Wild Cards: Could these technologies advance the energy transition?

In December 2022, scientists at the US National Ignition Facility achieved something that researchers had been attempting for 70 years: a nuclear fusion reaction that produced more energy than the laser pulse used to initiate it. The announcement made front pages around the world. Then, almost immediately, the caveats arrived. The lasers themselves consumed roughly a hundred times the energy delivered to the target, commercial fusion was still decades away, the engineering challenges remained formidable. The moment was both genuine and incomplete, a pattern that recurs across almost every frontier energy technology.

That tension between real scientific progress and the long road to commercial deployment defines where the energy transition stands today. Solar, wind, and batteries are scaling at extraordinary speed and will do most of the work. But the International Energy Agency (IEA) is explicit that nearly half the emissions reductions required to reach net zero by 2050 depend on technologies currently at demonstration stage or earlier. The question is what are some of the wild card technologies that are attempting to help bridge that gap?

Floating and airborne wind energy

The most dramatic recent progress has been in floating offshore wind, and China is where the pace is fastest.

The Global Wind Energy Council estimates that roughly 80 per cent of global offshore wind potential sits in waters too deep for fixed-bottom foundations. Until recently, floating turbines capable of accessing that resource were too expensive to build and too small to make the economics work. That is changing rapidly. In January 2025, state-owned CRRC commissioned the 20 MW Qihang prototype in Shandong Province[i], at that point the largest floating turbine in the world. By July, China Huaneng and Dongfang Electric had unveiled a 17 MW direct-drive unit with a 262-metre rotor diameter, designed to survive typhoon-force seas, and dispatched it for trials off Guangdong[ii]. In October, Mingyang Smart Energy announced a twin-rotor 50 MW design that would represent a step-change in unit capacity that normally takes the industry a decade to accumulate incrementally[iii].

The underlying logic is economic rather than merely technical: fewer, larger turbines reduce installation, cabling, and operations costs, which account for the bulk of offshore wind project expenditure. For countries like Japan, hemmed in by deep territorial seas with limited viable shallow-water sites, floating wind is less a technological curiosity than potentially a strategic necessity. European developers are approaching the same challenge from a systems perspective. Danish firm Floating Power Plant[iv] is deploying a combined wind-and-wave platform off the Canary Islands that integrates a turbine, wave energy devices, and up to 300 MWh of onboard hydrogen storage, targeting continuous dispatchable output from a single offshore structure.

Closely related, though still in an early stage of development, is airborne wind energy, which pushes the same idea—going higher—much further. A recent milestone in China saw a floating wind platform operating at extreme altitude (around 2,000 metres) successfully connected to the grid, signalling that the concept of harvesting stronger, more consistent high-altitude winds is moving beyond theory into early demonstration.[v]

Conventional turbines operate at hub heights of about 100–150 metres, while winds at several hundred metres and above are significantly stronger and more stable. Airborne systems take this further using tethered kites or wing-like devices that either generate electricity in the air and transmit it down a cable or convert aerodynamic tension into power on the ground. The approach eliminates the need for tall towers and heavy foundations, dramatically reducing steel and concrete use.

The operational challenge, however, remains substantial: these systems must autonomously manage launch, flight, power generation, and recovery under highly variable conditions. Companies such as Kitemill and Ampyx Power have demonstrated sustained generation but have not yet reached full commercial scale. The recent high-altitude grid-connected test in China is an important proof point, but still early in the technology’s development pathway.

Osmotic Energy

If the wind technologies feel relatively tangible, osmotic energy sits in a different register.

The concept is straightforward: At the point where freshwater and seawater mix, a large and constant amount of energy is released. Technologies like pressure-retarded osmosis and reverse electrodialysis capture this energy difference using special membranes, which allow ions to move and generate an electrical current. The core advantage, weather-independent baseload generation at every estuary on Earth, has been understood since the 1970s. Earlier commercial efforts failed because the membranes didn’t perform well enough. They couldn’t move enough ions, so the amount of energy produced wasn’t high enough to be economically viable.

