An autonomous drone with wheels can roll along the ground, only flying when needed to clear obstacles, which helps its battery last seven times longer, according to its developers.

Rolling robots are efficient and can travel long distances, but cannot traverse big obstacles, while flying drones can get past large obstructions, but have limited range. Now, Ruibin Zhang and his colleagues at Zhejiang University in Hangzhou, China, have attempted to combine the benefits of both types…

The UK’s 40-year-old fusion reactor achieved a world record for energy output in its final runs before being shut down for good, scientists have announced.

The Joint European Torus (JET) in Oxfordshire began operating in 1983. When running, it was temporarily the hottest point in the solar system, reaching 150 million°C.

The reactor’s previous record was a reaction lasting for 5 seconds in 2021, producing 59 megajoules of heat energy. But in its final tests in late 2023, it surpassed this by sustaining a reaction for 5.2 seconds while also reaching 69 megajoules of output, using just 0.2 milligrams of fuel.

This equates to a power output of 12.5 megawatts – enough to power 12,000 homes, said Mikhail Maslov of the UK Atomic Energy Agency at a press conference on 8 February.

Today’s nuclear power plants rely on fission reactions, where atoms are smashed apart to release energy and smaller particles. Fusion works in reverse, squeezing smaller particles together into larger atoms.

Fusion can create more energy with none of the resulting radioactive waste created by fission, but we don’t yet have a practical way to harness this process in a power plant.

JET forged together atoms of deuterium and tritium – two stable isotopes of hydrogen – in plasma to create helium, while also releasing a vast amount of energy. This is the same reaction that powers our sun. It was a type of fusion reactor known as a tokamak, which contains plasma in a donut shape using rings of electromagnets.

Scientists ran the last experiments with deuterium-tritium fuel at JET in October last year and other experiments continued until December. But the machine has now been shut down for good and it is being decommissioned over the next 16 years.

Juan Matthews at the University of Manchester, UK, says JET will reveal many secrets as it is dismantled, such as how the lining of the reactor deteriorated through contact with plasma and where valuable tritium – worth around £30,000 a gram – has embedded in the machinery and can be recovered. This will be vital information for future research and commercial reactors.

“It’s great that it’s gone out with a little flourish,” says Matthews. “It’s got a noble history. It’s served its time and they’re going to squeeze a bit more information out of it during its decommissioning period as well. So it’s not something to be sad about; it’s something to be celebrated.”

A larger and more modern replacement for JET, the International Thermonuclear Experimental Reactor (ITER) in France, is nearing completion and its first experiments are due to start in 2025.

Tim Luce, deputy head of the ITER construction project, told the press conference that ITER will scale up the energy output to 500 megawatts, or possibly even 700.

“These are what I usually call power plant scale,” he said. “They’re at the lower end of what you would need for an electricity generating facility. In addition, we need to extend the timescale to at least 300 seconds for the high fusion power and gain but perhaps as long as an hour in terms of energy production. So what JET has done is exactly a scale model of what we have to do in the ITER project.”

Another reactor using the same design, the Korea Superconducting Tokamak Advanced Research (KSTAR) device, recently managed to sustain a reaction for 30 seconds at temperatures in excess of 100 million°C.

There are other approaches to creating a working fusion reactor being pursued around the world as well, such as the National Ignition Facility at Lawrence Livermore National Laboratory in California. This bombards capsules of fuel with immensely powerful lasers, a process called inertial confinement fusion, and has managed to unleash almost twice the energy that was put into it.

Topics:

Scientists have confirmed that a fusion reaction in 2022 reached a historic milestone by unleashing more energy than was put into it – and subsequent trials have produced even better results, they say. The findings, now published in a series of papers, give encouragement that fusion reactors will one day create clean, plentiful energy.

Today’s nuclear power plants rely on fission reactions, where atoms are smashed apart to release energy and smaller particles. Fusion works in reverse, squeezing smaller particles together into larger atoms; the same process powers our sun.

