First Light: hoping to avoid a false dawn22 August 2018
UK-based start-up First Light is working on a novel approach to fusion that employs an inertially confined plasma combined with projectile-generated shockwaves as the driver, rather than the high-power lasers typically envisaged in inertial fusion. James Varley reports.
THE CONVENTIONAL WISDOM USED TO be that fusion could only be done, if at all, at huge scale, and this was seen as potentially its fatal flaw. In recent years, however, we seem to have seen a proliferation of projects focusing on “compact” fusion concepts, which, their proponents say, can be demonstrated and implemented at more manageable scales.
It’s very hard of course to judge the relative merits of the various approaches being proposed, but that being pursued by First Light, a UK based start-up, spun out of the University of Oxford in 2011, looks promising. The company, founded by Prof Yiannis Ventikos and Dr Nicholas Hawker, who is CTO, has raised £25 million to date, achieved “first plasma” (in 2015), and is now focused on “first fusion”, perhaps in 2019.
Alone among fusion start-ups, First Light is working on a purely inertial (as opposed to magnetic or hybrid inertial/ magnetic) plasma confinement concept, and is proposing radical new ideas aimed at inherent simplification of the technology and therefore cost reduction. In particular, instead of using laser energy as the “driver”, for initial heating and creation of the plasma, it proposes to use shockwaves, an idea inspired, bizarrely, by the pistol shrimp, which stuns its prey with a plasma-creating shockwave, and which First Light describes as “the only example of inertial confinement on earth.” It was the starting point for Nicholas Hawker’s PhD thesis. In the fusion device, the shockwaves will be provided by metal projectiles electromagnetically accelerated to high velocity – over 20km/s (about 45,000mph) to demonstrate fusion and around the 100km/s mark to achieve “gain” (maybe 2024 or thereabouts).
Temperature, density, time
As with most current fusion concepts, the basic process to be employed in the First Light reactor is the fusion of deuterium and tritium. The deuterium will be extracted from water and the tritium bred in a lithium blanket subjected to neutron bombardment. “This is the easiest fuel to use by a factor of ten”, says Nicholas Hawker. “Other fuels are possible, but they make the plasma physics much more difficult.” He notes in passing that the sun, which fuses hydrogen, a more difficult fuel, is “not very good at fusion” but “makes up for its poor performance by being enormous.”
To make fusion happen, the right combination of temperature, density and time is needed. The temperature must be such that particle velocities are high enough to overcome the electrostatic repulsion between positively charged nuclei, allowing the nuclei to get close enough for the strong force to come into play and fuse them. The nuclei must be close enough together to be likely to collide (ie there must be sufficient density), and the reaction must be sustained for a sufficient time for a useful amount of fusion to take place.
The temperature is “a kind of absolute” says Dr Hawker, but “density and time are interesting, they can be traded off to a huge degree.” He estimates that across the range of fusion technologies currently being proposed there is a factor of about a billion difference between the extremes of density and confinement time combinations, so “there’s a whole spectrum out there for those two parameters... Magnetic fusion [as exemplified by ITER and JET] holds the plasma for a very long time but the density is vanishingly small. In inertial fusion [as at the LLNL National Ignition Facility, which employs the world’s most energetic laser] the plasma exists only for nanoseconds but is incredibly dense. First Light is proposing a new idea for inertial confinement. Our scheme holds the plasma for a hundred times longer than mainstream inertial fusion, meaning we need less density.”
Dealing with instability
The challenge for mainstream fusion schemes, and “indeed with all fusion ever” is stability, says Dr Hawker, and finding a configuration that holds the fuel in a hot and dense enough state for a long enough time. Small deviations from the idealised geometry, or small variations in intensity of the driver, result in instabilities. These grow, and ultimately harm the expected performance, he explains. “Basically, every experiment that’s ever been built aims to achieve a particular time, but gets a shorter time because of the instabilities.”
The way this has manifested itself in fusion engineering is that “the machines get huge...you make it bigger to compensate for the fact you are losing energy faster than you expect” and to address the instability problem.
You basically end up with something like ITER, the €20 billion flagship of magnetic fusion. This is very likely to “succeed technically” and demonstrate net gain, Dr Hawker believes, but the enormous scale and complexity is daunting, resulting in huge costs and decades of delay.
First Light’s “starting point is very different”, he says, with “a radically different approach to the stability problem.” Instead of creating a fusion scheme based on an idealised theoretical approach, it took its inspiration from nature and the aforementioned pistol shrimp, which clicks its claw to produce a shockwave. “At the same time the water cavitates, and these cavities collapse. The air and water vapour inside is heated and confined as the cavity implodes, and a plasma is formed. The only other naturally occurring inertial confinement phenomenon is a supernova.”
Dr Hawker elaborates: “The pistol shrimp phenomenon is full of instabilities and yet creates an inertially confined plasma anyway – so rather than trying to make instabilities go away, which is what everyone else is trying to do, we are trying to work with how it happens in nature, to understand that, and to ask the opposite question: what do we need to make this work? How can we push this and live with the real-world complexity?” This led to a simplification of the way energy is put into the system, namely the use of a high-velocity projectile to create the shockwave to collapse a cavity in a target.