That bottleneck appears to be breaking.

French startup Sweetch Energy, named by the World Economic Forum as one of its 10 most important emerging technologies to watch in 2025[vi], has developed proprietary INOD (Ionic Nano Osmotic Diffusion) membranes using nanoscale pore architectures that allow ion migration at rates far exceeding earlier designs. Its OPUS-1 pilot began operations in late 2024, and the company is targeting €100 per MWh by 2030, which would be cost-competitive with nuclear on a baseload basis. Japan independently commissioned its first osmotic plant in Fukuoka in August 2024.

The Dubai Future Foundation estimates that energy from mixing freshwater and seawater could generate around 5,177 TWh of electricity each year, roughly one-fifth of current global electricity use. In total, about 30,000 TWh of energy is released annually at river mouths and coastal mixing zones, but almost none of it is currently captured. In other words, a huge potential energy source exists, yet it remains largely untapped due to technical and economic challenges.

Air Carbon Capture

Unlike the other technologies discussed, direct air capture (DAC) doesn’t produce energy, it removes carbon dioxide directly from the air. This matters because even with rapid renewable energy growth, there is still excess CO₂ in the atmosphere that needs to be actively removed to stay on track for 1.5°C warming targets.

According to the IEA net zero pathway, the world would need to remove about 80 million tonnes of CO₂ per year by 2030, rising to over 1 billion tonnes by 2050.[vii] Today, however, global DAC capacity is only around 10,000 tonnes per year, leaving a gap of several orders of magnitude.

The main challenge is cost. Early projects are expensive, at roughly $400–$700 per tonne of CO₂ removed, though costs are expected to fall over time as technology scales and improves, potentially to below $100 per tonne in the coming decades. More than 130 projects are currently in development, including a major 500,000-tonne-per-year facility in Texas being developed by 1PointFive[viii]. Early demand is being supported by corporate buyers like Microsoft, Amazon, and JPMorgan Chase, but large-scale public support will likely still be needed to close the gap fast enough to meet climate goals.

Solid-state batteries

Where DAC operates on a multi-decade vision, solid-state batteries are closing in on commercial reality fast enough to matter for this decade's EV transition.

At this year’s Consumer Electronics Show (CES), Finnish startup Donut Lab claimed to have produced the world's first production-ready all-solid-state battery: 400 Wh/kg energy density, a 100,000-cycle lifespan, and a full charge in five minutes.[ix] The announcement was met with strong scepticism from battery scientists, because impressive results at the individual cell level often don’t translate well when scaled up into full battery packs, and the company has not revealed exactly how its battery works.

However, independent testing by Finland’s VTT Technical Research Centre in February found that the fast-charging performance was real at the cell level[x]. A 94 Wh battery cell was able to charge from 0 to 80 per cent in under 10 minutes, while staying relatively cool and retaining most of its charge over time – 97.7 per cent over 10 days, which helps confirm it isn’t just behaving like a supercapacitor.

In later real-world testing on a Verge TS Pro in March, the system reached 50 per cent charge in about five minutes at high power levels, without needing liquid cooling.[xi]

The questions about full-scale pack performance remain open. But the broader context matters: Toyota, QuantumScape, CATL, and BYD are all advancing competing solid-state approaches, and BYD's five-minute lithium iron phosphate (LFP) charging, already tested in production vehicles in China, suggests that charging time is becoming a much smaller barrier to EV adoption, regardless of which technology succeeds.

Nuclear Fusion and Night Power Cells

These two technologies, which are at the furthest end of the speculative range are worth holding in view, precisely because the upside, if they work, is so large.

