M. Rajshekhar 
India’s new nuclear push differs from its predecessors.
This time around, the country is not solely banking on its atomic establishment to erect an array of nuclear power plants. As the first part of this series showed, that approach resulted in cost- and time-overruns, and yielded an outcome where India missed every single deadline to boost its nuclear capacity.
In a bid to avoid that fate, India’s new nuclear push stands on two new pillars. First, Nuclear Power Corporation of India will set up reactors in partnership with other state-owned companies like National Thermal Power Corporation. Second, the country is very bullish about Small and Modular Reactors (SMRs). These, goes the thinking, will be cheaper and faster to construct.
These claims – that NPCIL’s joint ventures or JVs will result in a faster rollout, and that SMRs will be competitive against rival forms of energy – have not received the broader scrutiny they deserve.
The JV, first.
Can Ashvini deliver?
The JV between NPCIL and NTPC named Ashvini is neither new nor unique.
It dates back to February 2016 when the NDA government amended the Atomic Energy Act to enable NPCIL to form JVs with other PSUs for setting up nuclear power plants.
The idea here is that JV partners bring their investible surplus to the alliance, while NPCIL brings in nuclear know-how. The controlling stake – in what are 51:49 arrangements – would be with NPCIL.
Between then and now, NPCIL has formed JVs with NTPC, Indian Oil Corporation, and Nalco. “JVs were also mooted with Hindustan Aluminium and Indian Railways, but those have stayed at a MoU level,” Shah Nawaz Ahmad, who oversees India for the World Nuclear Association, said.
Of these, the JV with NTPC seems to have made the most headway. NTPC, which will need to replace its coal fleet with cleaner alternatives, wants to install 2 GW of nuclear energy by 2032; 4.2 GW by 2035; and 20 GW by 2050. SMRs are a part of its strategy. It wants to limit its JV with NPCIL to large reactors – and form a separate JV (where it is the majority partner) to build SMRs with private partners. According to media reports, it has assembled a 15-member team in Mumbai to drive its nuclear ambitions. Beneath these announcements lie a mass of unknowns – like the JV’s capacity to resolve the problems that have hamstrung nuclear power till now.
While NTPC can raise funds more easily than NPCIL, as physicist MV Ramana writes in The Power of Promise, India’s atomic establishment has never really struggled for cash. It has always been well-capitalised.
What about competitiveness against rival forms of energy? Unlike coal-based power plants, nuclear power has little or no running costs. Unlike solar, whose installation cost now stands at Rs 3 crore to Rs 5 crore per megawatt but panels start leaching efficiency after 15 or so years, nuclear plants have a much longer life. A better comparison is with hydel and pumped storage projects (PSPs). They, too, have high upfront costs, can be used for base load generation as well as grid balancing, and have low running costs – dams more so than pumped storage.
This is where things get interesting. At NPCIL, the cost of a nuclear power plant today ranges between Rs 12 crore-Rs 15 crore/MW. Imported reactors cost more. At Mithi Virdi, for instance, the cost per MW for Westinghouse reactors has touched Rs 40 crore.
While the cost/MW for dams and PSPs (Rs 7 crore-Rs 8 crore/MW) is lower than that of nuclear, their costs/unit are similar – Rs 6/unit for a new dam; between Rs 5/unit and Rs 9/unit for a PSP, which incurs costs while pumping water back to the higher reservoir; and around Rs 7/unit for a reactor costing between Rs 15-20 crore/MW.
“Nuclear… remains the dispatchable low-carbon technology with the lowest expected costs in 2025,” says the IEA. “Only large hydro reservoirs can provide a similar contribution at comparable costs but remain highly dependent on the natural endowments of individual countries.”
Land acquisition is one bugbear – but that applies to hydel and PSPs as well.
Technical glitches, however, are a bigger risk. “India has had hardly any standardisation of nuclear plants thus far and without standardisation nuclear cannot compete,” a senior official familiar with the government’s clean energy plans told CarbonCopy on the condition of anonymity. The country has variously imported reactors from countries like France, Russia and the USA. “That is what NTPC is trying to do now—develop standardised designs,” he continues. “India has valuable experience with the standardisation of coal boilers which greatly sped up projects. That is how the idea of Ashvini came about.”