“The original research that led to the founding of the company was focused on the sophisticated simulations required to model this incredibly complex phenomenon.”
The company’s focus is now very much on design of advanced targets and testing with ever faster projectiles.
First Light has built one electromagnetic launcher already (Machine 1), – which achieved a projectile speed of 5km/s (using a simple “plate flyer” concept) – and leased a second, Machine 2, which was used to study and perform experiments on the physics of the electromagnetic launch, informing the design of Machine 3.
Construction of Machine 3, able to provide 200kV and 14 million amps over 2 microseconds, is underway, with commissioning expected by the end of 2018. It will be able to achieve a projectile velocity of 20 km/s, and should be able to demonstrate “first fusion.” Machine 3 will be a unique facility, says Dr Hawker, enabling pressures and velocities to be attained that will “massively extend the development of our fusion target designs.”
First Light has also worked extensively with a two-stage gas gun, a commercially available machine, which can get to around 8km/s. This was in fact used to demonstrate formation of a plasma and inertial confinement by projectile driven shock, with the emission of light from the hot dense plasma formed during cavity collapse (see Figure 1). This result, achieved with limited capital and confirmed by an expert third party, was a major milestone for the company, and showed that the initial computer modelling was producing accurate results.
Use of the gas gun has enabled continued target development, but would not be applicable in a fusion machine, for which electromagnetic launch is considered the only viable way of achieving the required projectile speeds.
Dr Hawker estimates the proposed machine architecture to be three orders of magnitude cheaper than the mainstream laser-driven approach typically used in inertial fusion.
By improving target design, First Light is aiming to reduce the speeds needed to achieve fusion or, for a given speed, to increase the energy delivered to the fusion reaction.
First Light regards target design as its technical unique selling proposition and “long term competitive advantage”, with the aim of partnering with other companies that “bring the big engineering capability required to build a power plant.”
The “heritage of the company is in cutting edge modelling and experimental capabilities. This is an advantage we believe a start-up can credibly defend against larger and older companies.” The company has already demonstrated the “excellent performance” of its modelling tool.
First Light sees itself as focusing on what it calls the “core technology”, shown in red in Figure 2, the other key parts of a postulated power plant being the “fusion island”, shown in blue, and “balance of plant” items. By limiting its scope to the core technology, the company believes it has “created a task that, whilst challenging, is of a magnitude that can be tackled by a start up.” Development of the other areas will entail partnerships.
Figure 3, bottom left, shows First Light milestones, achieved and planned, in terms of plasma temperature and driver velocity.
The £23 million funding secured, in addition to the initial £2 million (which enabled first plasma to be demonstrated), is considered sufficient to achieve first fusion (already demonstrated by both magnetically confined and laser-driven inertial systems).
The next phase, beyond first fusion, will be to demonstrate “energy gain”, which has not been achieved by any mainstream fusion project thus far, but which is the goal of both ITER and NIF.
Huge gain by inertial fusion has of course been achieved in the form of the H bomb, but this uses an atomic bomb as driver, which is not particularly practicable for a power station.
The capital requirement for First Light to get to first gain is estimated to be of the order of £400 million, which is cheap by mainstream fusion standards.
The company believes its approach, which lends itself to very rapid test iteration loop – six weeks from concept to experiment – makes development costs “substantially cheaper than the competition” and if it can achieve the key step of first fusion, thinks its pathway to gain and a power plant is likely to be “much simpler, quicker and cheaper than mainstream approaches.”
This claim might sound optimistic, but Dr Hawker insists that scientific rigour is a core value of his company. There has been a fair amount of hype in fusion, and developers “say a lot of stuff, but we try to stick to the evidence.”
One manifestation of this, Hawker believes, is the high calibre of the advisory board that the company has managed to attract. These include: a Nobel Prize winner, Steven Chu; a physicist trained by Enrico Fermi (Richard L Garwin); and a former scientific advisor to the UK government, Sir David King, who said he was “very excited to be joining a business working on a truly disruptive innovation, which could transform the world’s energy system for ever.”
A potentially troubling question remains: in an era of renewables and distributed generation, does fusion have a role, or is it a throwback to grandiose centralised generation schemes of the past?
Unusually for a fusion concept, Hawker sees the technology he is developing as a possible competitor with gas in what he calls the “flexible baseload” segment of the power market, providing carbon-dioxide-free power when there’s no wind and sun. It is also able to ramp up and down flexibly (much more so than conventional fission reactors), with a power plant installed capacity of around the 300MW mark.
This operational flexibility stems from the fact that inertial fusion is inherently a pulsed process – “put some fuel in, burn fuel, reset, and repeat” – somewhat analogous to an internal combustion engine, in contrast with the steady state “furnace” concept envisaged by magnetic fusion developers. In the case of inertial fusion, the power out is governed by repetition rate. “That’s why we think we can address flexible baseload, because we have this “free parameter” of repetition rate.”