In May 2017, India’s Union Cabinet approved the construction of ten 700MWe Indian Pressurised Heavy Water Reactors (IPHWR-700s) in ‘fleet mode’. This decision came just before India’s state-owned nuclear utility, the Nuclear Power Corporation of India Limited (NPCIL) concluded a ‘General Framework Agreement and Credit Protocol’ with Russia’s ‘Atomstroyexport’ (ASE, a subsidiary of Rosatom) for units 5&6 of the Kudankulam nuclear power plant.

Both decisions will help India consolidate its nuclear construction industry in different ways. However, these new-build plans also reveal that there are now two pathways for India to boost its nuclear generation capacity in the short to medium term – one based on domestic IPHWR-700s and the other centred around Russian origin VVERs. Other avenues for international collaboration are not really on the table at the moment, given supplier wariness about India’s nuclear liability law and high capital costs and general turmoil in the global nuclear industry.

India is also likely to commission the Prototype Fast Breeder Reactor (PFBR) located at Kalpakkam by next year and a Cabinet nod for constructing two 600MWe commercial fast breeder reactors is expected to follow.

IPHWR-700s – the future mainstay

With Cabinet approval, NPCIL, which is controlled by the Department of Atomic Energy (DAE), will set into motion activities at the various sites where the ten IPHWRs will be built. The sites for these units are to be Mahi Banswara in Rajasthan (units 1,2,3 &4), Chutka (units 1&2) in Madhya Pradesh, Gorakhpur in Haryana (units 3&4), and Kaiga (units 5&6) in Karnataka. All of these are inland sites, in keeping with India’s aim of building new IPHWRs in the hinterland while reserving the coastline for larger imported light water reactors.

Since project activities are already underway for Gorakhpur 1&2 and ‘in-principle’ approval for constructing two IPHWR-700s at Mahi Banswara has already been granted, this latest decision is a move to consolidate domestic IPHWR-700 build plans by looking to construct a fleet of reactors in one go, instead of the usual practice of giving piecemeal orders.

At the moment, four IPHWR-700 units are under various stages of construction in India. Of these four units, the most advanced is Kakrapar 3 in Gujarat, which is currently undergoing cold hydrostatic tests. Hot fuel commissioning is expected to commence by August 2017. The reactor is expected to reach criticality by November 2017 with commercial operations likely to start by early 2018. Its twin unit, Kakrapar 4, is supposed to begin commercial operations 6-7 months later.

The other two IPHWR-700s presently under construction are Rajasthan 7&8 in Rawatbhatta, Rajasthan. All major equipment for these two plants has already been set up and these reactors are likely to be commissioned in the 2019-20 timeframe.

In fact, construction at all of these units is actually experiencing time over-runs. At the time of Cabinet approval in 2009, it was believed that these units could be built in 66 months flat. However, that has turned out to be rather ambitious and each of these units is delayed by about 2-2.5 years with respect to expected completion dates at the time first concrete was poured. The delays have chiefly been on account of slow delivery of supplies and issues related to fine-tuning the IPHWR-700 design itself.

Nevertheless, the decision to build ten IPHWR-700 units in fleet mode does reveal a certain confidence in this design going forward. It seems that NPCIL’s membership of the CANDU Operators Group may have also come in handy in sorting out some of the residual issues with the design, especially now that India and Canada have a nuclear deal in place.

Though the IPHWR-700 design retains many features of the baseline IPHWR-540 units, such as two diverse and fast-acting shutdown systems, double containment reactor building, water-filled calandria vault and integral calandria/end-shield assembly, it also exhibits certain key improvements.

An uprating from 540MWe to 700MWe has been achieved by allowing partial boiling at the coolant channel outlet, even though the number of coolant channels stays the same as before. The design also has enhanced safety features such as the interleaving of primary heat transport system feeders, to reduce the core void coefficient and minimise reactor over-power during a loss of coolant accident (LOCA) or core meltdown. The IPHWR-700 also incorporates for the first time in IPHWR design history a passive decay heat removal system, regional over-power protection, containment spray system, mobile fuel transfer machine, and a steel liner on the inner containment wall. The passive decay heat condenser is capable of removing up to 3% decay heat.

Perhaps the greatest attraction of these reactors is that they can be wholly built via domestic supply chains. Indeed, the ‘fleet mode’ decision is driven by the need to place sizeable orders with Indian industry to achieve economies of scale and ensure that delivery timelines can be met, unlike in the past. The Indian government expects that the construction of these ten IPHWR-700s will generate orders worth $11 billion for Indian industry and result in the employment of some 33,400 people. The ‘fleet mode’ aims to break the ‘capacity creation vs existing orders’ conundrum that plagues any heavy industry and ensure that hard-won manufacturing ability does not atrophy due to long lean periods.

The chief beneficiaries of this new plan are expected to be companies like HCC, which is the lead engineering, procurement and construction (EPC) firm for Kakrapur and Rajasthan, as well as heavy engineering major Larsen & Toubro (L&T), which has supplied major heavy forgings based equipment such as steam generators for the same units.