These thoughts echo those of former environment minister Jairam Ramesh who said in an interview recently, “India should have standardised on its own design of a heavy water reactor. It should not depend on imported reactors. Two standard 700 MW reactors are coming up in Kakrapar and more elsewhere. India should standardise on those reactors. We should build on those reactors.
And so, the current expansion hinges on India’s 700 MW Pressurised Heavy Water Reactor. At Kakrapar, where this reactor was rolled out, it posted time-overruns on account of its “First Of a Kind” Systems and delays in procuring critical equipment.
At this time, it’s unclear if the NTPC-NPCIL alliance has ironed out these glitches. CarbonCopy contacted both NTPC and NPCIL for interviews. They did not respond. Subsequently, questions were emailed to both companies. This article will be updated when they respond.
Turn to SMRs and similar questions intensify.
SMRs: An idea whose time has come?
For some time now – especially once renewables’ prices began to drop – the nuclear industry has been trying to boost its own competitiveness. In the past, trying to bring down its costs, nuclear plants have bet on scale, spreading costs over a larger power output. With SMRs, the industry is heading the other way. Instead of building one large reactor, SMR makers want to mass-produce smaller reactors, and to use that scale to drive costs down. Modularity – where complete reactors are shipped out – is also expected to cut installation time and costs.
“If you take a large reactor like Kudankulam, on-site construction cost will be as much as 30-40% of the eventual cost,” said Vinod Sahni, a former director of the Raja Ramanna Centre for Advanced Technology at BARC. With SMRs, says manufacturers, installation time will drop from five or more years to less than two. The NDA echoes these claims. SMRs, it says, will yield significant savings in cost and construction time and fast-track nuclear power generation.
These claims by SMR makers have been challenged by others – including the nuclear establishment itself. Can mass production drive down reactor prices? The answer to that question lies in a thicket of deeper questions. In any manufacturing process, raw material costs account for a large chunk of costs. Efficiencies result from factors like cheaper sourcing of inputs by placing large orders; spreading the cost of the assembly plant, R&D, etc, over a larger output; and so on. And so, how many reactors will an assembly line need to produce to score meaningful savings on manufacturing costs? In a paper about the competitiveness of nuclear power, Ramana estimated the cost/MW of SMRs will match that of large reactors only after anywhere between 700 and 60,000 reactors have been manufactured.
Given that anywhere between 40 and 70 SMRs are in development across the world, how many reactors can each assembly line expect to manufacture each year? During installation, SMRs face the same risk as large reactors—of cost and time-overruns. Once installed, they produce the same externalities as large reactors (like accidents and radioactive waste disposal) and, ergo, have similar system costs as large reactors (boilers, turbines, safety systems, staff). All these costs, however, have to be recovered from lower power output. That will add to the cost per unit.
On the whole, SMRs might be much costlier than large reactors. In his talk at the India Nuclear Business Platform, A.K. Nayak, head (Nuclear Control and Program Wing) at the Department of Atomic Energy, pegged the “ballpark” cost of erecting an SMR at “Rs 200 million/MW” – or Rs 20 crore/MW.
Their running cost might be higher as well. “Given their smaller size, SMRs have a greater surface area,” said Grover. “Which means greater neutron leakage and therefore, more spent fuel.” This point has been made elsewhere as well. “A recent study published in the Proceedings of the National Academy of Sciences concluded that SMRs would create up to 30 times more radioactive waste per unit of electricity than conventional reactors”, reported Energy Monitor.
What about cost per unit? As IEEFA reported earlier this year, “NuScale’s first power plant… increased its initial estimated electricity cost from 5.8 cents per kWh to 8.9 cents per KWh. That increase [a jump from Rs 7.25/unit to Rs 10.3/unit] would have been even higher if not for $4 billion in federal tax subsidies and a $30/MWh subsidy,” added the Australian energy think-tank.
At ₹10.3/unit (or more), SMRs will be more expensive than pumped storage – which can also address intermittency and ensure grid stability. “I am not sure SMRs are a fruitful LT investment,” said author Prabir Purkayastha. “But pumped storage is. You can use it to cycle up and down. You can create these in small areas. They can do daily balancing as well as long-term balancing.”
For all these reasons, India’s nuclear community is not convinced. “Nowhere in the world have SMRs been used,” academic R. Rajaraman said in April. “We are yet to see a proof of concept. It’s all ambitious talk and exaggerated claims.”
