In the case of the most recalcitrant by-product of the nuclear industry, physics has never been the bottleneck. It has been engineering: How to transform a great idea on paper into hardware that is efficient, reliable and can run at a scale that redefines the meaning of “waste.” This is the working assumption of two current projects in the United States, on accelerator-driven systems in which particle accelerators are viewed not as scientific but as industrial mechanisms.
The core of these is a strategy that transforms spent nuclear fuel in a storage requirement into a feedstock. Weakly accelerated neutrons In an accelerator-driven system (ADS), a beam of protons hits a heavy target, usually liquid mercury, to produce neutrons by means of spallation. Those neutrons then react with the used fuel long-lived isotopes and transform them into energy as short-lived isotopes and heat that can be used to produce electricity. The value proposition is straightforward: with partitioning, recycling, the time scale of radiotoxicity of stored fuel, which in ARPA-E NEWTON program verbiages, is measured in the order of 100,000 years, changes to approximately 300 years.
“Instead of having a lifetime of 100,000 years in storage, for example, you can shorten the storage years down to 300,” said Rongli Geng, head of SRF Science & Technology at Jefferson Lab and principal investigator for both projects.
The energy efficiency is the long pole, since an ADS can only make sense on a systems level when it is clear that the gains are not lost due to the power requirements of the accelerator. The work of Jefferson Lab aims at the superconducting radiofrequency (SRF) cavities which support the beam. Traditional SRF cavities are manufactured of pure niobium and usually need large cryogenic facilities. The NEWTON-supported initiative moves cavities that are created with a layer of coating of niobium with tin, permitting them to work at elevated temperatures and permitting more common cooling approaches. Parallel to this, the group is developing more complicated “spoke” cavities which are expected to make the team even more efficient by facilitating accelerator designs which would be more suited to neutron production requirements.
The second constraint is power delivery. A 10 megawatts radiofrequency power is required of an ADS-scale accelerator, although it also requires frequency-locking to the cavities at 805 MHz. The engineering twist of the project is to transform magnetrons, which are hardware most commonly seen outside the lab context into high-power, combinable source that can satisfy the stability and efficiency criteria. Collaboration with Stellant Systems is intended to demonstrate units capable of being stacked whilst maintaining strict frequency control, a factor that will make the expression “cheap RF power” a mere slogan or a subsystem that can be made a reality.
Through the mentioned objective of the NEWTON program, the U.S. is seeking an avenue that may solve the whole commercial used nuclear fuel inventory in 30 years. It is important to note that the presence of industrial partners at the beginning will indicate that such a project is not focused on a single demonstration, but instead, it will result in a repeatable architecture that can be constructed and maintained.
It is possible to see the importance of that reliability threshold with international work. The pre-industrial ADS pilot currently under construction in Belgium, called MYRRHA, is aimed at proving integrated operation, i.e. subcritical control which prevents fission when the accelerator is shut, and specifically with a goals of minimizing the volumes of waste and the storage life to the same time horizon of a few hundred years as ADS-based transmutation.
Beyond the policy debate concerning repositories and reprocessing lurks a more technical issue: is it possible to make accelerators act like grid infrastructure? “The challenge is to really translate the accelerator science from where we are right now in terms of technology readiness to where the technology needs to be for this application,” Geng said.