Overwhelmingly domestic sources of supply are expected to keep capital costs for these reactors under control, although these IPHWR- 700s are likely to be more expensive than the IPHWR-700s already under construction. While each of those units will end up costing over Rs61.60 billion (taking Kakrapur 3 as a benchmark), the average build cost for the new units will be around Rs82.75 billion (based on the projected cost for Chutka 1&2). With these figures, IPWHR-700s are going to cost about $1850 per kilowatt to build at current exchange rates. In terms of levelised cost of electricity (LCOE) estimates, these plants will be competitive with new coal-fired units.

Presently, NPCIL operates a fleet of two IPHWR-540s and 14 IPHWR-220s, besides two smaller PHWRs in Rajasthan, one of which is now essentially a research reactor. The total number of new IPHWR-700s approved for construction is now twelve to add to the four already being built. Taken together, these 16 new IPHWR-700s will add 11,200MWe (gross) or 10,080MWe (net) to the Indian grid system.

Up to 40% of their fuel could be slightly enriched uranium (SEU) – with about 1.1% U-235, to achieve a higher fuel burn-up of about 21,000 MW-day per tonne (MWd/t) instead of around 7000MWd/t. Initially, the required SEU can be imported given that some of these reactors will be placed under safeguards, but by 2025 India’s new gas-centrifuge enrichment plant in Chitradurga in Karnataka is expected to supply the fleet’s needs.

The Russian option

The IPHWR-700 would be classified as a medium-sized reactor by the IAEA. For true large capacity reactors, Russian pressurised water reactors (PWRs) or VVERs are currently the only game in town as far as the Department of Atomic Energy is concerned.

In Tamil Nadu, two 1050MWe (gross) AES-92 units, Kudankulam 1&2, are currently operational while Kudankulam 3&4 (which are also AES-92 designs) are under construction. In June 2017 NPCIL signed a general framework and credit protocol agreement (GFA) with Atomstroyexport for two more AES-92 units at the site (Kudankulam 5&6.) With the GFA, the contract negotiations for the two units has entered what Rosatom calls the ‘practical phase’ with ‘the obligations of the two sides, costs and other important conditions of their cooperation’ being delineated, apparently setting the stage for the project to take-off.

Overall, the Kudankulam project seems to have stabilised now despite a huge delay and a variety of problems, not the least of which were massive protests in 2011-12 when Kudankulam 1 was undergoing pre-commissioning tests. The first two units were also delayed due to issues related to the turbine sets supplied by Russia’s Silmash. In addition NPCIL’s engineers, who were mostly conversant with PHWRs, took time to absorb Russian PWR technology, and only 80 Russian supervisory staff were present at the site at any given time.

Owing to the significant delay in their commissioning, Kudankulam 1&2 ended up being some Rs90 billion over budget. Subsequently, haggling related to the vendor liability sub-clause in India’s Civil Liability for Nuclear Damage Act 2010 (CNLD 2010), with ASE demanding India provide re-insurance for future reactors, meant that India and Russia signed contracts for Kudankulam 3&4 only at the end of 2014 and for a project cost of Rs398 billion, which is more than double what was agreed for the initial two units. The sharp escalation in build costs was in no small measure due to liability issues. So while Kudankulam 1&2 are supplying power at around the Rs4.0 per kWh mark, it is unlikely that the power from Kudankulam 3&4 will cost anything less than Rs6.0 per kWh when they are ready for commercial operations sometime in the mid-2020s.

It seems that Kudankulam 5&6 will be even costlier, with the contract for building these two reactors expected to be worth Rs500 billion ($7.76 billion at current exchange rates). The Indian side is keen that costs do not escalate and that the spend also ends up boosting India’s nuclear industry. So the final contract for these units is linked to the implementation of the ‘Programme of Action for Localisation in India’ signed by India and Russia on 24 December 2015, to increase the domestic value content of Russian VVERs built in India.

India hopes that local content will rise to as much as 60% by the time Kudankulam 7&8 are built. Incidentally, Kudakulam 7&8 will be 1200MWe AES-2006 units, taking the total gross capacity envisaged for Kudankulam to nearly 10GWe. All Russian reactors built in India will come under safeguards with Russia guaranteeing the supply of enriched fuel. India has the ‘right to reprocess’ the spent fuel from these reactors. India and Russia are also in talks to set up a fuel fabrication plant for Russian origin reactors on Indian soil.

Since NPCIL now feels confident about Russian VVERs, and the fact that the Russians offer substantial credit lines for their reactors in addition to the promise of localisation, another six 1200MWe AES-2006 units are being projected for construction at Kavali in Andhra Pradesh on India’s East Coast. Kavali is an alternate site that is being offered to the Russians following the West Bengal state government’s reluctance to allow the already surveyed Haripur site be used to set up a nuclear plant.

