Engineers at the Titan-2 Holding, the construction company tasked with building the third unit at Leningrad Nuclear Power Plant in Russia, recently oversaw the installation of a 120-tonne (132 U.S. tons) core catcher in the reactor containment building. This construction is part of phase II of the Leningrad plant, one of the largest nuclear facilities in Russia. 

With an eye on climate goals, countries are turning back to nuclear energy to meet their energy demands without releasing carbon into the atmosphere. While this is a major shift for other countries, for Russia, it is just a matter of upgrading its nuclear facilities with more powerful home-built reactors. 

The Leningrad Nuclear Power Plant, for instance, is one of the largest power facilities in Russia, providing about 55 percent of the electricity demand in the city of St. Petersburg and the Leningrad region at large. With an installed capacity of 4,400 MWe, the nuclear power plant singlehandedly meets 30 percent of the electricity demand of northwest Russia. 

After over five decades of operation, the original nuclear reactors have been shut down, and their replacement is being built as Phase II of the Leningrad Power Plant. 

Construction in Phase II

The Leningrad Nuclear Power Plant consisted of four RBMK units, one of the earliest Generation II reactors, some of which are still operational elsewhere. At Leningrad, the first two units were shut down in 2018 and 2020, after 45 years of operation each. 

In Phase II, Russia is building VVER-1200 reactors with an estimated service life of 60 years, with a possibility of an extension of another 20 years. This will see the reactors in operation, well beyond the 2080s. The first two units at Leningrad Phase II (Leningrad II-1 and II-2) are commonly known as Units 5 and 6, while the other two units, Leningrad II-3 and II-4, are referred to as Units 7 and 8. 

While Units 3 and 4 at the Leningrad Nuclear Power Plant remain operational, work to replace them has already begun, with concrete pouring for Units 7 and 8 scheduled to start in March 2024 and March 2025, respectively. The core catcher for Unit 7 (Leningrad II-3) was completed recently. 

What is the core catcher? 

Made out of thermally resistant steel, the core catcher is a safety device that springs into action in the event of an emergency. In a situation where the core has melted, the core catcher is designed to hold the radioactive material within the containment of the reactor and not let it spill out. 

Engineers at Titan-2 oversaw the installation of the core catcher into a concrete shaft of the reactor containment building for Unit 7. 

“The likelihood that this passive safety system will ever be truly needed is extremely low,” said Alexey Mochalov, Deputy Chief Engineer for the Construction of New Power Units at Leningrad. 

“All the decisions made in the design of domestic power units with VVER-1200 reactors, the high-quality construction, installation, and commissioning work, as well as the highly qualified and responsible personnel of the nuclear power plant, will leave the trap in standby mode for the entire design life of the unit.”

The construction will now proceed to install the thermal insulation, support and thrust trusses, and other equipment and measuring channels to enable reactor vessel assembly as early as next year. 

Ferveret, a startup from the Massachusetts Institute of Technology (MIT), is using cooling technology deployed in nuclear reactors to solve challenges faced by data centers today. The waterless cooling technology can also help establish data centers in areas where water is scarce while abundant solar energy powers the chips and cooling infrastructure. 

The recent surge in artificial intelligence (AI) applications has been made possible by high-performance computing chips from companies like NVIDIA. When AI needs to be deployed at scale, companies need to house many chips capable of processing information quickly enough. To save costs, they hire computing power from private data centers that buy and deploy the high-performance chips and rent them out over the long term. 

As more and more companies deploy AI in their workflows, the demand for data centers has shot up. With new data centers, energy demand is also growing. Estimates suggest that by the end of the decade, up to 17 percent of the US electricity usage will be consumed by data centers. The power consumption is not just for high-performance chips but also for cooling them. 

Liquid cooling chips

When data centers were first set up, chips were cooled using air. As their computing capacities increased, they began to generate more heat, which needed better cooling approaches. 

Liquids are better than air at removing heat from a chip. Their performance is even better when the heat is used to bring about a phase change. At the liquid’s boiling point, it can extract more heat from the chip without increasing the temperature difference between the liquid and the chip. 

The newest data centers use an immersion cooling approach in which the entire server rack is immersed in the cooling liquid to maximize heat transfer. However, with AI adding more computational power to the server pack, heat generation has increased, necessitating newer cooling solutions. 

Nuclear reactor-inspired tech

MIT spin-off Ferveret is looking to deploy cooling technology used in nuclear reactors and adapt it for data center use. Called subcooled boiling, it uses a liquid with a low boiling point, which produces smaller bubbles than immersion cooling approaches. 

While immersion cooling has to contend with issues such as capturing the bubbles and reliquefying them under constrained pressure and temperature conditions, in Ferveret’s approach, the bubbles recondense quickly into the surrounding liquid. 

