IBM Approaches Quantum Advantage and a Liquid Sodium Next Generation Reactor Gets the Go Ahead


Episode Summary:

Moore's Law famously predicted the rapid rate of increase in integrated circuit gate density. As gate sizes are falling into the low single digit nanometre range however, physical constraints on future development are likely. Quantum computing promises to deliver orders of magnitude better performance than conventional binary devices, and IBM has launched a new quantum processor called Eagle, which promises to approach quantum advantage, the point at which quantum computers clearly outperform their conventional counterparts.

Nuclear energy startup TerraPower and GE Hitachi Energy Systems have chosen the site for the first demonstration reactor of Natrium liquid sodium fission technology. The pilot plant will be built in western Wyoming on the site of a coal plant nearing decommissioning. The Natrium concept uses a simplified reactor design that uses substantially less material and low enrichment uranium fuel to produce power more cost-effectively than current pressurized water designs. A prominent investor is Bill Gates, and the pilot plant should be ready in seven years.

Access all episodes of This Week in Engineering on engineering.com TV along with all of our other series.

Transcript of this week's show:

To see any graphs, charts, graphics, images, and/or videos to which the transcript may be referring, watch the above video.

Segment 1: Moore’s Law famously predicted the dramatic increase in transistor density seen in modern integrated circuits, but as gate dimensions drop into the single digit nanometre range, physics suggest that there are limits to how dense those chips can be. Game changing improvements in computer performance will require radically new technology, and quantum hardware is the best candidate to achieve breakthrough performance improvement. 

IBM has announced a breakthrough 127 quantum bit, or “qubit” processor called “Eagle”. According to the company, the Eagle processor represents a tipping point in hardware development: where quantum circuits cannot be reliably simulated on a classical computer. What makes quantum computing different? Compared to current technology, which uses very small networks of transistors as gates to process binary signals, quantum computing harnesses the very non-Newtonian behaviour of matter at the subatomic level. 

Compared to the switching behaviour of transistors, quantum computers can operate at states other than the simple binary one and zero. One consequence of these expanded states is that increasing the size of an array of qubits dramatically increases the computational power of the system as a whole. The current performance goal is termed Quantum Advantage, the point at which quantum machines outperform traditional equivalents. 

While IBM has declared that Eagle is not yet the device to achieve quantum advantage, expanded states create a big problem for conventional binary systems to test or simulate. The number of classical bits necessary to represent the states possible on the Eagle processor exceeds the total number of atoms in the bodies of all 7.5 billion people on Earth today. While quantum computers are still error-prone, and require cryogenic temperatures to operate, the potential advantages in simulation for industries like advanced materials and pharmaceuticals make quantum computing possibly the most important area of technological development so far in the 21st century. 

How will humans develop code for machines of this power? Will it require AI just to harness the power of quantum computers? At this point, these are unknowns, but earlier generation chips running on IBM’s Quantum System One have been deployed at research centers at universities in Germany, South Korea and Japan as well as at the Cleveland Clinic for medical research. 

The space is developing fast, and IBM’s next generation of quantum chips, with 433 and 1121 cubits, will be installed in Quantum System Two, for deployment in 2023. With complexity scaling exponentially in quantum devices, what these more powerful systems can do is anyone’s guess, but fast simulation of complex and difficult processes like protein folding and nuclear reactor behaviour could be the gateway to entirely new technologies in a decade or less.

Segment 2: Advancements in carbon free energy generation are coming fast in the wake of the United Nations climate change conference in Glasgow. While solar and battery storage get much of the attention, nuclear, once dismissed as a viable future energy source, is rapidly regaining popularity due to radical new technologies that may replace conventional reactor designs. While multiple novel designs exist on paper, very few make it to the hardware stage and fewer still become running demonstrators. 

One radical design that appears to be overcoming those hurdles is TerraPower, who are teaming with GE-Hitachi to build a demonstration reactor in Kemmerer Wyoming, the site of the state’s Naughton Power Plant, where two remaining coal units are scheduled to retire in 2025. Kemmerer, in Western Wyoming near the Utah border was chosen for access to infrastructure, ready connection to the grid due to the pre-existing coal plant, community support and successful licensing from the Nuclear Regulatory Commission. 

The core of the project, the Natrium reactor is rated at 345 Mwe and according to the company is four times more fuel-efficient than conventional light water reactors. Typical of new generation designs, requires it also less site infrastructure, using 80% less nuclear grade concrete per Mwe. The key elements of the design are liquid sodium fast reactor operating at atmospheric pressure, coupled to large-scale thermal energy storage similar to systems used in renewable projects. 

The fuel is High Assay Low Enriched Uranium in the system operates with 8% higher thermal efficiency than conventional reactors, with a greatly simplified design. The thermal storage system is key to system simplicity. The reactor operates continuously at a high-capacity factor without throttling, diverting power not immediately needed by the grid into heat storage, and releasing it for power generation at times of peak demand. The combination can also be used for process heat for hydrogen, petrochemical or steel production as well as space heating applications for buildings. The project will employ 2,000 in the construction phase in approximately 254 reactor operations. Project completion is expected in seven years.