Page 363 – Physics World

Home » Page 363

Stretching the limits: novel material produces highly conductive, stretchable bioelectronics

Wearable and implantable healthcare technologies can enable remote, multimodal monitoring for disease diagnosis and tracking, provide programmable, ongoing therapies for heart or retinal diseases, and even restore lost sensation via prosthetics. Yet, these technologies are limited by their material make-up, as they use rigid electrode materials that have limited mechanical compatibility with soft tissues and must be patterned to induce stretchability, limiting device density.

Conducting polymers are intrinsically soft, overcoming problems of tissue compatibility. They also benefit from a large volumetric capacitance that ensures high-quality signal recording and enhanced charge injection for stimulation. Despite these advantages, conducting polymers are not intrinsically stretchable, and existing chemical modifications used to induce stretchability result in conductivities that are too low for relevant bioelectronic applications.

That is why researchers at Stanford University, led by Zhenan Bao, used rational design to build a highly conductive, stretchable and photo-patternable material for use in bioelectronics, publishing their results in Science.

“We are the first to apply this supramolecular structure design into the field of stretchable conducting polymers,” explains Yuanwen Jiang, first author of the study.

Molecular engineering to meet complex design requirements

PEDOT:PSS – or poly(3,4-ethylenedioxythiophene):polystyrene sulfonate – is one of the highest performing and most investigated conducting polymer for use in bioelectronic devices. Previous attempts to modify the mechanical properties of PEDOT:PSS relied on blending approaches with ionic or molecular additives. But because they are not cross-linked to the conducting polymer, the additives are typically washed away in wet, biological environments, leading to large drops in conductivity and device performance.

“It a grand challenge for a single conducting polymer system to simultaneously possess high mechanical stretchability and electrical conductivity at the same time, especially when we micro-fabricate them down to cellular-level feature sizes for real device level applications,” says Jiang.

Polymer design — Supramolecular network: A polymer designed at the molecular level enables ultra-stretchable devices. (Courtesy: Yuanwen Jiang, Yi-Xuan Wang, Jian-Cheng Lai)

To meet this challenge, Jiang and colleagues had to design a supramolecular polymer additive for PEDOT:PSS molecule-by-molecule. Hypothesizing that high stretchability required mobile junctions with a large number of possible arrangements, they designed a polyrotaxane (PR) structure with cyclodextrin rings that can “slide” freely along the polymer backbone. This is the first work to introduce topology, or spatial arrangement, into the design of conducting polymers.

Having achieved stretchability, the researchers designed the polymer chemistry to meet all the requirements of integration with PEDOT:PSS. To enable water solubility and enhance interaction with the conducting polymer, they used polyethylene glycol as the backbone and the side chain. To introduce photo-patternability, they added methacrylates onto the side chains, allowing 2-µm feature sizes. Finally, they note that this multifunctional material could survive sulphuric acid treatment, which enhanced chain packing and thus overall conductivity.

This new material, which the team calls TopoE-S, outperforms all previous PEDOT:PSS modifications, achieving stable conductivity of 6000 S/cm, even at strains of up to 100%.

Outstretching the competition

The design of the bioelectronic interface is essential for applications in the body. “We want the best possible contact on interface, so that we can have the highest quality or signal-to-noise ratio in challenging dynamic environments,” said Jiang. The team has achieved that with the TopoE-S material.

When compared with microcracked gold, one the best stretchable electrodes, TopoE-S had a lower electrochemical impedance in all frequency ranges. This is because of the much larger electrochemical contact area of TopoE-S, which significantly reduces the interfacial impedance. All of these properties are essential for low-power electrical stimulation, the researchers note.

The TopoE-S array can be patterned into electrodes 100 µm in size, while its soft and elastic properties allow for conformal attachment to the skin. Thus, the team was able to fabricate devices to perform high-spatial-resolution electromyography (measurement of muscle’s electrical activity).

To demonstrate the capabilities of this advanced material, the researchers turned to octopuses, which provide a perfect test platform given the large deformation of their muscles. They found that upon electrical stimulation of the octopus arm, the electromyography device recorded muscle activity dynamics with consistently good signal-to-noise ratio, which was impossible with more rigid probes.

New medical therapy applications

Looking to the future, Jiang and colleagues are thinking about new tools that could improve clinical practice. “Through communication and collaboration [with doctors], we understand that there are unmet clinical needs. We try to design tools that can help improve care delivery,” he explains.

Stretchable organic photodiode could improve the performance of wearable devices

The TopoE-S material is designed to solve such unmet clinical needs. One example is the ability to map the brainstem prior to neurosurgery, to determine safe entry points for operation. Currently, this procedure is still performed using bulk electrodes that clinicians must manually move across the surface, greatly extending the time of the procedure.

The researchers patterned the TopoE-S into a high-density array and placed it on the brainstem of a rat, to allow location mapping with extremely high precision. By delivering current pulses to individual electrodes, they could control specific organ functions with little leakage current. Once again, their material outperformed rigid plastic probes that quickly caused severe damage to the fragile brainstem.