Nuclear fusion has moved into early demonstration territory, with the 2022 milestone at the National Ignition Facility and France’s WEST tokamak sustaining plasma for over 22 minutes in 2025. But the main barriers are now engineering ones: materials that can survive extreme neutron exposure, superconducting magnets at commercial scale, and a workable tritium fuel cycle that is still largely unproven outside laboratories.[xii]

Companies such as Commonwealth Fusion Systems, TAE Technologies, and Helion Energy are pursuing different routes, with the most optimistic timelines pointing to the mid-2030s for first commercial plants. The appeal remains straightforward: potentially limitless, zero-carbon baseload power, enough to keep capital flowing despite the long odds.

Thermoradiative diodes are a more unusual type of energy technology. While a typical solar panel generates electricity by absorbing sunlight, these devices work in reverse, they produce electricity by releasing heat as infrared light into the cold of outer space. A team at UNSW Sydney demonstrated this concept in 2022 using materials similar to those found in night-vision equipment.[xiii]

Right now, the amount of electricity they generate is extremely small, about 100,000 times less than a standard solar panel, and researchers think it could take another decade before the technology becomes practical. Even in the best-case scenario, the maximum output would be around one-tenth of a solar panel. However, because they work by emitting heat rather than absorbing sunlight, they could potentially generate power at night, complementing daytime solar energy.[xiv]

The silver lining

Floating solar panels installed on hydroelectric reservoirs have been up and running in countries like Thailand and Malaysia for several years, and the idea is steadily gaining ground across the region. It works well because it builds on infrastructure that already exists: the water helps keep the panels cool, which improves their performance; the panels reduce evaporation, helping conserve water; and the dam itself acts like a built-in battery. During the day, solar power can meet demand while saving water, which can then be used to generate electricity later when it’s needed. There’s no need for new grid connections or major approvals, just adding more clean energy to what’s already there.

That practicality is a useful way to wrap up a broader look at energy technologies, ranging from those ready to deploy now to those still in early development. Breakthroughs like the 2022 fusion ignition at the National Ignition Facility didn’t happen overnight - they were the result of decades of work. Similarly, technologies that once seemed unviable, like osmotic power, could become competitive within the next decade. Solar itself is a reminder of how quickly things can change: what was once one of the most expensive forms of electricity has become the cheapest. The broader lesson is that the energy sector has a strong track record of rapid and unexpected progress, and with significant investment now flowing into emerging technologies, some of these wild card technologies may yet prove to be more effective than once thought.

[i] https://www.offshorewind.biz/2025/01/20/crrc-installs-worlds-largest-floating-offshore-wind-turbine-in-china/ 

[ii] https://www.chinadaily.com.cn/a/202507/13/WS68735453a31000e9a573bb3b.html 

[iii] https://en.myse.com.cn/news/info.aspx?itemid=2534 

[iv] https://floatingpowerplant.com/ 

[v] https://www.euronews.com/next/2026/01/29/a-floating-power-station-chinas-flying-wind-turbine-hits-milestone-with-grid-connected-tes 

[vi] https://www.weforum.org/stories/2025/09/what-is-osmotic-energy-and-how-could-it-generate-one-fifth-of-the-world-s-energy-needs/ 

[vii] https://www.iea.org/energy-system/carbon-capture-utilisation-and-storage/direct-air-capture 

[viii] https://www.1pointfive.com/about 

[ix] https://www.technologyreview.com/2026/02/26/1133722/solid-state-batteries-donut-lab/ 

[x] https://www.theverge.com/transportation/882993/donut-labs-solid-state-battery-charge-speed-vtt-test 

[xi] https://www.donutlab.com/battery-pack-announcement/ 

[xii] https://theconversation.com/nuclear-fusion-could-one-day-be-a-viable-clean-energy-source-but-big-engineering-challenges-stand-in-the-way-237544

[xiii] https://www.unsw.edu.au/newsroom/news/2022/05/night-time-solar-technology-can-deliver-power-in-dark

[xiv] https://www.abc.net.au/news/2022-05-17/australian-researchers-show-solar-power-can-be-generated-at-nigh/101070388