Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Ring of Steel: France's Iter Nuclear Fusion

A world-first visionary nuclear fusion project that transcends other construction projects with its multi-national collaboration effort demands a suitably forward thinking plan for delivery. Alexandra Wynne visits the complex project in the south of France to see how it is benefiting from clever digital delivery tools.

Collaboration and integration are two words whose increasingly frequent appearance in the world of digitally engineered infrastructure could ensure they are consigned to the corporate-speak scrap heap. But visiting the International Thermonuclear Experimental Reactor (Iter,) project it becomes all too obvious why both are so vital to the ultra-ambitious scheme’s successful delivery.

NCE first visited Iter, which also means “the way” in Latin, three years ago when construction had got underway and the decades-long hope of solving the world’s future energy conundrum had begun to come to fruition.

It is a major international experiment involving 35 nations and aims to uncover a potentially limitless energy resource by heating frozen pellets of deuterium and tritium isotopes to 150M°C – or 10 times the heat of the Sun. If it works, it will provide a potentially limitless source of energy that has huge benefits over nuclear fission because most of the radioisotopes in any waste will have a half life of less than 10 years. Conventional nuclear fuel decays over thousands of years.

The challenges are not to be underestimated - the project has suffered rising costs and delays. The original hoped for nuclear fusion start date of 2019 has slipped to what is more likely to be around 2024 and the construction cost of €15bn (£11bn) is significantly higher than the much earlier 2001 price tag of £3.7bn. But for the Iter Organization (IO), which is running the project, much of this is completely justified given the significant design work required in the subsequent years and the necessity of stringent regulatory oversight by French nuclear safety organisation ASN.

This project is not just influenced by scientists, it’s influenced by regulatory approval, and regulatory approval

Roger Holt, Engage

Under new director general Bernard Bigot, IO is determined to get the project back on track. “We currently know we are behind schedule,” said IO spokeswoman Sabine Griffith during NCE’s March visit. “We don’t know by how long because everything is being reviewed as we sit here. [Bigot] is coming up with a new schedule. The last estimate was to have the first plasma by 2023. But now it could be 2024. The current management has decided to bring it back on schedule.”

Anyone familiar with new nuclear power schemes might assume there is something familiar about the site in Cadarache in the south of France – the expanse of the space being filled; the baffling number of operational, safety and construction buildings to be built and massive cranes dotting the landscape. It has them all.

Digital engineering on Iter

Digital engineering is smoothing the path on France’s visionary Iter nuclear fusion project

But there is a great difference. This is an experimental, research facility that has perhaps more in common with other world-first projects like the God-particle searching Cern laboratory in Switzerland or the International Space Station – all represent in some way a great leap of faith.
Fusion has been done before: notably in the 80m3, 16MWh Joint European Torus (Jet) machine at Culham in Oxfordshire. “But the big breakthrough is up to us to do,” says Griffith.

What is being built at Cadarache is a much larger version of the Tokamak reactor at Jet. The 30m wide by 30m tall machine will weigh 23,000t – or roughly three times the Eiffel Tower – and will be able to hold 840m3 of plasma and is intended to be capable of generating 500MW from an input of 50MW. This ambitious project remains an experiment and any future commercial scale development of the technology, including a demonstration project, hinges on the success of Iter.

The 30m wide by 30m tall machine will weigh 23,000t – or roughly three times the Eiffel Tower

Housing the Tokamak is a vast and intricately designed complex. Appointed in 2010 to an eight-year, £110M contract as architect engineer, the Engage consortium comprises Atkins, France’s Assystem and Egis and Spain’s Empresarios Agrupados. It is responsible for design work that includes 18 reinforced concrete buildings, 16 steel framed buildings and myriad associated infrastructure.

One of the most crucial features of Engage is its integration team. It works at the interface between the physicists, whose machine needs housing, the contractors building the complex, IO and the project’s delivery organisation Fusion for Energy (F4E).