Solar cell keeps going after sunset

A new type of photovoltaic cell continues working after sunset thanks to an additional generator that harvests electricity from the temperature difference between the cell and its surroundings. The combined device can generate 50 mW or more of power per square metre at night while providing additional power during the day, and its developers say that it could provide a continuous source of renewable energy for people without access to the electrical grid.

When the Sun is shining, conventional photovoltaic (PV) cells absorb energy over a wide range of wavelengths. At night, the cells lose heat to their surroundings, leaving them several degrees cooler than the ambient air due to radiative cooling.

Thermoelectric effect

The new device uses this difference in temperatures to generate a voltage and current via the thermoelectric effect. Here, a temperature gradient across a solid material containing free charge carriers (electrons and holes) produces a voltage as the carriers migrate from the hot side of the material to the cold side. The strength of this thermoelectric effect is characterized by its thermopower, or Seebeck coefficient, which is the ratio of the voltage difference to the temperature difference across a material.

The new device, which was developed by Shanhui Fan and colleagues at Stanford University in the US, can generate up to 100 mW/m² of power depending on location and weather conditions. It also has an open-circuit voltage of 100 mV, which is orders of magnitude higher than previous demonstrations. “This improvement comes thanks to improved thermal engineering, especially in terms of enhancing the thermal coupling between the solar cells and the thermoelectric module,” Fan tells Physics World.

Simple and inexpensive setup

The researchers designed their device around a commercial SunPower C-60 monocrystalline silicon solar cell with a surface area of 153 cm². To improve thermal conduction across the cell’s surface and enable effective heat transfer to and from the smaller thermoelectric generator (TEG) module, they placed an aluminium sheet underneath the cell. They then connected the underside of the TEG to a base support and a standard heat sink repurposed from a personal computer. This configuration, together with silicone thermal grease applied at each component interface, facilitates efficient heat transfer across the device.

To minimize so-called parasitic heat flow due to inherent inefficiencies in the system, the researchers enclosed the PV cell within a chamber made from polystyrene sheets, with an opening at the bottom for connecting the heat sink. They then covered the top of the chamber with a low-density polyethylene (LDPE) film, which is transparent in both the visible and the mid-infrared wavelength range.

Fan and colleagues tested their device on a rooftop at Stanford, taking measurements on 10–13 October 2021. During this period the sky turned from clear to cloudy. They measured the power of the TEG and the PV cell, as well as the ambient temperature and the temperatures of the PV panel and heat sink. They also measured the electrical power the cell generated using a maximum power point tracking algorithm.

The set-up is simple and inexpensive and could, in principle, be incorporated into existing solar cells, says team member Sid Assawaworrarit. Since none of the components in the device were specifically engineered for this purpose, there is room for improvement, he adds.

Off-grid applications

According to the researchers, the power generated could be used for night-time standby lighting and power in off-grid and mini-grid applications (an area in which PV cell installations are becoming more popular); for charging low-power devices; or for providing power to sensors in remote locations without the need for battery storage. Lighting, for example, would require a PV cell with an area of 20 m². Thanks to the long service lifetime of thermoelectric generators, the proposed setup may also cost less to maintain than battery storage in the long run.

Fan and colleagues say they are now working on improving the power density of their device – for example by further engineering the thermal emissivity of the solar cell and by optimizing the PV-to-TEG area ratio. Previous work showed that adding a selection of dielectric to the top of a PV cell could increase the average emissivity in the 8–13 μm solar radiation window from its current value of 0.65 to 0.90. Assawaworrarit adds that the power density could be increased to 70 mW/m² in weather conditions similar to this previous work and to over 100 mW/m² when operated under clear skies.

The device is described in Applied Physics Letters.

Mountain climbing on the Moon

One night in November 1609, Galileo turned his telescope to the Moon and realized that the changing shadows implied it had mountains. I find myself wondering: what would climbing them involve?

For my expedition, I have chosen the central peaks of the spectacular Theophilus crater in the south-east. The summit is around 1800 m above the crater floor and after eons of meteor bombardments the peak has modest gradients so it should be a walk without any rock climbing. Despite a significant altitude change, my initial thought was that with the reduced gravity of the Moon, it would be relatively easy to accomplish the ascent in a reasonable time – but would it?

First, let’s consider terrestrial hill walking. On Earth we have Naismith’s rule (1892) for estimating walking times: 1 hour for 5 km of distance plus 1 hour for every 600 m of ascent. With a summit 1800 m above us and a round trip of 15 km, the rule would give us a journey time of 6 hours (excluding rests and food breaks) if the climb was on Earth.

The gravity on the Moon is one sixth of that on Earth but I would have to wear a heavy space suit with a life-support pack that adds 80 kg, so overall my weight would be about one third of my terrestrial weight. Consequently, I thought that for the same energy expenditure I could travel three times faster on the Moon, giving a speedy ascent of just 2 hours. However that means I would be progressing at an average speed of 7.5 km/h – I’d have already adopted a running gait if on Earth. With lower gravity reducing traction of the foot on the lunar regolith, the resultant “scree-running” down the mountain would most likely cause falls, risking tears in the space suit, which would be fatal. While the Apollo 17 crew experimented with kangaroo hops down a steep slope, the motto would have to be “walk don’t run”.