These numbers – coupled with the absence of commercial models in the market and reluctance within the nuclear establishment—raise a large question. Support for SMR comes from elsewhere in the government. “In the DAE (Department of Atomic Energy) and Niti Aayog, some see SMRs as an opportunity,” Grover said. “They think SMRs can be deployed quickly and economically.”
As this article was being finalised, news came too that NuScale had terminated its agreement with Utah. “NuScale’s Utah plant was expected to be the first SMR to win a license from the U.S. Nuclear Regulatory Commission for construction,” reported Reuters. Concerns about uncompetitiveness were a major reason.
The question writes itself. Why is India so bullish about an untested technology?
A question of risk
At this time, the world is seeing as many as 70 SMR models under development.
“NuScale is creating a small reactor by downsizing an existing reactor,” said Ahmad. “Others are innovating new designs. In the next three to four years, an array of SMRs will be offered as the first of a kind. These will range from micro (200 to 500 kw) to small (below 300 MW).” (India has its own small reactor. This is the 300 MW AHWR300-LEU. Developed by BARC, it runs on thorium and plutonium. It, however, is not modular).
Most SMR makers are eyeing developing countries which need to decarbonise but don’t have enough solar, wind and hydel potential to quit coal. Take NuScale. It hopes to build SMRs in Romania, Kazakhstan, Poland and Ukraine.
And India. As Nayak said at his conference: India has to retrofit 22O GW of thermal capacity. “With a ballpark figure of 200 Mn INR/ MW, business has a potential of INR 44 Trillion (USD 550 Billion) over the next 20 years,” he said. Add industrial demand for SMRs and that number will swell further.
One outcome is lobbying. At this time, a bevy of foreign SMR manufacturers – chiefly Westinghouse, Holtec, NuScale, Rosatom — are eyeing India. Hardwired into their lobbying is the worry that India might relax vital safeguards and invest billions in an uncompetitive source of electricity.
Unlike thermal power plants (TPP), SMRs cannot be installed in seismic zones. This means NTPC’s TPPs around Delhi, for instance, cannot be retro-fitted. Also, nuclear reactors are required to have an exclusion zone—a 1.6 km radius—around them. Most thermal power projects, however, have towns and villages pressing up against their walls. In other words, to install SMRs, NTPC will have to either acquire land for a safety zone around power plants being retro-fitted or the government will have to do away with this stipulation.
The government is also soft-pedalling the civil liability issue. “That has been pushed to the backburner,” said the senior government official. “It is a part of the law but it is not at the centre of these discussions. It will not be a deal breaker.”
The questions write themselves. Should nuclear or renewables and pumped storage take the base-load? How much nuclear power should India build? Should it focus on large reactors or SMRs?
These are vital queries. To set up 10,000 MW of SMR capacity, India will need to spend Rs 200,000 crore. For this sum, it could set up 16,000 MW of large reactors. Or, at Rs 7.5 crore/MW, 25,000 MW of pumped storage capacity. Alternately, should the country invest in smarter grids?
The bigger question
Modern grid operators, wrote Ramana recently, are trying to create 100% renewable grids by addressing intermittency without falling back on “baseload” generation.
“The first and foremost is energy efficiency, which reduces demand, especially during periods of peak use…. A second option is demand flexibility… wherein utilities compensate electricity customers that lower their use when asked – often automatically and imperceptibly – helping balance supply and demand.”
One recent study found, he writes, that “the U.S. has 200 gigawatts of cost-effective load flexibility potential that could be realised by 2030 if effective demand response is actively pursued.” Before embarking on the nuclear buildup, India needs tno ascertain the equivalent number for itself – and then see how much additional base load it needs. The costs of not doing so are stranded nuclear plants – or expensive power for users.
Instead of having those conversations, however, India is rushing to embrace SMRs. Not only does all this draw the country’s ill-fated adventure with Enron to mind, it also highlights a larger flaw in India’s response to climate change. As journalist Joydeep Gupta had commented in a recent article for ThirdPole, secrecy pervades South Asian countries’ responses to climate change.
In the case of nuclear, India is embarking on an expensive buildup – even thinking of retooling safety norms – without much of a discussion on viability, alternatives and attendant costs.