On the heels of the pact signed earlier this month between state-owned Nuclear Power Corporation of India Ltd (NPCIL) and US nuclear vendor Westinghouse Electric for deploying six atomic reactors units in India, the Toshiba Corp-owned American reactor firm is expected to put out a revised techno-commercial proposal shortly. If negotiations proceed on schedule, the commercial contract should be signed by June 2017, while site design and engineering work is expected to start immediately for the project that has been relocated to Andhra Pradesh and will deploy Westinghouse’s new AP1000 reactors. Each of these “third-generation” reactors has a capacity of 1,150 megawatt electric (MWe).
There seem to be two impediments, though — Westinghouse’s patchy record on AP1000-based projects that are currently under construction globally and the lack of clarity on the cost viability of projects based on the untested reactor technology. Westinghouse currently has eight AP1000 reactors under construction — two in South Carolina, two in Georgia and four in China.
Incidentally, just three months ahead of India inking a broad-based atomic power agreement with the US to commence engineering and site-design work for six Westinghouse’s AP1000 reactors, there was some heartening news from China. The world’s first Westinghouse AP1000 nuclear reactor, the China National Nuclear Corporation announced in March, will go into operation in June 2017.
This, however, comes after construction delays of over three years — the original commissioning date for the project being December 2013. The reactor has been plagued by delays due to design flaws and problems with key components, including new coolant pumps for the two reactor units that arrived only at the end of last year.
The uncertainty on the pricing stems from the fact that the AP1000 is a new design approved by the US nuclear regulator (the NRC) in December 2011, which has not been completed and brought on line anywhere in the world. While nuclear power construction has been subject to significant cost escalations and regulatory delays, adding a new reactor technology only serves to make the cost estimation even tougher. While clear pricing benchmarks are not available for the AP1000, costs for two new proposed reactors at a Levy County site in Florida, US are reported to have surged from $3.5 billion to $22 billion in five years since initial estimate, translating into a project cost of about Rs 1.45 lakh crore (at an exchange rate of Rs 66 to a US$). This would entail a capital cost of over Rs 70 crore per MWe.
In India, the cost benchmark for projects based on the indigenous Pressurised Heavy Water Reactor (PHWR) technology is about Rs 9 crore per MWe. For new imported light water reactors such as the AP1000, costing benchmarks are derived broadly from the two Russian designed VVER-1000 reactors which are to be deployed at Kudankulam. The two new VVER reactor units (KKNPP 3 & 4) to be set up in Tamil Nadu, which would come up at the Kudankulam site where two identical units (KKNPP 1 & 2) are nearing commissioning, entail a sanctioned project cost of Rs 39,849 crore for the two new reactors.
With the cost uncertainties looming large, the Government has initiated two tangible steps to ensure the cost viability of the project, the first involving an American firm since India and the US signed a landmark civil nuclear deal in 2008. Next, the Department of Atomic Energy has formed a panel to thrash out the broad financing pattern for the six Westinghouse reactors and a finance negotiating committee has been tasked with getting the details of the funding package with the US Exim Bank, which has agreed to part-finance the project.
In case of the Westinghouse project, the original site, Mithi Virdi in Gujarat, had to be abandoned in favour of Andhra Pradesh, where land acquisition is at an advanced stage. This site was originally earmarked for a project based on GE-Hitachi’s new ESBWR reactor based project.
While the land acquisition was stuck for over half a decade, NPCIL is learnt to have made a down payment on 2,000 acres of land in the coastal district of Srikakulam in Andhra Pradesh. A majority of the farmers in the area have agreed to sell their land and the entire land acquisition is expected to happen over the next few months, according to assurances handed out by the state government to NPCIL.
DAE officials indicated that it would be wrong to set the project cost benchmark based on the cost of projects to be implemented in the US. A better cost indicated, they said, would be ongoing projects in China, where local manufacturing of parts promises to drive down overall costs. “The Chinese project cost would have better correlation to the costing being worked out for Indian project,” an official said.
In China, though, reworking quality-related niggles have repeatedly led to the implementation schedule being derailed. The startup of the first AP1000 unit, Sanmen-1, initially expected in 2014, has been delayed by issues related to components, including the sealed reactor coolant pumps built by component manufacturer Curtiss-Wright Corp. The pumps used, touted as the largest of their type ever built for a reactor, are reported to have developed heating issues and then faced a problem with bearings that required them to be shipped from China for refabrication. Then a new problem, having to do with retaining clips in the pumps, was identified and addressed subsequently.
Added to the worries is the news surrounding the poor financial health of Westinghouse’s parent firm, Toshiba. The Japanese conglomerate has slashed thousands of jobs in the nuclear business in the wake of a series of setbacks that started with the Fukushima Daiichi nuclear disaster in 2011, alongside competition from natural gas plants and a crash in the appetite for nuclear plants. Toshiba wrote down $2.3 billion of Westinghouse’s value in April.
In less than two years after the Fukushima meltdown, Toshiba and Westinghouse Electric lost contracts at 60 reactors. Between 2012 and 2014, according to Toshiba, Westinghouse’s cumulative operating losses reached $1.43 billion.
There could be yet another stumbling block — the fact that Japan continues to hold out on signing a nuclear deal with India. This could be a crucial determining factor as the two US reactor vendors, as well as a range of other global nuclear reactor manufacturers, source the most critical equipment in a reactor — the calandria or reactor vessel — from Japanese heavy forging major Japan Steel Works (JSW).
Experts say that the transfer of Japanese technology to India for civilian use requires a nuclear pact, but Tokyo has so far desisted from initiating one as India has not signed the Nuclear Non-Proliferation Treaty (NPT).
Globally, apart from Japan’s JSW, there are just three major heavy forging capacities in operation worldwide that can fabricate large single-piece pressure vessels for nuclear reactors — Creusot Forge (a subsidiary of France’s Areva group), Russia’s OMZ Izhora and Chinese China First Heavy Industries. But JSW dominates the nuclear forgings business, accounting for nearly 80 per cent of the market for large forged components for nuclear plants. Also, Toshiba-Westinghouse’s AP1000 pressure vessel closure head and three complex steam generator parts can only be made by JSW. Globally, all the suppliers are booked with orders for at least the next five years as the most critical issue for accelerating nuclear power plant construction is the availability of heavy engineering plants to make the reactor components, especially for units of over 1,000 MWe.