So what is the fastest I could walk on the Moon? Looking at the data from the Apollo missions (see “Lunar gaits”, Apollo 11 Lunar Surface Journal, November 2010), the astronauts were walking at 2.2 km/h on the flat, although Neil Armstrong achieved 3.6 km/h by adopting a loping gait. The low walking speeds and tendency to lope may have been due to the inflexibility of the suits impeding limb movement. While there are no data on sustaining a reasonable pace over a long distance, Buzz Aldrin thought the peculiar movement of the body within the suit would make distance walking tiring. But let’s assume that nowadays space suits could be more suitably designed for longer walks.

A simple mechanical model of walking is based on an inverse pendulum. One’s centre of mass (m kg) is assumed to rotate at velocity v m/s in an arc of radius r m about the contact point of the foot with the ground. The downward force due to weight is mg (on Earth g = 9.81 m/s²) and this supplies the centripetal force (F_c) to keep the foot in contact with the ground F_c = mv²/r. Hence v²/rg (a dimensionless ratio dubbed the Froude number) has to be less than unity otherwise the foot lifts off the ground and we break into a run. For a 1 m “leg” to the body’s centre of mass we thus have a limiting terrestrial walk velocity of 11 km/h. Now, while Olympic speed walkers can achieve such speeds they do so by adopting a special gait, rotating their hips and taking many short steps. However, with a natural gait empirically most people break into a run at a Froude number of about 0.5 (around 8 km/h). While less than unity, walking faster causes discomfort in the ankle and associated musculature, and though running is initially less favourable energetically, it is more comfortable at these speeds.

In the reduced gravity of the Moon (g = 1.62 m/s²) a Froude number of 0.5 would give a maximum walking speed of only 3 km/h, so applying a modified Naismith’s rule (pessimistically assuming the same distance/ascent ratio applies) our expedition would take 10 hours, which exceeds the extra vehicular activity (EVA) times of the Apollo missions (7 hours). But does the walk-to-run transition occur at the same Froude number on the Moon?

Treadmill tests have been done both with simulated lunar gravity, using overhead suspension, and real lunar gravity, as experienced in 30 s time slots during the parabolic flight phase of a DC-9 (J. Exp. Biol. 217 3200). The researchers got similar results from both tests and found that the volunteers actually only broke into a run at a Froude number of 1.25 ± 0.37 (average of both sets of results), corresponding to a speed of about 5 km/h on the flat. The greater-than-unity Froude number was thought to be due to the down forces generated by swinging arms and the free leg being more significant in reduced gravity than on Earth, thereby making the inverse pendulum model inaccurate.

The Apollo moonwalkers in their inflexible space suits opted to lope rather than walk at speeds lower than that corresponding to the Lunar Froude walk/run transition number. However, if more flexible space suits can be designed, in the reduced gravity on the Moon, a walking speed of the same value as used in the terrestrial Naismith’s rule should be possible. The time to climb Theophilus peak would therefore be only 6 hours, but with rests it would probably exceed the allowed Apollo EVA times and use up the consumables (oxygen, water) and battery life in the Apollo support packs, which was limited to 8 hours. Perhaps modern technology could improve these time-limiting factors to take the climb beyond the realms of speculation when we once again walk on the Moon.

Freeze–thaw battery could help store solar and wind energy

Freeze-thaw battery figure — Energy cycles: schematic showing how the battery can be frozen and thawed to store an excess supply of energy and release it when demand is high. (Courtesy: *Cell Reports Physical Sciences*/CC BY 4.0)

A battery that “freezes” its stored chemical energy for several months has been developed by Minyuan Li and colleagues at Pacific Northwest National Laboratory in the US. Their battery uses a molten salt electrolyte, which remains solid at room temperatures, and thaws out when heated.

Our capacity to generate electrical energy using wind and solar energy is growing in leaps and bounds, but our ability to store this energy has not kept pace. This is a problem because seasonal trends in wind and solar generation often do not coincide with the energy needs of consumers. At higher latitudes, for example, abundant solar energy can be harvested in the summer, when requirements for heating and lighting are low. Conversely, solar energy can be scarce in the winter when energy demand can be high.

This problem could be solved by storing energy for several months and releasing it when demand begins to outpace supply. While lithium-ion batteries could do the job (while losing as much as 5% of their energy in the first month), there are supply and geopolitical concerns regarding materials used in their manufacture. Other battery types have different challenges that would have to be overcome.

Reduced mobility

To develop a better storage technology, Li and colleagues explored the use of molten salt as a battery electrolyte. When their aluminium-nickel salt was heated to 180 °C, its ions were allowed to flow freely between electrodes immersed in the liquid. But when cooled to room temperature, the salt froze into a solid. This drastically reduced the mobility of its ions – which locks in the chemical energy of the battery. After an indefinite period, the salt could then be heated and thawed, allowing the battery to be discharged.