Fusion can create more energy with none of the radioactive waste involved in fission, but finding a way to contain and control this process, let alone extract energy from it, has eluded scientists and engineers for decades.

Experiments to do this using capsules of deuterium and tritium fuel bombarded with lasers – a process called inertial confinement fusion (ICF) – began at the Lawrence Livermore National Laboratory (LLNL) in California in 2011. The energy released was initially only a tiny fraction of the laser energy put in, but it gradually increased until an experiment on 5 December 2022 finally passed the crucial milestone of breaking even. That reaction put out 1.5 times the laser energy required to kickstart it.

In one paper, the lab’s National Ignition Facility (NIF) claims that trial runs since then have yielded even greater ratios, peaking at 1.9 times the energy input on 4 September 2023.

Richard Town at LLNL says the team’s checks and double-checks since the 2022 result have proved that it “wasn’t a flash in the pan”, and he believes there is still room for improvement.

Even with the hardware currently installed at NIF, Town says it is likely that yields could be improved, but if the lasers can be upgraded – which would take years – things could be pushed even further. “A bigger hammer always helps,” he says. “If we can get a bigger hammer, I think we could get to target gains of about roughly 10.”

But Town points out that NIF was never built to be a prototype reactor and isn’t optimised for boosting yields. Its main job is to provide critical research for the US nuclear weapons programme.

Part of this work involves exposing electronics and payloads from nuclear bombs to the neutron bombardment that takes place when ICF reactions occur, to check that they will function in the event of all-out nuclear war. The danger of an electronics failure was highlighted during a test in 2021 when NIF fired and wiped out all lights across the site, plunging researchers into darkness. “Those lights were not hardened, but you can sort of imagine a military component that has to survive a much higher dosage,” says Town.

This mission means some research from the project remains classified; even the concept of ICF was a classified secret into the 1990s, says Town.

The announcement that ICF had reached the break-even point in 2022 provided hope that fusion power was drawing closer, and this will be bolstered by news that further progress has been made. But there are caveats.

Firstly, the energy output falls far short of what would be needed for a commercial reactor, barely creating enough to heat a bath. Worse than that, the ratio is calculated using the lasers’ output, but to create that 2.1 megajoules of energy, the lasers draw 500 trillion watts, which is more power than the output of the entire US national grid. So these experiments break even in a very narrow sense of the term.

Martin Freer at the University of Birmingham, UK, says these results are certainly not an indication that practical fusion reactors can now be built. “There’s still science to be done,” he says. “It’s not like we know the answers to all of this and we don’t need researchers any more.”

Freer says that as scientific experiments progress, they throw up engineering challenges to create better materials and processes, which will allow better experiments and more progress. “There is a chance that we will have fusion,” he says. “But the challenges that we have are pretty steep, scientifically.”

Aneeqa Khan at the University of Manchester, UK, agrees that recent progress in fusion research is positive, but stresses that it will be decades before commercial power plants are operational – and even that will hinge on global collaboration and a concerted effort to train more people in the field. She warns against interpreting progress in fusion research as a possible solution to tackle our reliance on energy from fossil fuels.

“Fusion is already too late to deal with the climate crisis. We are already facing the devastation from climate change on a global scale,” says Khan. “In the short term, we need to use existing low-carbon technologies such as fission and renewables, while investing in fusion for the long term, to be part of a diverse low-carbon energy mix. We need to be throwing everything we have at the climate crisis.”

Topics:

• nuclear energy/
• fusion power

A cooling system with no moving parts or environmentally damaging refrigerant liquids or gases can work almost twice as efficiently as a standard air-conditioning system, which could slash electricity use.

Most air conditioners and fridges rely on compressing and expanding a fluid to either absorb or release large quantities of heat. While these systems are relatively cheap and simple to produce, they aren’t very efficient and so require lots of energy – about a fifth of the electricity used in buildings globally – and many of the coolants used are environmentally harmful.