Engage’s Tokamak integration team leader Frederic Laugier makes a clear distinction between his role and that of a designer. He says his role is paramount in detecting clashes between what has been designed and what is being built – along with managing requirements of the physicists and other interested parties.

Yes, the 42ha site for the new facility is vast. But that does not mean the scheme is not challenged by space constraints. Not only is there a keenness for Engage to come up with an efficient design, but accommodating strict safety requirements and extremely intricate requirements for positioning the Tokamak all add to the potential for mind boggling clashes.

When they arise, Laugier points out that a contractor’s first response is to seek more room to ensure all these details can be built. But most are denied. “The first cost in a nuclear project is the civil works so you wouldn’t make that bigger,” he explains.

There are some very large pieces of machinery that need to be fitted into the complex So we need to design a simple structure

Roger Holt, Engage

Some of the clashes are “normal” and “temporary” throughout installation and are resolved swiftly on site. The key is in interpreting the model. “I’m not a designer, but I scrutinise the model and the whole site is 3D,” he adds. “I’ve never seen a project where someone says: ‘The building is too big, I need less space’. Contractors often come here to say ‘I have a clash, I need more space’. But they need to be educated – the clash is a problem that the integrated whole team needs to solve. I say to them: ‘So then what is your solution, what are the options?’”

Reimagining the complex scheme so that it becomes as simple as possible is central to the design principles. “There are some very large pieces of machinery that need to be fitted into the complex,” explains Atkins principal engineer and design manager for Engage Roger Holt.

“So we need to design a simple structure that not only can support the loads that are applied to it, but we need to also be able to put large temporary openings into the sides of the buildings and their slabs to allow the fit out of machinery as the complex grows.”

And growing it is. The main construction activity has been well underway since the award of a £170M contract to the VFR joint venture of French firms Vinci and Razel-Bec and Spain’s Ferrovial in late 2012. This covers 11 buildings including the Tokamak complex.

A 3D Catia model is the other key player in the project – providing all the data and visual models that are helping the integration team uncover potentially derailing clashes.

Who’s who

Around 550 people work for the Iter Organization which represents the interests of the 35 nations, speaking 40 different languages, leading the project.

Added to those are 450 staff working for contractors, consultants and experts. The project is run and funded by seven member entities – the European Union (EU), India, Japan, China, Russia, South Korea and the United States – but Europe has responsibility for the delivery of the project’s construction and its costs, equating to almost half of the project.

Fusion for Energy (F4E) is the European project promoter representing the 28 EU member states plus Switzerland. “We take for granted how much collaboration is going on,” F4E information and communication group leader Aris Apollonatos says of the international relationships.

Engage (the UK’s Atkins, France’s Assystem and Egis and Spain’s Empresarios Agrupados) is the architect engineer for the buildings, cranes, infrastructure, power and associated buildings and processes. It is responsible for the design and support procurement and construction supervision.

Around 96% of contracts are now out with industry. Since 2012, main construction activity has been led by the VFR consortium (France’s Vinci and Razel-Bec with Spain’s Ferrovial) but Vinci subsidiary GTM Sud carried out the complex early works that involved excavation and preparation of the Tokamak Complex foundation slab.

The scheme had attracted interest from companies wanting to work on it, but Apollonatos says that more intriguing thing is that many firms have been nervous of it. “It is a project that people are initially afraid to get involved with,” he says. “But it opens up such opportunities – with the opportunity to benefit from technology improvement, for example.”


The 120m long by 80m wide Tokamak complex is without doubt where the majority of these clashes pose the greatest challenge. As expected with any nuclear facility, seismically isolating the reactor is vital. The seven-storey, 60m tall structure is recessed 17m below ground level and has a dual base slab system that contains 493 anti-seismic bearings. Vinci subsidiary GTM Sud carried out excavation work between 2010/11 in readiness for its follow on work to create the foundation and basement slab sandwiching the bearings.