To ensure that their battery could be a practical storage system, a key concern for Li’s team was to use low-cost, widely available materials wherever possible. This involved choosing a suitable material for their battery’s inert, porous separator – which separates the anode and cathode while allowing ions to pass through. So far, separators have typically been made from ceramics, but these materials are costly, and could be easily damaged during the battery’s freeze-thaw cycle.

As an alternative, the researchers used porous fibreglass as a separator, which fares far better at widely varying temperatures. In addition, they doped their electrolyte with sulphur – another easily attainable material. This addition both boosted the battery’s energy retention further, and activated its nickel cathode.

So far, Li’s team has developed a small prototype of the battery, around the size of a hockey puck. After a storage period of up to eight weeks, the device retained over 90% of its stored energy, following a single freeze-thaw cycle. In the future, the researchers hope that the low cost and simplicity of their design will allow them to ramp up its size and capacity. If achieved, this could allow electricity grids to store energy and eliminate differences between supply and demand.

The battery is described in Cell Reports Physical Sciences.

Oscillating temperature gradient boosts heat flow in fluids

Physicists in the Czech Republic have shown that heat flows more efficiently through a fluid when the fluid’s temperature gradient oscillates over time. The discovery by Ladislav Skrbek at Charles University and colleagues could help climate scientists to better understand the flow of heat through Earth’s atmosphere, and may also lead to better designs of heat exchangers.

From tidal ocean currents to the pumping of blood through our veins, many processes in nature are affected by turbulent flows that are driven by periodically oscillating forces. However, these flows have not been studied to the same extent as those driven by continuous forces. Now, Skrbek’s team has examined a simple example of an oscillating flow by doing an experiment first developed by the British physicist Lord Rayleigh in 1916.

In his experiment, Rayleigh set up a closed, fluid-filled cell; and then established a temperature gradient across the fluid by heating a plate at the bottom of the cell, while cooling another plate at the top. Over time, he observed that the fluid rises and falls in alternating columns: a pattern now known as Rayleigh-Bénard convection (RBC).

Efficiency booster

In a theoretical analysis carried out in 2020, physicists led by Detlef Lohse of the University of Twente in the Netherlands examined a more complex version of this set-up, where the fluid’s temperature gradient is made to oscillate periodically. Surprisingly, their simulations predicted that this oscillation would increase the efficiency of heat transfer between the plates – due to changes in RBC flow patterns.

In their study, Skrbek’s team tested this prediction by doing an experiment involving a cylindrical cell of helium gas, which they cooled to just 6 K. To establish an oscillating temperature gradient, they periodically varied the temperature of the cell’s bottom plate over time. They examined different oscillation frequencies in the range 0.006–0.2 Hz. By constructing the plates out of copper, which is particularly responsive to temperature changes, they could also induce oscillation amplitudes as high as 0.4 K.

By placing temperature sensors throughout the cell, the researchers examined the RBC convection patterns that developed inside. Just as the 2020 study predicted, they found that oscillating the temperature of the bottom plate boosted the heat transfer efficiency by as much as 20%. This enhancement occurred because thermal waves were able to propagate through the helium gas with virtually no attenuation.

Climate models

The discovery could have widespread implications for our understanding of heat transfer in many natural systems. Among the most important is the heat exchanged between Earth’s surface – which oscillates in temperature throughout the day – and outer space, where temperatures sit at just 2.7 K.

Accounting for this oscillating temperature gradient could help researchers to better understand how heat flows through the atmosphere – enhancing the accuracy of existing climate models, and potentially even improving our understanding of other planetary atmospheres. Elsewhere, the results could inform designs for more efficient heat exchangers, which are vital components of heating and cooling systems in many engineering applications.

The research is described in Physical Review Letters.

Quantum physics sets a speed limit for fastest possible optoelectronic switch

Optoelectronic switches can operate up to 1000 trillion times a second – a rate of 1 petahertz – before quantum processes wreck their effectiveness, say researchers in Germany and Austria. The result places a fundamental speed limit on classical information processing, while the experimental techniques used to achieve it could help physicists obtain a better understanding of a wide range of phenomena with applications in coherent electronics.

While today’s computer chips are faster than ever before, the laws of quantum physics limit how much faster they can get. Traditionally, increases in processing speed have come from shrinking transistors and other chip components so that data has a shorter distance to travel. The physical limit here is the size of an atom.

Another approach is to increase the switching rate. One way to do this is to use light rather than a transistor to control the flow of current – for example, by applying a laser pulse to excite the electrons from the valence band of a semiconducting material into the conduction band so that the material becomes a conductor. The energy required for this excitation depends on the semiconductor and corresponds to light frequencies in the infrared to visible range, which ultimately sets the maximum switching speed possible with these materials.

Higher-frequency light

In the new work, a team led by Martin Schultze of Austria’s Graz University of Technology studied a dielectric, lithium fluoride, rather than a semiconductor because the excitation energy of dielectrics is much higher. This enabled the researchers to use higher-frequency light and thereby achieve even faster data transmission. However, as team member Marcus Ossiander explains, there is a drawback: most dielectric materials cannot conduct electricity without breaking.