In a recent study, Ferveret found that its cooling solution resulted in up to 15 percent improvement in computational power efficiency compared to state-of-the-art immersive cooling techniques. This, combined with Fereveret’s proprietary power control system, allows data centers to extract 35 percent more tokens from the AI models without any increase in power usage. 

Ferveret’s Adaptive Phase Cooling (APC) system is deployed in small boxes, each capable of housing one server. This modular approach allows its deploy and maintain its system much more easily. 

“The physics enable us to get to form factors that weren’t possible in the past,” said Reza Azizian, co-founder at Ferveret, in a press release. “Most immersion cooling solutions are large tanks that people submerge the servers in. We have a smaller, modular rack-mounted solution that makes it adaptable to the current infrastructure, so it’s easier for people to deploy our technology.”

The waterless cooling approach is also beneficial in areas where water is scarce, but sunlight is plentiful. Setting up solar power plants in these areas is simple, but with low water, data centers could not be cooled down. 

“This technology can help deploy data centers in regions where normally you wouldn’t have the resources to do so, including Africa, the Middle East, and of course parts of America. It’s a huge unlock,” added Matteo Bucci, associate professor at the Department of Nuclear Science and Engineering and also Ferveret co-founder. 

Scientists at the Argonne National Laboratory (ANL) in the US found an elusive type of vibration, called the Higgs mode, in a semiconductor material for the first time. The collective movement of atoms resulted in a change in the symmetry of the crystal, which can be tuned to optimize the material’s properties. 

If you recall science taught at school, you will remember that in gases molecules are loosely packed and are in motion, whereas in solids, molecules are densely packed and rigid, which gives them a defined shape. Yet, even within solids, atoms are in a constant motion. Much of this is vibrational motion, which is random and uncoordinated. 

Scientists know that certain inputs can turn this random motion into coordinated, in-sync movement. Collectively, these vibrations make a sound, referred to as phonons. By tuning phonons, scientists have also found a way to influence the structure and behavior of materials. 

Finding the Higgs mode

Researchers at ANL were using light to change the phonon activity in a semiconductor material called metal halide perovskites. These materials are crucial as we look to build the next generation of solar cells, advanced sensors and quantum technologies. 

During one of their experiments, where the scientists exposed a two-dimensional perovskite crystal to ultrafast laser pulses, they noticed a complex and elusive type of vibration that modulated the symmetry of the crystal itself. This is called Higgs mode. 

Throughout history, researchers have found mathematical analogs of Higgs mode in various materials. The Higgs boson in particle physics, or that in superconductivity, which allows electricity to flow through a material with no resistance. 

The Higgs mode is an oscillation in a system to a degree where it shows some symmetry or order. It typically emerges when a system undergoes a phase transition due to a phenomenon known as spontaneous symmetry breaking. 

On this occasion, ANL researchers observed Higgs mode for the first time in a semiconductor material. 

Higgs mode in semiconductor material

The researchers were working with butylammonium lead iodide, a metal halide perovskite which is used to make single-layer semiconducting crystals. Finding its Higgs mode allowed researchers to tune their band gaps, making them better suited for solar cell applications. 

Band gaps of a semiconductor material determine which light frequencies it can absorb to make electricity in a solar cell and which ones they scatter. In their experiments, when the researchers exposed the semiconductor to ultrafast laser pulses, small groups of atoms in it began to oscillate. 

The interactions between the electrons of these oscillating atoms changed their angles too. Interestingly, this also changed the bandgap of the material. 

“We found that the bandgap increased and decreased periodically and rapidly,” said Richard Schaller, a scientist at ANL who was involved in the work. ​“Essentially, the color of the sample oscillated as it rocked through different crystal symmetries — turning redder and then bluer, over and over.”

The researchers also found that this Higgs mode steered the crystal toward a phase that cannot be achieved by heating it. This demonstrates that light-based excitation of material is different from heat-based excitation, thereby offering a new way to explore material phases and properties. 

“If we can use light to control structural and electronic changes in materials on ultrafast time scales they might find use as optical switches in modern microelectronics and quantum technologies,” added Ayushi Shukla, postdoctoral researcher at ANL in a press release. ​

“Also, stabilizing novel, high-symmetry phases with low band gaps could open exciting opportunities for photovoltaic applications.”

The research findings were published in Nature Materials. 

Researchers at the John Hopkins Applied Physics Laboratory (APL) teamed up with their counterparts at John Hopkins University to develop a comprehensive framework to model quantum noise for superconducting quantum processors. This is expected to help develop better error correction strategies leading to fault tolerant quantum computers in the future. 

Quantum computing is the next big leap in computing that researchers are keen to develop thanks to its exponentially fast computational speeds. Even the world’s fastest supercomputers fall exceedingly short when compared to a quantum computer with just a few hundred qubits. 