Other options?

Unlike the Russian VVERs, other foreign designs are unlikely to be built in India any time soon despite the projections that were made at the time of India’s waiver from the Nuclear Suppliers Group. India’s CNLD is still a major deterrent for Western and Japanese suppliers, despite New Delhi instituting an Indian Nuclear Insurance Pool in June 2015 to serve as a workaround to supplier liability.

For example, even as EDF submitted a revised proposal to NPCIL in July 2016 for six EPR-1650s to be built at Jaitapur In Maharashtra, it sought guarantee of ‘the same level of protection’ in relation to liability that is available at the international level, while citing the Vienna convention on liability. Earlier in September 2015, GE-Hitachi also said that it would not proceed with any project in India unless the latter’s liability regime was consonant with international standards. Toshiba-Westinghouse has said in the past that it would wait to supply equipment to India till such time the country ratified the Convention on Supplementary Compensation, something that has now happened.

But liability is not the only issue getting in the way of these reactor types being built in India. Even though both coastal zone and environmental clearances are in place for the Jaitapur site, DAE is now rather wary of building this type given the high LCOE projections that are being made on the basis of estimated project costs for EPR-1650 construction in India, even with appreciable localisation levels. LCOEs north of Rs12.0 per kWh are being projected and this would be prohibitive in India’s increasingly liberalised power market. In fact, DAE had earlier sought to obtain a commitment from Areva that the cost of power from Indian EPRs would not exceed Rs7.0 per kWh. There are now also doubts about the quality of certain French- origin forgings after reports about them from Flamanville in France and Taishan in China.

The prospects for GE-Hitachi’s Economic Simplified Boing Water Reactor (ESBWR) to be built in India are dimmer. In June 2016, DAE stated that it would not support the construction of any reactor design in India that did not have a reference plant, basically ruling out the ESBWR for the foreseeable future. At the same time, DAE re-assigned the Kovvada site in Andhra Pradesh, which was earlier reserved to host GE-Hitachi-supplied reactors, to Toshiba-Westinghouse’s AP1000 design.

The allocation of Kovvada to Westinghouse happened due to land acquisition issues ruling out the Chhaya-Mithi Virdi site in Gujarat that had been originally earmarked for building AP1000s. However, the AP1000’s prospects for construction in India have also fallen, due to Westinghouse filing for Chapter 11 bankruptcy in the USA. Though Westinghouse insists that the Kovvada project is ‘structured in a manner that does not include construction risk’, DAE insiders are now sceptical about Westinghouse’s staying power for this venture. No commercial contract has followed the submission of a techno-commercial offer by Westinghouse in 2016 for building six AP1000 units at the Kovvada site, thereby making Westinghouse’s plan to start the project in 2018 very unlikely.

Capital cost is also an issue for the Kovvada project. India fears a major escalation in capital costs as the project gets stretched out. The massive increase in estimated project costs for the AP1000 build proposals at the Levy County site in Florida over a period of just five years are adding to these fears. DAE has formed a panel to flesh out the broad financing for the six AP1000s and has tasked a finance negotiating committee with getting the details of a possible funding package from US Exim Bank, which has agreed to partly finance the Kovvada project. DAE is also concerned about the delays that have plagued the commissioning of the first Chinese AP1000 unit, Sanmen 1.

Given these developments, it is seeming more and more likely that the first non-Russian LWR to be built in India will be the indigenous 900MWe Indian PWR or IPWR, whose design is now complete. The IPWR has the support of India’s heavy industry and boasts generation III+ safety features including a core-catcher. Indian industry is keen to build up capacity to support this design and DAE is confident of sourcing fuel from both domestic and international sources, since it is likely to be placed under safeguards.

Second stage

What is not likely to be placed under safeguards anytime soon is the 500MWe PFBR awaiting criticality at Kalpakkam in Tamil Nadu. India’s first medium-sized FBR is being built by Bharatiya Nabhikiya Vidyut Nigam or Bhavini, which is a DAE-controlled utility specifically set up to manage India’s future fleet of FBRs. Bhavini expects that the PFBR will attain first-criticality sometime in late 2017 and the reactor will be fully commissioned by mid-2018.

Commissioning of the project, on which about Rs54.00 billion has been spent thus far, has already been delayed by almost seven years. The delay to some extent is on account of Indian scientists looking to ‘get things right the first time over’, given typical fears of a sodium coolant leak from a liquid-metal cooled FBR of this design and the negative publicity such an event would entail for the second-stage of India’s three-stage nuclear programme.

Once the PFBR attains criticality, a Cabinet nod is expected for two 600MWe CFBRs which will boast a better breeding ratio and enhanced safety features – besides being easier to produce as a more standardised design.

For the time being, however, India is banking on IPHWR-700s and Russian-designed VVERs to boost nuclear generation capacity to 25,000MWe by 2030 at a time when India is looking to chart a carbon-light growth path for itself.