To overcome this problem, the researchers increased the frequency of the switching light pulse to the extreme ultraviolet range and decreased its duration to one femtosecond (10^-15 s). This is so short that the dielectric does not have time to break. By bombarding a sample of the dielectric with this ultrashort laser pulse, they excited the electrons into the conduction band where the particles can move freely. They then applied a second, slightly longer pulse to accelerate these excited electrons in the desired direction, creating an electric current that they could detect thanks to electrodes connected to both sides of the material.

“The method we employed – sampling ultrafast currents injected by extreme ultraviolet radiation and then driven by light fields – allows us to follow what electrons do when the dielectric switch is operated,” explains Ossiander, who is now a postdoctoral researcher at Harvard University in the US. “Our technique allows us to push electrons from the valence band to the conduction band in a dielectric within one femtosecond – that is, we switched the dielectric from an insulator to a conductor at the speed needed to realize a petahertz switch.”

Combination of two rules in physics

By tracking what the electrons do after they are put in the conduction band, Ossiander and colleagues showed that if switching times got any faster, the electrons would be pushed into regions of the band structure where they would “harm” the signals the researchers were trying to transmit. Ossiander attributes this result to a combination of two factors: the band structure of the material and the Fourier limit. As the light pulses used in switching become shorter, the Fourier limit means that the pulses must stretch across a broader range of wavelengths. As such, the pulses end up promoting electrons into regions where they have undesirable effects.

“Switching beyond one petahertz could cause the electrons to go to band structure regions in which they would suddenly begin to move in a direction that opposes the electric field we apply, which is of course terrible for the switching performance,” Ossiander explains. “By comparing the band structure of lithium fluoride with that of other materials, we were able to show that we pretty much reached the maximum of what is theoretically possible for any existing material before these reversal effects occur.”

A switch the size of an apartment

Real-world applications of the new maximum-speed switch are still a long way off, notes Ossiander. “Currently, the experimental set-up we used for this experiment is roughly the size and cost of a one-bedroom apartment – and this to realize a single switch,” he says. “It will still take us a while to miniaturize the lasers and other parts of the apparatus and construct the millions of parallel switches required for a processor at a size and cost that can be integrated into a smartphone, for example.”

The good news is that according to Ossiander, the technique developed in this work can be applied to study most materials. This should ultimately allow researchers to analyse phenomena such as ultrafast charge carrier transport, elastic and inelastic electron scattering and cooling of (quasi-) free charge carriers that are important for implementing coherent electronics. Ossiander tells Physics World that because the team has already demonstrated the approach for lithium fluoride, which has the largest band gap (the distance between the valence band and conduction band) of all known materials, applying it to other materials “will be relatively straightforward”.

And that is not all: Ossiander adds that the method should enable researchers to measure the electromagnetic field of laser pulses – coveted information that only a handful of techniques can currently provide. The way the laser field evolves after interacting with materials can reveal the time-resolved nonlinear polarization light created in them and therefore can also reveal how bound carriers behave in solids, explains Ossiander. “It will thus provide the full picture of the comportment of electrons interacting with light,” he concludes.

The research is detailed in Nature Communications.

Researchers exploit astronomy technology for biomedical imaging

<i>In vivo</i> imaging — *In vivo* imaging: A reconstructed image clearly separates three radionuclides: ¹¹¹In in lymph nodes below the jaw (red), ¹²⁵I in the thyroid (green) and ^99mTc in lymph node below the ear (blue). Scale bar: 1 mm. The diagram shows where the radionuclides accumulate in the mouse. (Credit: Yagishita *et al.*/Kavli IPMU)

In vivo imaging plays an essential role within medicine, pharmaceutics and biology. Imaging with fluorescent tracers is one promising approach, as multiple tracer types are available to simultaneously visualize the distribution of numerous molecules. Biological tissues, however, will scatter and attenuate visible light, making image reconstruction and quantitative measurements of fluorescent dyes difficult.

Instead, researchers can employ X-ray-emitting radionuclides for in vivo imaging, using techniques such as single photon emission computed tomography (SPECT) and positron emission tomography (PET) to image tracers deep inside the body. Multi-radionuclide imaging, however, is tricky. There are far fewer radionuclides available for imaging compared with the number of fluorescence dyes, and the energy ranges of some of their emission lines can be similar, or even overlap.

What’s needed is a high-sensitivity detector, with high spatial resolution and good enough energy resolution to resolve potential overlaps between multiple radiotracers. To achieve this, biomedical researchers in Japan have teamed up with their colleagues in astronomy to adapt technology used in space observations for biomedical research.

Led by Atsushi Yagishita, from Kavli IPMU and the National Cancer Center Japan, the researchers developed the IPMU imager – a biomedical imaging system based on a cadmium telluride detector originally developed for hard X-ray and gamma-ray space observations. The imager exhibited high spatial and energy resolution, but could not completely eliminate all noise, especially when imaging radionuclides with similar emission lines. So the researchers improved it further by employing a spectral analysis method used in X-ray astronomy.