Yet, quantum computers aren’t being deployed to solve real-world problems. The reason is the error accumulation in quantum computation. Quantum bits or qubits are naturally prone to noise, where a change in temperature, electric or magnetic field can lead to the accumulation of errors, prompting the need to build robust error correction mechanisms. 

Modeling noise in quantum computers

A research team led by Gregory Quiroz, a physicist at APL decided to study quantum noise in real operating environments for multi-qubit quantum computers. So, they accessed a quantum cloud consisting of 39 qubits across seven superconducting devices. 

The researchers were keen to study transmons, a specific type of superconducting qubit which is known for its reduced sensitivity to noise from electric fields. Working with a cloud-based system also presented another challenge since the researchers did not have low-level access to the hardware. 

However Quiroz noted that the challenge was much like a real world simulation, where users do not have low-level access to hardware in proprietary systems either. Instead of attempting to study noise effects on a single operation in detail, the researchers ran repeated computations on the quantum cloud to accumulate errors. 

By looking into how errors accumulated and how deviant the answers were from those expected, the researchers were able to get valuable insights into how the physical system was working. 

Building a comprehensive model

Attempts at studying errors in quantum computation have traditionally worked with only one type of error, either coherent or incoherent. The latter are errors where information stored on qubits is lost and not much can be done about them. 

However, the coherent ones are caused by flaws in the hardware and are therefore fixable. Interestingly, APL’s approach of working with a quantum cloud allowed them to study both incoherent and coherent errors in the same framework. 

We were able to put a wide variety of errors together into one model, which is simple in terms of parameters but also comprehensive in the types of phenomena it can describe — even predicting the performance of small quantum algorithms,” added Yasuo Oda, postdoctoral researcher, who worked with Quiroz on the model. “That’s our biggest contribution.”

“Now that we have this low-weight noise model, we have the opportunity to apply it across all levels of the quantum computing stack, from hardware design to algorithm design to error correction,” said Quiroz in a press release. “The information we can get from the model can inform every level of the quantum computing stack.”

The research findings were published in the journal PRX Quantum.

Scientists at the Ames National Laboratory in the U.S. have developed a systematic path to the discovery of new approaches to building permanent magnets without relying on rare-earth elements. The approach combines physics-based modeling and high-throughput simulations with reasoning-based artificial intelligence (AI) tools to guide the discovery process even before materials are made in the laboratory. 

Permanent magnets are a crucial part of our everyday lives. From aiding data storage to helping us drive electric vehicles (EVs) to enabling high-quality medical imaging, permanent magnets are used everywhere. 

High-performance permanent magnets made with rare-earth elements are used in defense applications and energy generation. However, the U.S. is highly dependent on other countries to refine its rare earth elements, which increases costs and security risks. Scientists at the Ames National Laboratory have therefore been tasked with developing permanent magnets that do not use rare-earth elements. 

AI, AI everywhere

At a time when AI adoption is soaring and no organization wants to remain behind, another story about AI workflows is not a major surprise. However, for an AI model to perform perfectly, it must be trained on the right data. 

For a general use case, an AI model can be trained on general data. However, for an AI that needs to excel in materials science, a research team led by Prashant Singh, a scientist at Ames, ensured that models were trained on experimentally measured and scientifically calculated material properties. Only this can ensure that the predictions the AI model is making are grounded in real-world behavior. 

“Understanding the physics of materials is important to include in AI frameworks when you are trying to design new materials. If you just use the data to train your models, you are going to get only the predictions within the range of information you have,” said Singh in a press release. 

“But once you understand the physics of what controls specific properties, then you and your agentic tools or AI frameworks can search arbitrary material space,” added Singh. 

Addressing the full pipeline

The advantage of using such an AI-based approach is that it uses a material’s atomic structure and electronic behavior to determine its properties. Magnetization strength, resistance to demagnetization, energy storage capacity, and the permanent magnet’s behavior at high temperatures are major parameters that determine its usefulness. 

Using an AI model that accounts for this allows researchers to identify the most promising material candidates and arrive at likely outcomes through computations rather than real-world iterations. 

“Ames Lab’s strength comes from its deep expertise and a long history of data in the magnet space that no other institution has,” elaborated Singh in the press release. 

“In any material design problem, you need to know how combining two elements will change their performance before you ever run an experiment. We have been building both theoretical and analytical tools to answer that question, and we are now bringing AI into that process to make it faster and broader in scope.”

However, the researchers did not just stop there. The AI tools also take into account the availability and cost of the material. Since the frailty of supply chains has been repeatedly exposed in the past decade, the AI models also factor in these conditions during the discovery process itself to ensure that the material developed is practically possible and can be scaled. 

In doing so, the researchers have addressed the complete pipeline from discovery to industrial availability as the U.S. looks to reduce its dependence on other countries for critical components. 

The research findings were published in the journal Materials Science and Engineering.