Now, Emmanuel Defay at the Luxembourg Institute of Science and Technology and his colleagues have developed a coolant-free refrigeration device made from the metals lead, scandium and tantalum. It can reach maximum efficiencies of more than 60 per cent, almost double that of typical single-room air conditioning units.

The technology relies on a principle called electrocaloric cooling, which is when an electric field applied across a material changes the direction of electric charges, causing a temporary increase in temperature and a subsequent decrease when the electric field is removed.

To make their cooling system, Defay and his colleagues stacked eight strips of the material known as lead scandium tantalate, which is electrocaloric, on top of one another and immersed them in a heat-carrying fluid, silicone oil. When an electric field is switched on and the strips heat up, the fluid moves to the right, and when it cools down, it moves to the left, creating permanent regions of hot and cold of about 20°C difference.

These regions can be used as hot and cold reservoirs from which the oil can be circulated through pipes to cool or heat rooms or objects as desired.

Although the efficiency of the device is theoretically 67 per cent, the current design is around 12 per cent efficient. This could be improved if a better thermal conductor than the lead scandium tantalate were found, says Defay.

Topics:

• energy and fuels/
• materials science

Amid the many debates about the potential dangers of artificial intelligence, some researchers argue that an important concern is being overlooked: the energy used by computers to train and run large AI models.

Alex de Vries at the VU Amsterdam School of Business and Economics warns that AI’s growth is poised to make it a significant contributor to global carbon emissions. He estimates that if Google switched its whole search business to AI, it would end up using 29.3 terawatt hours per year – equivalent to the electricity consumption of Ireland, and almost double the company’s total energy consumption of 15.4 terawatt hours in 2020. Google didn’t respond to a request for comment.

On one hand, there is good reason not to panic. Making that sort of switch is practically impossible, as it would require more than 4 million powerful computer chips known as graphics processing units (GPUs) that are currently in huge demand, with limited supply. This would cost $100 billion, which even Google’s deep pockets would struggle to fund.

On the other hand, in time, AI’s energy consumption will present a genuine problem. Nvidia, which sells 95 per cent of the GPUs used for AI, will ship 100,000 of its A100 servers this year, which can collectively consume 5.7 terrawatt hours a year.

Things could, and probably will, get much worse in time as new manufacturing plants come online and dramatically increase production capacity. Chip maker TSMC, which supplies Nvidia, is investing in new factories that could provide 1.5 million servers a year by 2027, and all that hardware could consume 85.4 terawatt hours of energy a year, says de Vries.

With businesses rushing to integrate AI into all sorts of products, Nvidia will probably have no problem clearing its stock. But de Vries says it is important for AI to be used sparingly, given its high environmental cost.

“People have this new tool and they’re like, ‘OK, that’s great, we’re gonna use it’, without regard for whether they actually need it,” he says. “They forget to ask or wonder if the end user even has a need for this in some way or will it make their lives better. And I think that disconnect is ultimately the real problem.”

Sandra Wachter at the University of Oxford says consumers should be aware that playing with these models has a cost. “It’s one of the topics that really keeps me up at night,” says Wachter. “We just interact with the technology and we’re not actually aware of how much resources – electricity, water, space – it takes.” Legislation to force transparency about the models’ environmental impact would push companies to act more responsibly, she says.

A spokesperson for OpenAI, the developer of ChatGPT, tells New Scientist: “We recognise training large models can be energy-intensive and is one of the reasons we are constantly working to improve efficiencies. We give considerable thought about the best use of our computing power.”

There are signs that smaller AI models are now approaching the capabilities of larger ones, which could bring significant energy savings, says Thomas Wolf, co-founder of AI company Hugging Face. Mistral 7B and Meta’s Llama 2 are 10 to 100 times smaller than GPT4, the AI behind ChatGPT, and can do many of the same things, he says. “Not everyone needs GPT4 for everything, just like you don’t need a Ferrari to go to work.”