Spectroscopic analysis: The data obtained by the IPMU imager (solid black line) and spectra from individual radionuclides (coloured dotted lines) identified and separated using the fitting method. (Credit: Yagishita *et al.*/Kavli IPMU)

When imaging multiple radionuclides, the goal is to generate spectra and images from individual radionuclides, without contamination from other emitters in the sample. The spectroscopic analysis method involves fitting the observed spectra to a model of radionuclide emission lines in specific energy bands. Then, using the determined intensity ratios between the nuclides, any contamination in the image can be calculated and eliminated. The researchers, also from RIKEN and the Institute of Space and Astronautical Science/JAXA, describe the new imaging technique in Nature Biomedical Engineering.

System tests

To verify their new approach, the researchers measured solutions of the gamma-ray emitters iodine-125 (¹²⁵I) and indium-111(¹¹¹In). They examined a pure ¹²⁵I sample, and a sample with the same activity of ¹²⁵I mixed with ¹¹¹In. Their analysis method completely separated spectra from the individual radionuclides and the background component. After processing, ¹²⁵I images from both samples exhibited the same signal intensity, demonstrating the ability to remove noise from other radiation sources and obtain accurate images of only the target radionuclide.

“We assessed the quantitative performance of the images when using multiple radionuclides and confirmed that the intensity of the images correctly represented the radioactivity,” Yagishita tells Physics World.

The researchers also evaluated the spatial resolution of the IPMU imager, by imaging a phantom containing various sized holes filled with solutions of ¹¹¹In, ¹²⁵I or technetium-99m (^99mTc) and reconstructing images in the 21–25, 26–29 and 135–143 keV energy bands, respectively. For all samples, the images clearly showed the 350 and 500 μm holes, while some of the 250 μm holes were also distinguished. The team note that this spatial resolution of roughly 300 μm is similar to that of state-of-the-art small-animal SPECT systems.

In vivo experiments

Finally, Yagishita and colleagues used the IPMU imager to simultaneously visualize thyroid tissue, mandibular lymph nodes and parotid lymph nodes in mice, using ¹²⁵I, ¹¹¹In and ^99mTc, respectively. The spectrum obtained with the IPMU imager clearly showed the separate emission lines from the three radionuclides.

The team then reconstructed images of the three tracers in the relevant energy bands. While the raw images contained noise and ghosting caused by emission from the other radionuclides, the fitting technique identified unwanted radiation sources and produced separate images of each radionuclide in vivo.

In images of ¹²⁵I (in the 26–29 keV energy band), the thyroid gland where iodine accumulates could be accurately delineated by subtracting background noise and ¹¹¹In contamination. The researchers note that the processed image showed no activity in the lymph nodes, where ¹²⁵I does not accumulate.

For comparison, they imaged the same mouse using a conventional small-animal SPECT/CT scanner. The low-energy-resolution spectrum only showed a single peak, while SPECT images in the ¹²⁵I energy window showed the lymph nodes more clearly than the thyroid. These results suggest that the IPMU image is of higher quality than the SPECT image.

In future studies, the team hopes to use the IPMU imager to visualize targeted radionuclide therapy (TRNT) drugs in vivo. “TRNT drugs target specific landmarks on cancer cells and accumulate in cancer cells,” Yagishita explains. “Although the drug pharmacokinetics is ordinarily invisible, it is theoretically possible to visualize TRNT drugs because some emit gamma rays or X-rays. So currently, we are focusing on visualizing the pharmacokinetics of TRNT drugs.”

Why we need to tackle renewable energy’s storage problem

On 16 September 1910 the Canadian inventor Reginald A Fessenden, who is best known for his work on radio technology, published an article in the journal The Electrician about energy storage. “The problem of the commercial utilization, for the production of power, of the energy of solar radiation, the wind and other intermittent natural sources is a double one,” he wrote. “The energy of the sources must first be charged so as to be suitable in form, it must next be stored so as to be available in time.”

As we attempt to lessen the impact of climate change, Fessenden’s comments remain just as pertinent today – if not more so. That’s because many countries are reducing their reliance on fossil fuels by boosting renewable-energy output, whether solar, wind or hydro. And given that so many European nations rely on fossil fuels from Russia, this trend towards green energy is likely to speed up following the invasion of Ukraine that began in late February, having already resulted in a raft of economic sanctions and measures against Russia’s economy.

We can expect that future electric power generation in the UK will be based on variable renewables. This will be primarily wind and solar, possibly supported by nuclear. Unlike fossil fuel and nuclear generation, such renewables will, by their very nature, often produce less power than required – an extreme case being the lack of solar electricity generated at night. One of the biggest challenges for renewables will therefore be to deal with the periods when “the Sun doesn’t shine and the wind doesn’t blow”.

This situation is beautifully captured by the German word Dunkelflaute (meaning dark doldrums). It is critical that we store enough renewable electrical energy that has been produced during periods of excess generation – such as those during favourable wind conditions – for the inevitable Dunkelflaute periods that follow. But this is far from easy. And thanks to detailed studies on future electricity storage requirements and cost, we know it is not cheap either.

