium-uranium 233 reactor sounds like a good idea, but it has only ever been demonstrated on a tiny, laboratory scale. Unless its techno-economic viability is proved on a pilot and then industrial scale, it will essentially remain only a concept.

The first assumption is equally fraught. The world's experience with fast-breeder reactors has been an unhappy one and impelled most countries to abandon their programmes. France persisted with breeders the longest of all, investing huge sums in large reactors like the Superphenix (1,200MW), launched with great fanfare. Superphenix shut down in 1996 after working at a lifetime average capacity factor of 6.6%. 'Fast-breeder' is a bit of a misnomer. In practice, nuclear reactors often do not yield more fissile material than they consume.

The problem with fast-breeders is that they concentrate enormous amounts of fissile material in small spaces and their chain reaction is sustained by 'fast neutrons'. This produces enormous amounts of heat in the core, which cannot be sufficiently drawn out by water, but needs materials like liquid sodium. Sodium explodes on contact with moisture or air.

India's own experience with breeders has been embarrassingly unsatisfactory. It built a small 14MW Fast Breeder Test Reactor (FBTR) with French assistance. This was expected to be commissioned in 1976, but achieved criticality nine years later and its steam generator began operating only in 1993. It has so far operated at a load factor of 20% and has repeatedly shut down due to sodium leaks and explosions. The first time it continuously operated for more than 50 days was in 2001. Yet, the DAE claims to have 'successfully demonstrated the design, construction and operation' of a fast-breeder reactor through the FBTR. In a profoundly irrational and unwise decision, the DAE is proceeding to build a much larger (500MW) Prototype Fast Breeder Reactor.

Safety concerns

Finally, India's record in respect of nuclear safety, or what is known of it, is deeply unsatisfactory, with numerous accidents and cases of exposure of occupational workers to radiation doses well in excess of the officially stipulated maximum permissible limits, at least 350 of which were documented from India's first nuclear stations at Tarapur.

Some of the accidents involved a fire in the turbine room, collapse during construction of a containment dome - a concrete shell which is meant to protect the environment against potential leaks from the reactor - and flooding of a reactor and its building during a maintenance shutdown. Not enough is known about the DAE's safety procedures and preparedness for mishaps, or its record of accidents and over-exposure to radiation because the Department operates under a veil of secrecy thanks to the Atomic Energy Act of 1962. This allows it to suppress any information that it does not wish to disclose. However, what is known about the way it operates its uranium mines, its transportation of nuclear materials and its waste storage practices raises serious concern.

These failures in safety management are compounded by a general lack of an industrial safety culture, and by the absence of independent oversight, safety audit and public accountability. The true social, health-related and environmental costs of nuclear power in India will only be known - on the basis of which alone can a rational judgment be exercised about the desirability of nuclear power - if the Atomic Energy Act is amended and an independent licensing and safety regulatory agency is created which reports to Parliament and exercises complete authority over the DAE and its subordinate agencies.

Such an agency must formulate transparent rules, procedures and norms on the basis of the Precautionary Principle, expert advice and state-of-the-art understanding of the best practices prevalent in the nuclear industry. It must subject them to public debate. It must make a serious environmental impact assessment based on transparent public consultation and hearings before approving a project site. And it must conduct health surveys both before project construction and periodically thereafter to assess health impacts. None of these is on the cards as India rushes into nuclear power expansion.

India is trying to do this by luring foreign investors, who want a law which limits the liability of the nuclear industry in the event of accidents. The Indian government has drafted the Civil Liability for Nuclear Damage Bill to do this. But it makes no sense to cap the liability for a potentially catastrophic mishap in an accident-prone, highly hazardous industry, whose radioactive fallout can produce cancers and contaminate large areas for centuries.

Liability limits

Each of the 430-odd nuclear reactors worldwide can experience a reactor-core meltdown, like Chernobyl. Till 2007, 63 potentially catastrophic nuclear accidents were documented in them, including hair-raising Loss of Coolant Accidents (LOCAs). In an LOCA, the coolant - usually water, which must continuously draw out heat from the core - is lost through leaks, evaporation or chemical reaction. Unless the LOCA is contained, the core overheats, and a runaway chain reaction can lead to a meltdown.

Although this probability is low, its consequences are catastrophic -hundreds of early deaths from burns and acute radiation poisoning, and tens of thousands from cancers over decades; environmental contamination, and poisoning of vegetation and animal life.

The economic damage from Chernobyl, in which an estimated 65,000 people died from cancers, is $390 billion. Should a Chernobyl occur in Germany, the damage, according to an independent expert study, would be $2,400 to 6,000 billion -equivalent to Germany's GDP.

Capping the liability for such large-scale damage violates two vital safety tenets: the Precautionary Principle and the Polluter Pays Principle. The first says no activity with inadequately understood hazards should be undertaken. Under the second, those causing damage must compensate the victims. These principles and the absolute liability notion have been upheld by the Supreme Court of India in many judgments as deriving from Articles 21 (right to life), and 47 and 48A (improving public health and safeguarding the environment) of the Constitution.

The nuclear liability Bill violates these principles. It artificially caps total liability for an accident at 300 million Special Drawing Rights, or about Rs 23 billion, and the operator's liability at Rs 5 billion. The difference is to be made up by the Indian taxpayer. This is outrageous.

The Bill lets nuclear equipment suppliers and designers off the hook. The notions of strict liability and product liability demand that they pay damages in case the equipment is poorly designed or manufactured. Equally obnoxious is the 10-year limit to liability: many forms of radiation injury, including cancer and genetic damage, reveal themselves only 20 years after exposure.

These flaws stem from two 1960s nuclear conventions meant to promote and subsidise nuclear power when it was seen as safe and deserving of subsidy. But we now know that nuclear power is inherently hazardous, because it involves high-pressure, high-temperature processes and great energy intensity. A nuclear reactor is a complex system whose sub-systems are tightly coupled. A mishap in one sub-system gets instantly transmitted to others, potentially causing a runaway reaction. Nuclear power poses the radiation danger at every step -routinely, even without accidents. The costs of the damage, including treatment, are hard to estimate.

That's why developed countries like Germany, Japan, Austria and Sweden impose unlimited liability on the operator, supplier and transporter, etc., and often demand a $3 billion security deposit.

However, the Indian government has latched on to the 1997 Convention on Supplementary Compensation for Nuclear Damage (CSC) sponsored by the International Atomic Energy Agency, as if it enjoyed wide acceptance. It isn't actually in force yet - five states need to ratify it, but only four (Argentina, Morocco, Romania and the US) have. The IAEA's mandate is to promote nuclear power as safe and economical. It trivialises Chernobyl. The CSC follows the Paris-Vienna model and raises total liability per accident to a miserly $986 million.

The sole justification offered for India's nuclear Bill is that without a low liability cap, no foreign nuclear operator will invest in India. But this is a question-begging argument. Indians don't need nuclear power at the expense of safety or Constitutional principles. The Rs 5 billion operator liability (even if raised, according to a new proposal) won't remotely compensate for Indian lives.

The Bill represents capitulation to US and Indian corporate pressure, and a retreat from the state's responsibility to protect citizens against hazards. The US, having given India the nuclear deal, is now furiously lobbying to extract nuclear contracts for American corporations. This must stop.            

*Third World Resurgence No. 235, March 2010, pp 18-20