A Nvidia spokesperson says running AI on its GPUs is more energy-efficient than on an alternative type of chip called a CPU. “Accelerated computing on Nvidia technology is the most energy-efficient computing model for AI and other data centre workloads,” they say. “Our products are more performant and energy efficient with each new generation.”

Topics:

Xcel Energy announced today that it has received the necessary approval from the Minnesota Public Utilities Commission to extend operations of the Monticello Nuclear Generating Plant through 2040.

The approval is a major step toward continuing to provide reliable, carbon-free energy for more than 500,000 customers across the Upper Midwest while meeting the state’s carbon reduction goals.

The Minnesota Public Utilities Commission unanimously approved Xcel Energy’s application to allow increased storage of spent nuclear fuel at the site, opening the door for at least 10 more years of operation.

“We thank the Commission, Minnesota Department of Commerce and other stakeholders for their careful review and recognition of the importance of the Monticello nuclear plant to our shared clean energy goals,” said Chris Clark, president, Xcel Energy – Minnesota, North Dakota, and South Dakota. “Nuclear power is crucial to achieving those goals because of its unique combination of reliability, affordability, and zero emissions.”

Together, the company’s Monticello and Prairie Island nuclear plants provide more than 30% of the electricity that the company’s customers use in the Upper Midwest. In 2022, the two plants together produced nearly 14,700 gigawatt hours of energy, a record for the sites. They’re among the company’s most reliable generating facilities, and key to reaching Minnesota’s goal of 100% carbon-free electricity by 2040.

As part of extending the use of the Monticello plant, Xcel Energy has also filed with the U.S. Nuclear Regulatory Commission to request an extension of the plant’s operating license. The federal license review and approval process is extensive, and a decision from federal regulators is expected in late 2024.

In addition to the environmental benefits of nuclear power, the Monticello plant also provides a significant economic boost to the local community. The plant is Monticello’s largest employer and local taxpayer, and it supports hundreds of businesses and jobs in the region.

“The Commission’s decision helps ensure many high-quality jobs in our local community for years to come,” said Marcy Anderson, executive director of the Monticello Chamber of Commerce and Industry. “The economic impact is tremendous. Workers live, shop and dine locally, contributing to a thriving Monticello community.”

The approval of the Monticello plant extension is a major victory for Xcel Energy and for the clean energy movement. It shows that nuclear power can be a key part of the solution to climate change, while also providing reliable and affordable energy to communities across the country.

A rhythmic pumping method inspired by the human heart could slash the energy used to move fluids through domestic and industrial pipes.

Forcing fluids through such systems – be it moving oil and gas from drilling rigs to refineries or circulating water in our home heating systems – is estimated to use between 10 and 15 per cent of the world’s electricity supply.

Turbulence inside pipes causes friction, which vastly inflates the energy needed to pump liquids. Previous attempts to reduce turbulence have included complex coatings on the inside of pipes, which would be costly to roll out on a wide scale.

Björn Hof at the Institute of Science and Technology Austria says copying the human heart is a natural starting point to address this problem because it has the benefit of millions of years of evolution. He and his colleagues discovered that pumping liquids through a pipe in pulses – much like the human heart moves blood – can reduce the friction in the pipe and therefore also the energy consumed.

To find out more, the researchers laced water with reflective particles and pumped it through transparent pipes while shining a laser into them, allowing them to visualise the swirls and eddies in the liquid.

They tried numerous rhythmic pulsing patterns, most of which actually increased the energy required to pump the water. But when they introduced a short resting phase between pulses – like that between heartbeats – they found that turbulence in the water diminished. The best experiments yielded a 25 per cent decrease in friction and a 9 per cent overall reduction in energy demand.

To capitalise on this in the real world, pumps would have to be modified to pulsate, which would have a cost, but would be much cheaper than upgrading the lining of often lengthy and awkwardly placed pipes, says Hof. However, as a scientist he says the practical applications are best left to engineers.

“What I don’t know at all is how happy my central heating pump would be if it was permanently switched on and off, so to speak. Then it may not last winter – I have no idea,” says Hof.