We will need adequate excess renewable generation capacity pre-Dunkelflaute to ensure that stored electricity is available over any such period. On a cold winter’s day in the UK, for example, the country requires at least 40 GW of electricity, which equates to about 1 terawatt-hour (TWh). If half of that comes from variable renewables, then on a challenging Dunkelflaute day we will need to have stored 0.5 TWh – assuming that the other 50% is composed of non-renewable sources of gas, nuclear and biomass.

The situation is starker still for a period of 10 successive Dunkelflaute days – a not-uncommon situation in a typical British winter – where we would need some 5 TWh of battery storage. To get an idea of the price tag, we know that the energy company InterGen is currently building a 1 GWh lithium-ion battery-storage facility at DP World London Gateway, a new port on the Thames Estuary in south-east England. It will cost about £300m to build, so a simple extrapolation would mean that having a 5 TWh capacity would be £1.5 trillion. If we depended entirely on renewable electricity, the corresponding battery storage cost would be £3 trillion. This is clearly unfeasible, so what else could we do?

Keeping the lights on

Lithium-ion technology has remained the benchmark for batteries since its discovery four decades ago. Some hope that battery costs will continue to drop – as they have done for the last two decades – or that a new, cheaper kind of battery technology will surely emerge over the next few years. However, there is a worldwide shortage of lithium for building battery storage at scale, while cobalt mining – the material that provides a stabilizing effect in lithium-ion batteries – comes at a heavy environmental price.

Another possibility for storage is hydrogen, which is produced by electrolysis from excess renewable energy generation. It can be converted into electricity through fuel cells or internal combustion engines and can also be used for a range of industrial processes. There are several major hydrogen projects under way in the UK, but it is hard to directly compare the cost of hydrogen with other large-scale storage technologies given the unknown costs of associated conversion technologies and the diverse range of applications.

It is possible, however, to make some qualitative estimates for Dunkelflaute events. If we assume that the storage mechanism is hydrogen and it is converted into electricity by a combination of fuel cells and internal combustion engines with an overall efficiency of 50%, then we would require 10 TWh of stored energy if we are to generate half our electricity from renewables. Using a value of 50 kilowatt hours per kilogram for electrolyser efficiency, that’s equivalent to about 200 million kilograms of hydrogen. So, at a cost of £2 per kilogram that would be about £400m. If we assume an equivalent cost for the process of converting hydrogen to electricity, then the cost of storage and generation in each 10-day Dunkelflaute period would be about £800m. If there are two periods annually and the electrolysers and engines last for 30 years, the total cost would be about £50bn.

Net-zero carbon targets require almost all energy to be provided by renewable electricity. It is essential therefore that we find ways of storing renewable energy during Dunkelflaute periods, but this rather obvious issue is all too often neglected by governments around the world. The costs of either battery storage or energy storage via hydrogen are huge – and even if the costs of batteries can be reduced, big questions about the space, security and safety of such storage installations remain. Decisions are urgently required about the way forward since electricity storage must evolve alongside plans for variable renewable energy.

Honeycomb-like nanopatterning boosts efficiency of ultrathin solar panels

HUD solar cells — **How it works** A HUD nanotextured silicon film surface (left) significantly improves light absorption as shown on the right. (Courtesy: CC BY 4.O/Tavakoli *et al.*)

While solar farms offer clean and reliable energy production, they are space-consuming and remain unfeasible for use in urban centres and outer space. An alternative solar solution for the near-future rests in micron-thin photovoltaics. These devices couple the low-cost manufacturing of silicon with the low weight and flexibility of thin films.

Led by Marian Florescu and Esther Alarcon-Llado, researchers from the University of Surrey, Imperial College London, and AMOLF’s Center for Nanophotonics partnered to address the shortfalls of existing thin film solar panels. Their newly released paper in ACS Photonics describes the properties of the photovoltaic device they devised and tested.

Absorbance through light trapping

The primary issue with increasingly thin solar cells is that with lower thickness comes lower light absorbance efficiency — a majority of light either reflects off the surface or passes through entirely. “Nearly a third of light bounces straight off [the material] without being absorbed and the energy harnessed,” says Florescu, adding, “a textured layer across the silicon slab helps tackle this problem.”

By texturing the surface of the silicon, incident sunlight is instead diffracted into the plane of the silicon thin film, “trapping” the light in the material. Regular ordered patterns are insufficient for this task, as they only diffract sunlight at specific wavelengths and angles of incidence. Alternatively, disordered patterns can scatter a broad range of wavelengths into the film but are not particularly efficient.

Instead, the team employed a specific surface design called hyperuniform disordered (HUD) patterns. HUD media have constrained randomness, which causes them to behave like ordered solids rather than amorphous materials. As such, they are particularly useful in engineering light scattering and diffraction.

HUD solar cells figure 2 — **Test patterns** These three hyperuniform disordered textures were tested for improving thin solar film absorbance. (Courtesy: CC BY 4.O/Tavakoli *et al.*)

The researchers etched HUD patterns into the top 200 nm of a 1-micron-thick silicon film (see above figure). Then they topped the film with a spin-coated polymer resist, which reduced its reflectance by making its index of refraction closer to that of air. Through this combination of techniques, the team recorded a solar absorbance of 66.5% – a staggering increase from the 25.5% absorbance of unadulterated films. This improvement resulted in an absorption equivalent photocurrent of 26.3 mA/cm², which tops the next-best reported result of 19.72 mA/cm² for a cell of similar thickness.