BIM model: Computer generated image of the main Iter reactor building

BIM model: Computer generated image of the main Iter reactor building

The first 1.5m thick concrete slab forms the basis for the 1m2, 1.8m high concrete bearing plinths on which the 180mm2 steel isolation plates sit. Above this is another 1.5m thick slab – the B2 level – forms the true 9,600m2 first level of the Tokamak complex.

It was at this point that the team faced the first test of its integrated and collaborative process.

“Both the high reinforcement density and the tolerance of accuracy present challenges in construction,” explains F4E deputy head of site, buildings and power supplies Romaric Darbour.

The high reinforcement density and the tolerance of accuracy present challenges in construction

Romaric Darbour, F4E

There is an extraordinary volume of rebar demanded – some 3,600t encased in 14,000m3 of concrete form the B2 slab – all vital for supporting what will be 400,000t of pressure bearing down from the building and equipment above.

“We’ve got quite a high rebar density in there,” says Holt. “If you look at the forces we’re having to take from the [Tokamak] machine, it’s around about 26M.N in a vertical direction, up to 8M.N radially and a horizontal overturning of about 8M.N. So you can understand that this structure, not only does it support the weight of the Tokamak machine, but it’s also got to be designed to take those forces.”

The contractor, under Engage’s supervision, 3D modelled each piece of rebar to detect clashes and iron them out. Each piece of rebar is typically 50mm in diameter and up to 15m long, so significantly or repeatedly adapting rebar layout on site was not only logistically inefficient but the precision demanded in positioning would not allow it.

While the modelling gave a great reassurance to many in the team, forced delays came about as the regulator sought further evidence of the robustness of the 400kg/m3 concrete and its reinforcement.

“This project is not just [externally] influenced by scientists, it’s influenced by regulatory approval, and regulatory approval is never on a dedicated timeline,” explains Holt. “I think we’ve managed the timeline pretty well - recently the only slippage we’ve seen is from a regulatory requirement to substantiate part of the building… It’s a challenging environment.”

As a result of the 3D modelling and involvement from the regulator, a full scale mock-up of a section of the basement reinforcement was constructed on site in March last year.

“Why we have done that is first to detect any clashes and also to demonstrate that the construction is feasible,” says Darbour. “It demonstrated the constructability to us, but also to the regulator.”

“We are walking down this road hand in hand with the nuclear regulator – it’s the first time they have had to license a nuclear fusion project,” explains Griffith.

The “Iter-grade” concrete is the same for the bearing plinths – and tests were carried out over 20 times to ensure its stability, water permeability and confinement properties.

Following on from the iterations of digital design and physical trials, site workers completed the B2 slab in 15 concrete pours between July and August last year.

The next part of the elaborate design starts with the floor slab but doesn’t end there. Holding the vast and multi-component Tokamak machine in place while considering its insitu assembly and how it will operate is yet another mind-bending design and build task.

This is where interaction between the physicists and engineers on such a scale has been another first – for both parties.

For IO physicists, like Steve Lisgo, it has been a one-off experience to see the experiment begin to come to life while working closely on a large scale alongside engineers who have to influence how the machine might be put together and how it will operate in its complex building.

“Once we start using [tritium], humans can’t go into the machine,” Lisgo explains. “From then on robots have to do the work [of maintenance and refurbishment].”

Understanding this constraint, and the fact that the installation of any element within the Tokamak complex will be made more difficult during operation, has been carefully catered for.

Deeply anchored into the centre concrete of the B2 slab is a large, circular steel plate that the machine will be founded on. Confinement of the reactor is founded on the “Russian doll” principle, where the most highly radioactive materials are contained in the very centre – in this instance in the doughnut-shaped, hermetically sealed vacuum chamber.

Tokamak reinforcement

Growing structure: Reinforcement for the support for the Tokamak structure rise from the foundation slab

Multiple protective layers of steel and then a 2.1m thick cylindrical concrete wall form a bioshield that completely surrounds the Tokamak for vital protection.

“One of the key elements we have in the Tokamak complex … is that we will not be able, after construction, to drill holes and install equipment as we would do in a usual building,” Darbour explains. “So in anticipation of the construction we have to know exactly where each of the pieces of equipment to be installed will be.”