Topics:

An analogue computer chip can run an artificial intelligence (AI) speech recognition model 14 times more efficiently than traditional chips, potentially offering a solution to the vast and growing energy use of AI research and to the worldwide shortage of the digital chips usually used.

The device was developed by IBM Research, which declined New Scientist’s request for an interview and didn’t provide any comment. But in a paper outlining the work, researchers claim that the analogue chip can reduce bottlenecks in AI development.

There is a global rush for GPU chips, the graphic processors that were originally designed to run video games and have also traditionally been used to train and run AI models, with demand outstripping supply. Studies have also shown that the energy use of AI is rapidly growing, rising 100-fold from 2012 to 2021, with most of that energy derived from fossil fuels. These issues have led to suggestions that the constantly increasing scale of AI models will soon reach an impasse.

Another problem with current AI hardware is that it must shuttle data back and forth from memory to processors in operations that cause significant bottlenecks. One solution to this is the analogue compute-in-memory (CiM) chip that performs calculations directly within its own memory, which IBM has now demonstrated at scale.

IBM’s device contains 35 million so-called phase-change memory cells – a form of CiM – that can be set to one of two states, like transistors in computer chips, but also to varying degrees between them.

This last trait is crucial because these varied states can be used to represent the synaptic weights between artificial neurons in a neural network, a type of AI that models the way that links between neurons in human brains vary in strength when learning new information or skills, something that is traditionally stored as a digital value in computer memory. This allows the new chip to store and process these weights without making millions of operations to recall or store data in distant memory chips.

In tests on speech recognition tasks, the chip showed an efficiency of 12.4 trillion operations per second per watt. This is up to 14 times more efficient than conventional processors.

Hechen Wang at tech firm Intel says the chip is “far from a mature product”, but experiments have shown it can work effectively on today’s commonly used forms of AI neural network – two of the best-known examples are called CNN and RNN – and has the potential to support popular applications such as ChatGPT.

“Highly customised chips can provide unparalleled efficiency. However, this has the consequence of sacrificing feasibility,” says Wang. “Just as a GPU cannot cover all the tasks a CPU [a standard computer processor] can perform, similarly, an analogue-AI chip, or analogue compute-in-memory chip, has its limitations. But if the trend of AI can continue and follow the current trend, highly customised chips can definitely become more common.”

Wang says that although the chip is specialised, it could have uses outside the speech recognition task used by IBM in its experiments. “As long as people are still using a CNN or RNN, it won’t be completely useless or e-waste,” he says. “And, as demonstrated, analogue-AI, or analogue compute-in-memory, has a higher power and silicon usage efficiency, which can potentially lower the cost compared to CPUs or GPUs.”

Topics:

In a partnership announcement, IP3 Corporation and Green Energy Partners, LLC (GEP) have declared their collective intentions to lead Virginia into a greener future. The two companies are coming together to inaugurate the Surry Green Energy Center, LLC. Their mission? To address Virginia’s industrial and baseload energy needs innovatively.

Central to their collaboration is a sophisticated land acquisition strategy that focuses on establishing green data centers. These plans also encompass potential future ventures into carbon-free hydrogen and nuclear energy production facilities.

To understand the magnitude of this initiative, one must consider that Virginia’s data centers are responsible for processing a staggering 70% of global internet data. The appetite for data storage and management in the state is booming, and these data centers consume roughly a quarter of Virginia’s energy, primarily derived from carbon-based fuels. Enter the Surry Green Energy Center — a revolutionary data center hub powered by pure green energy and enhanced with on-site small-modular reactors (SMRs). The blueprint entails a deployment of 4-6 SMRs that will not only power 20-30 data centers but also generate hydrogen fuel and offer backup power to Virginia’s grid.

Governor Glenn Youngkin highlighted the state’s relentless commitment to advancing energy-reliant sectors in the Virginia Energy Plan, noting the essentiality of consistent and cost-effective power.