Which HUD pattern?

The team tested multiple HUD patterning designs, including a HUD network, HUD holes and a spinodal pattern (left to right, above image). Each design far improved upon the efficiency and light absorption of unetched native films, with the spinodal pattern slightly edging out the others in efficiency. However, the large-scale implementation of a spinodal texture requires further technological developments to increase patterning resolution of its aperiodic structures. On the other hand, HUD holes and network patterns arise naturally in many physical systems, pointing to their simpler and more scalable fabrication. The team estimates that using similar patterning, researchers will be able to achieve photovoltaic efficiencies above 20% for a 1-micron-thick crystalline silicon cell.

Alarcon-Llado adds, “thinner silicon absorbers are more tolerant to electronic defects compared to the thick counterparts. This means that thin silicon cells with high efficiency could also be made from lower grade silicon, thereby reducing the energy needs for raw silicon purification and reducing their energy pay-back time.” Pairing these qualities with the team’s texturing breakthrough provides hope for future applications where heavier, more rigid bulk-silicon cells are not feasible, including on buildings, cars and in outer space.

Tricks learnt from eye imaging enable video-rate LiDAR

Using lessons learnt from eye imaging, researchers at Duke University have developed a high-precision, high-speed LiDAR (light detection and ranging) system that they claim is 25 times faster than previous LiDAR demonstrations. This provides the system with a frame rate comparable to that of video cameras, greatly improving its potential for use in imaging systems for autonomous technologies such as driverless cars and robots.

Real-time, high-resolution 3D imaging is desirable in many fields, including biomedical imaging, robotics, virtual reality, 3D printing and autonomous vehicles. Currently, a lot of the attention for autonomous systems focuses on LiDAR. Most LiDAR systems are based on time-of-flight: they send out laser pulses and measure the time it takes for reflections to return, using that information to map their surroundings.

In recent years, to try to improve the speed of LiDAR and reduce errors caused by ambient light, researchers have been looking at frequency-modulated continuous wave (FMCW) LiDAR. This technique is like traditional LiDAR, but the system continuously changes, or modulates, the frequency of the laser. When the reflected light returns, as well as measuring the time taken, the system also analyses how its phase has shifted. This provides a far more accurate measure of distance. The frequency pattern created by the modulated laser also makes it easier for the system to distinguish the reflected laser light from other light sources.

FMCW LiDAR shares similar working principles with the medical imaging technique swept-source optical coherence tomography (OCT). Swept-source OCT is primarily used in ophthalmology to obtain high-resolution cross-sectional images of the retina. The technique uses a laser that sweeps over a range of frequencies to achieve 3D volumetric imaging of the back of the eye with a depth of several millimetres.

In recent years, several developments have helped increase the range, speed and resolution of swept-source OCT. In their latest work, biomedical engineer Joseph Izatt and his colleagues applied some of the lessons they have learnt with OCT to create high-speed and high-precision FMCW LiDAR. They discuss their results in Nature Communications.

First, the researchers used an akinetic all-semiconductor sweep source. This does away with any form of mechanical movement in the mechanism that creates the sweep of laser frequencies. In the past, techniques like this have been shown to extend the imaging range of OCT and FMCW LiDAR.

Many FMCW LiDAR systems have a limited scanning speed due to the mechanical mirrors, or other electromechanical systems, that are used to move the laser beam across the field-of-view. One way to address this, and enable high-speed imaging, is to again shift away from mechanical systems to non-mechanical beam scanners with no moving parts. Izatt and his team used a diffraction grating, which separates the laser into multiple beams of different wavelength, all travelling at different angles away from the source. Along with the sweep source, this creates a laser that scans its environment much faster than is possible with mechanical systems.

While these techniques can improve the range of imaging systems, they impact the depth resolution. OCT systems combat this by using more sampling points, faster photodetectors and longer acquisition times. But LiDAR systems do not need the same depth resolution, as they are not probing structures several millimetres deep in a patient’s retina. They just need to scan an object’s surface.

The researchers found that by reducing the number of spectral sampling points and narrowing the laser’s range of frequencies, they could achieve real-time 3D imaging, with a frame rate as high as 33.2 Hz. With this compressed sampling approach, they demonstrated video-rate imaging of various everyday objects, including a moving human hand, with an imaging range of up to 32.8 cm.

The system currently has a few limitations, particularly the short imaging range. This is due to the bandwidth of the available components, but could be extended to about 2 m using a higher-specification commercial photodetector and digitizer. The researchers emphasize, however, that their approach shows the great potential of FMCW LiDAR for a number of imaging applications.

“The biological cell-scale imaging technology we have been working on for decades is directly translatable for large-scale, real-time 3D vision,” Izatt says. “These are exactly the capabilities needed for robots to see and interact with humans safely.” In a 3D world, he explains, “they need to be able to see us as well as we can see them.”

Browse articles by content type

Stretching the limits: novel material produces highly conductive, stretchable bioelectronics