“If you drill, you run the risk of running into rebar if you’re in your ultimate confinement boundaries,” elaborates Holt.

We will not be able, after construction, to drill holes and install equipment as we would do in a usual building

Romaric Darbour, F4E

So before concrete pours were completed, embedded plates had to be set into precise locations in the design.

Once the buildings are in place, equipment and machinery can then be welded to these plates, which mostly range from 100mm2 to 900mm2, and are on average 300mm2 – there are some that are even larger but need a more bespoke design.

If the scale of rebar and concrete appears overwhelming, the number of steel plates does not disappoint. Around 2,500 are embedded deep – around 1.2m – into the B2 slab and as the project advances, so does the potential for more plates that will go into walls and floors of the complex.

Estimates by some of the team put quantities of steel plates at around 80,000 to 90,000 - but the design team already expects that number will breach 100,000. All of the plates are being carefully positioned as part of the digital engineering job too – a critical role given that the tolerances while mostly around plus or minus 20mm can be as tight as plus or minus 2mm.

Construction milestones

June 2011 Start of main construction
2010-2011 Excavation for the 17m deep Tokamak complex basement area.
2011-2012 Some 493 anti-seismic bearings positioned on concrete plinths on foundation base mat
July 2014 Slab poured for B2 basement slab that forms the first floor of the Tokamak complex
September 2014 First columns erected for assembly hall building
May-July 2015 Basement walls due for completion
Summer 2015 Assembly hall roof lifted into place
End 2015 Installation of crane rails in assembly hall
End 2015 Construction of nine other buildings begins
June 2016 B1 level, above B2 level, concrete pours due to start
February 2017 Civils construction design complete
March 2019 Main civil work complete (building weather tight)
December 2019 Hand over of buildings


Adjoining the Tokamak complex is possibly the most significant steel structure on site. The 60m tall Tokamak assembly hall starts at the ground floor just at the point at which the Tokamak complex comes out of its basement recess, so the roofs of both will be level. Erection of the first columns began last September – each is between 12m and 18m long, and typically weighs 30t.

Five elevations of the columns, representing 6,000t of steel will be needed for the expanse of space in which the Tokamak can be assembled. A 900t roof is due to be raised in the next few months.

Iter Assembly Building

Source: Iter IO

The steel frame assembly building

Here work is “progressing well”, according to Darbour and the aim is to begin installing the dual crane rails that are so vital in ensuring the build of the Tokamak happens. Suspended from these rails will be a 1,500t crane needed to carry the machine elements from assembly into their final position in the adjoining Tokamak complex.

Summing up the epic evolution of this scheme, Holt says the collaboration has become increasingly successful as time goes on. “As we’ve grown our relationship with the contractors and IO and F4E it’s becoming quite painless,” he says. And as new contractors come on board we will take what we’ve learned to them.”

If [the start of operation] slips to 2025 if will be very hard to keep our stakeholders and governments on board

Sabine Griffith, Iter Organization

The objective is to deliver the civils and building work by 2019 at which point the project will be back in the hands of the scientists.
Once built, the Tokamak is expected to begin operating by 2024 with the hope that fusion will be delivered between 2040/50.

“If [the start of operation] slips to 2025 if will be very hard to keep our stakeholders and governments on board,” warns Griffith. “We have one more bullet in our gun … We believe it can be done.”

Tokamak transportation

Tokamak components will arrive from all the international partners and the heaviest shipped via the Mediterranean Sea.

They will be transported along 104km of specially modified road known as the Iter Itinerary.

The dimensions of these components are impressive: the heaviest will weigh nearly 900t including the transport vehicle; the largest will be approximately four storeys — or 10.6m — high. Some will measure 9m across, others 33m long. The first exceptional load, an 87t transformer provided by the US and manufactured in Korea arrived in January.

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Please note comments made online may also be published in the print edition of New Civil Engineer. Links may be included in your comments but HTML is not permitted.