The collaborative venture between IP3 and GEP is strategically situated in Surry County, Virginia, ensuring optimum services for digital infrastructure and future power supplies to government bodies and military bases from Washington, DC to Norfolk, VA. The projections estimate the creation of over 3,000 jobs in the Surry vicinity, promising a substantial surge in the local economy through annual revenue and salary distributions.

Delineating the partnership further, IP3 will spearhead project development, financing frameworks, and funding solutions. This holistic agreement encompasses all stages from inception to operation, aiming to magnetize private capital and avant-garde energy corporations to Virginia.

Michael Hewitt, Co-Founder and CEO of IP3, lauded Virginia’s proactive approach to its energy trajectory. Simultaneously, Mark Andrews, CEO of Green Energy Partners, emphasized the amalgamated strength of GEP and IP3 in leading an America-first strategy for low-carbon energy security.

A breakthrough fusion experiment has produced a net gain in energy for only the second time ever and with improved performance over the first successful attempt. But before you get excited about the coming era of unlimited clean energy, there are some important caveats to keep in mind.

What’s so good about fusion power?

Today’s nuclear power plants rely on fission reactions, where atoms are smashed apart to release energy and smaller particles. Fusion works differently, by squeezing smaller particles together into larger atoms – the same process that operates within our sun. Fusion can create more energy, with no radioactive waste, but containing and controlling such a reaction has proven to be a monumental problem for both physicists and engineers.

What has happened and what is ignition?

In December 2022, researchers at the Lawrence Livermore National Laboratory (LLNL) in California reached a historic milestone: they got more energy out of a fusion reaction than they put in. The lab’s National Ignition Facility (NIF) fusion reactor used lasers to create enough heat and pressure to turn deuterium and tritium – isotopes of hydrogen – into a plasma in which fusion could occur. These lasers output 2.1 megajoules of energy, but the reactor produced about 2.5 megajoules, roughly a 20 per cent increase. While those numbers are nowhere near the sort of ratio you would need to run a commercial reactor, it offered a vital glimmer of hope that fusion reactors were a viable goal.

Now the lab has reportedly created a second ignition – the term for a reaction that surpasses break-even – and improved on those numbers with the reactor producing around 3.5 megajoules. The experiment occurred on 30 July, according to a report in the Financial Times.

Does this mean fusion power has been solved?

In short, no.

One problem is that while the reactor’s output is higher than the laser’s output, the lasers themselves are very inefficient. To create 2.1 megajoules of energy they draw 500 trillion watts, which is more power than the output of the entire US national grid. So a significant challenge for the future is to create a reaction that breaks even with its total energy requirements, and not just the final laser stage.

Another issue is that the NIF reactor can fire only once, for a few billionths of a second, before it has to spend several hours cooling its components in order to be switched on once more. A commercial reactor would have to run nearly continuously with multiple ignitions a second.

And, of course, even once a reactor can run for long periods and offset its true energy requirements by the lasers, it would still only be breaking even. For fusion to become a viable alternative to existing power sources, we must be able to extract a large amount of net energy – enough to make the enormous cost of building it worthwhile.

Will fusion be solved in the future?

While it is still impossible to tell for sure, as there could be insurmountable problems ahead, there is more cause for optimism than ever before. The ignition milestone effectively proves that the science is sound, and makes the problem one of engineering rather than physics.

There are two main research approaches aiming to achieve viable nuclear fusion. One uses magnetic fields to contain a plasma, while the other uses lasers. NIF uses the second approach, known as inertial confinement fusion, where a tiny capsule containing hydrogen fuel is blasted with lasers, causing it to heat up and rapidly expand. But there are a host of startups working on unusual designs that all have the potential to break through. All these experiments are helping us to better understand the problem, and the best way forward.

But one thing is clear: with a working fusion reactor still many years away at the very least, we cannot rely on the technology to solve the climate change crisis. Fusion reactors will be perhaps the greatest pay-off ever received from a coordinated research effort stretching back more than a century, but clean and abundant energy will have to come from renewable sources for the short and medium term.

Topics: