Recent scientific achievements in continental Europe

07.11.2025

Recent scientific achievements in continental Europe

Academician Oleg Figovsky

A group of Finnish scientists conducted an experiment: they restricted the consumption of red and processed meat in the diets of healthy men for six weeks, replacing them with legumes. Six weeks of this diet proved sufficient to promote weight loss and improve blood counts. The randomized controlled trial was conducted by nutrition and health specialists from the University of Helsinki (Finland). The results showed that it is possible to lose weight by partially eliminating sausage, beef, bacon, and similar foods from the diet, replacing them with peas and beans. The article was published in the European Journal of Nutrition. The experiment involved men without bowel, liver, or kidney problems, or other serious illnesses such as diabetes and cancer. Smoking, alcohol abuse, certain medications, and recent international travel were also prohibited. A total of 102 working-age individuals (21 to 61 years old) participated in the trial.

Before and after the experiment, all men underwent a medical examination and provided urine and blood samples for analysis. Participants were divided into a "meat" and a "bean" group. The first group was allowed to eat 760 grams of red and processed meat per week, which was supposed to account for a quarter of their total protein intake. Otherwise, they were advised to maintain a normal diet. In the second group, consumption of such meats was reduced to 200 grams per week. This reduction was compensated for by adding legumes, excluding soy. For example, subjects were allowed to eat pea soup, green peas, beans, vegetable patties made with legumes, and so on. Other familiar protein sources were also allowed: chicken, fish, eggs, and so on.

Otherwise, all participants were free to eat as before, without any special restrictions. The researchers provided recommended foods (red meat and legumes) along with recipes. Subjects were required to consume them within a designated time frame, which they reported by filling out a food diary in a mobile app. The app was also used to track their compliance with the experiment. After six weeks, men in the legume group lost an average of one kilogram. They also experienced lower levels of total and "bad" cholesterol (LDL), which is detrimental to heart and vascular health. Iron levels increased, despite the reduction in red meat.

Austrian and French physicists have developed a method for detecting objects hidden in thick layers of granular media and suspensions using random matrix theory and the fingerprint of the target object. The scientists tested their mathematical model in three experiments, demonstrating its high accuracy (in one experiment, the false-positive rate was less than 10-8). They also proposed using the model in medicine for diagnosing neuromuscular diseases. The results were published in the journal Nature. To peer beneath the surface of a disordered medium, such as a layer of bulk material, the physicists use multi-element devices—arrays of detectors that record transmitted or reflected signals. Since heterogeneous media can be viewed as the realization of random processes, random matrix theory also aids the researchers. All this allowed scientists to visualize objects hidden beneath a layer of porous or granular material: in optics, wavefront shaping methods were used, and in acoustics, time-reversing mirrors were adapted.

However, in more complex cases, these approaches are inapplicable, as it is generally impossible to verify a one-to-one correspondence between each matrix eigenstate and each target in the material. This is especially true if the targets embedded in the bulk medium do not reflect the signal very well, and the surrounding environment is quite chaotic. In other words, in the case of strong multiple scattering, it is impossible to fully compensate for the influence of disorder. Physicists from France and Austria, led by Alexandre Aubry of the Higher School of Industrial Physics and Chemistry, addressed this problem by proposing to search for correlations in scattered waves instead of canceling out the chaos in a thick bulk material.

To do this, the scientists constructed an invariant operator from the transmission matrix in a disordered medium and the conjugate transposed matrix for a homogeneous reference medium. To locate an object within a material, it is sufficient to apply this operator to the object's supposed imprint. The researchers proved that the eigenstates of the proposed transformation are modes that remain unchanged after scattering and demonstrated a pattern of the transmitted field behind the medium under study. This pattern, in turn, turned out to be identical to that of purely ballistic waves, regardless of the number of re-emissions and reflections.

The physicists validated their mathematical model using three experiments. The first involved ultrasonic imaging of two metal spheres, 10 and 8 millimeters in size, placed in a granular suspension of water and glass beads 300-315 micrometers in diameter. The authors placed a two-dimensional array of 1024 detectors on top of the medium, which detected both objects with high reliability (the false alarm rate was 10-8) but with a strong echo signal. In addition to this experiment, the physicists studied how the invariant operator changed when one of the spheres sank in the suspension—the scientists tracked the object's trajectory with good accuracy.

In the second experiment, the researchers used a lesion marker commonly used to monitor breast tumors in clinical settings. They placed this marker in water and then calculated the ultrasound trace generated by the water foam. Finally, the third experiment involved mapping the local anisotropy of the fibrous medium: the physicists visualized muscle tissue using an array of 256 ultrasound transducers operating in the 5.5-9.5 megahertz frequency range. The result allowed the scientists to almost completely determine the spatial characteristics of each muscle. The authors of the new method noted that the second and third experiments could prove extremely useful for medical applications in the future. For example, the results of the latter experiment could be used to monitor neuromuscular diseases, as well as myocardial fiber abnormalities that occur in the early stages of cardiomyopathy and fibrosis.

An international team of researchers has developed a new method for producing aniline derivatives—compounds used in the manufacture of drugs, dyes, and electronic materials. Instead of toxic and expensive reagents, they proposed using tetrahydrofuran, which can be obtained from renewable raw materials, and the reaction was carried out in the presence of readily available cobalt salts and syngas. This approach reduces hazardous waste and makes production simpler and more environmentally friendly. The study was published in the journal ChemSusChem. Anilines are compounds in which the nitrogen atom is bonded to an aromatic ring. This structure makes the molecule stable and easily modified: reactions are possible at both the nitrogen atom and the aromatic ring, while the core framework is preserved. This allows the basic structure to yield a wide range of derivatives with diverse properties—from painkillers and decongestants to dyes, polymers, and electronic materials.

Aniline derivatives are typically produced using alkyl halides — reagents with chlorine or bromine atoms attached to the carbon chain. This method is reliable, but comes with a number of problems: alkyl halides are expensive, toxic, and produce waste that must be disposed of. Furthermore, such substances corrode the metal of industrial reactors and pipelines, causing equipment to fail faster. Scientists have proposed a new method for synthesizing aniline derivatives. Instead of alkyl halides, they used tetrahydrofuran (THF), a common solvent that typically creates a reaction medium without itself participating. In the new method, THF acted as a reactant: its ring opened, forming an intermediate compound that reacted with aniline to produce N-alkylanilines.

The reaction was catalyzed by cobalt salts — a more readily available and inexpensive metal than the noble metals often used as catalysts. Additionally, the scientists used syngas—a mixture of hydrogen and carbon monoxide. In industry, it's used, for example, to produce fuel and methanol, and here it helped initiate the desired transformations and replaced more hazardous reagents. Thanks to this set of conditions, the researchers obtained the target compounds with good efficiency.

"The method proved to be versatile. Depending on the reaction conditions, it's possible to obtain compounds with either one or two carbon chains. This is important because these variants have applications in various fields," explains Associate Professor Evgeniya Pod'yacheva. Furthermore, the researchers succeeded in synthesizing a compound used to produce tetracaine, a well-known local anesthetic. Thus, the method is suitable not only for laboratory experiments but also for pharmaceutical production. "We've shown that a common solvent can become a valuable reagent. The use of THF, which can be obtained from biomass, makes the process sustainable and less dependent on non-renewable resources. Furthermore, the method reduces waste and eliminates toxic reagents. In the future, it can be expanded to synthesize other aniline derivatives in demand in medicine, materials science, and electronics,” says Professor Denis Chusov.

Many technologies, when they first appeared, seemed like mere gimmicks, nothing more than entertainment. However, over time, their development and refinement have shown their enormous potential for application across a wide range of industries. Such technologies are so widespread that they are no longer surprising. Cameras are a prime example of this. They are used in mobile devices, security systems, research, medicine, space exploration, and more. In other words, they are everywhere. The lens plays a crucial role in any camera's operation, and improving its performance can expand the device's capabilities. A group of scientists from Friedrich Schiller University Jena (07743 Jena, Germany) have developed a new type of lens — a multilayer metal lens — that could revolutionize photo and video recording devices.

Many technologies, when they first appeared, seemed like mere gimmicks, nothing more than entertainment. However, over time, their development and refinement have shown their enormous potential for application across a wide range of industries. And the prevalence of such technologies is so widespread that they are hardly surprising. Cameras are a prime example of this. They are used in mobile devices, security systems, research, medicine, space exploration, and more. In other words, they are everywhere. The lens plays a vital role in any camera's performance, and improving its performance can expand the device's capabilities. A group of university scientists has developed a new type of lens — a multilayer metalens — that could revolutionize photo and video recording.

Metalens, composed of flat arrays of subwavelength scatterers known as meta-atoms, have quickly become an advanced alternative to traditional diffractive and refractive lenses. By precisely engineering these scattering elements, metalens provide unrivaled subwavelength control over the polarization, amplitude, phase, and frequency of incident light. This advanced control could radically improve the efficiency and compactness of next-generation optical systems. Enormous progress has been made in monochromatic operation: high-performance devices have already been demonstrated for many different wavelengths and sizes up to tens of centimeters. However, the main limitation in achieving these goals for broadband or multi-wavelength metalenses is the difficulty of scaling to centimeter-sized apertures, which are critical for practical devices. This difficulty arises because the maximum required linear phase dispersion, or group delay (GD), which must be applied to the incident field for perfect achromatic focusing, is proportional to the lens diameter.

The limited GD values currently achievable in single-layer nanostructured metalenses with existing dielectrics impose an upper limit on the diameter, which is significantly below the centimeter-scale dimensions required in many real-world applications. This leads to significant chromatic aberration as the lens diameter increases (1a). To overcome this limitation, a recent study proposed the concept of a multi-zone dispersive metalenses, in which the applied GD is folded into discrete zones limited by the achievable maximum value of meta-atoms. Each zone then achieves achromatic focusing independently, and the phase shifts between the zones are optimized so that they constructively interfere at the focal plane over the entire range of design wavelengths. A key limitation of this approach is the need to use geometric phase to independently control the phase and GD across the surface. Thus, the focusing response of the lens is sensitive to polarization. For systems operating with unpolarized sources, polarization sensitivity effectively halves the focusing efficiency and results in out-of-focus scattered light, which degrades system performance.

Although a truly large, broadband, polarization-insensitive metalens platform has not yet been demonstrated, progress has been made with multi-wavelength devices. Here, the GD limitation is relaxed, and focusing over scalable, large areas can be achieved simply by applying a phase profile to the lens at each wavelength of interest, using an arbitrary phase dispersion characteristic. Since most artificial sources operate at discrete wavelengths, multi-wavelength rather than achromatic operation also does not significantly limit the variety of potential applications. Researchers have proposed polarization-sensitive designs based on geometric phase and narrow q-BIC resonances, as well as several multi-wavelength polarization-insensitive variants applicable to large areas. These include ring interference, multilayer meta-atoms, spatial alternation, and multi-resonance approaches. For designs based on ring interference, the maximum expected efficiency is low and limited by the 10% efficiency of an ideal binary Fresnel zone plate. Multilayer meta-atom (1c) designs, when simulated, are capable of achieving focusing efficiencies greater than 60% at a numerical aperture (NA) of 0.2; however, they suffer greatly from manufacturing errors associated with interlayer spacing and alignment variations.

Spatially alternating and multiresonant designs (1b) tend to suffer from a large spread of focusing performance across the calculated wavelengths, typically with minimum focusing efficiencies below 40% even in simulations. This effect can be attributed to a combination of unaccounted near-field interactions due to the violation of the locally periodic approximation (LPA) and the sparser spatial phase sampling at short wavelengths. To overcome these limitations, a recent study utilized an improved LPA-based inverse design scheme for large areas, enabling the creation of centimeter-scale visible-light metalenses operating at red, green, and blue (RGB) wavelengths, insensitive to polarization, and exhibiting nearly identical focusing efficiencies across the operating wavelength range. However, the maximum focusing efficiency achieved in the simulations remained low—approximately 24%—due to the use of cross-polarization-converting meta-atoms, which have inherently lower efficiency compared to isotropic meta-atom platforms.

In the present study, the researchers presented a new design for a multi-wavelength, polarization-insensitive metalens operating in the near-infrared (NIR) range. The design consists of multiple layers of Huygens metasurfaces, where each layer modulates only a specific wavelength, while providing high throughput and low phase distortions at other wavelengths (1d). This approach effectively eliminates the problem of sparse phase sampling characteristic of spatially interleaved designs and minimizes the difficulties in maintaining the applicability of the local periodicity approximation (LPA) simultaneously over a wide range of wavelengths. Furthermore, since the layers are assumed to be separated by a macroscopic distance in the far field, this method is inherently robust to layer misalignment. It is also highly compatible with modern large-area lithography technologies, as each layer can be fabricated as a separate metasurface chip and then simply combined into the final device. As a proof-of-concept, a meta-lens operating at 2000 and 2340 nm was designed, achieving focusing efficiencies of 65% and 56% at a numerical aperture (NA) of 0.11 in simulations.

In Huygens metasurfaces, an array of meta-atoms supports local spectrally overlapping electric (ED) and magnetic (MD) dipole resonances that interfere to form directional forward scattering with a phase span of 2π. The archetypal meta-atom is a disk with a small aspect ratio, where the spectral position of the resonances and, consequently, the phase of the transmitted field are controlled by varying the disk radius.

To implement multi-wavelength functionality in this mode, two characteristics of this interaction were exploited. The first is that in the immediate spectral region near the resonance, the scattering cross section is small and the transmission is high. This is due to the fact that dipole resonances are low-order resonances, and within the local frequency band, the structure does not support other Mie resonances.

The second characteristic is that far from the resonance, the field is not strongly localized within the meta-atom and is not sensitive to its geometry. In this case, the metasurface behaves more like a thin film with a small linear phase dispersion. Together, these properties allow Huygens metasurfaces to simultaneously provide 2π phase coverage at the resonant wavelength and limit phase distortions far from the resonance. This is achieved primarily by decoupling the fill factor of the meta-atom array from the spectral position of the resonances. Under these conditions, the phase accumulated off-resonance (determined by a thin-film-like interaction) can remain approximately constant within the meta-atom library.

Unfortunately, the well-known Huygens disk geometry is not optimal in this case, as it requires the fill factor of the array to be proportional to the applied resonant phase, since the disk radius remains the only geometric degree of freedom. To overcome this, meta-atom libraries were constructed using an inverse projection scheme, which allows for significantly more geometric degrees of freedom while maintaining rotational symmetry and polarization insensitivity.

In this study, as a demonstration of principle, a design for a multilayer Huygens metalens system with two operating wavelengths in the near-infrared (NIR) range was presented. The system can be expanded to include additional wavelength channels in any optical range and material system supporting Huygens resonances. However, there are physical limitations on the number and relative positions of operating wavelengths. Based on an estimate based on a minimum separation between operating wavelengths equal to twice the full width at half maximum (FWHM) of the Huygens resonance mode, the multilayer Huygens metasurface system can be expanded to a maximum of approximately five operating wavelengths using the material parameters employed in this study.

To create multiwavelength libraries, an inverse projection scheme with shape optimization within the local periodicity approximation (LPA) was used. This scheme is based on an efficient meta-atom parameterization method, in which the meta-atom geometry is formed by interpolating an array of control points within a period to construct a control surface. A level-set operation is then applied to this surface to define the meta-atom boundary. This highly flexible method allows for the implementation of arbitrary symmetries and is ideal for creating polarization-insensitive meta-atoms. Additional technological constraints are imposed after obtaining the geometry using morphological or other algorithmic operations.

Meta-atoms are modeled using rigorous conjugate wave analysis (RCWA) at the wavelengths of interest. The resulting complex transfer parameter is then used to construct a multi-objective function that quantifies the meta-atom's characteristics. The formulated objectives take into account the required multi-wavelength operation both at the operating wavelengths and within ±10 nm to avoid strong frequency dependence in the final designs, as this increases sensitivity to process defects. Multi-wavelength libraries are then generated using the meta-atom checkpoint array as optimization parameters in a gradient-free multi-objective evolutionary algorithm (MOEA). This allows a library with 2π phase coverage, including approximately 8000 objective function evaluations, to be constructed in a single optimization run.

To accelerate the MOEA optimization step, meta-atom hyperparameters (array period, height, and target phase for the off-resonant wavelength) were first determined using a simple grid-based optimization using Latin hypercubic sampling (LHS) to obtain the initial data set. The final libraries are then formed using an additional grid-based search over the entire MOEA solution archive. This is explained by the fact that upon completion of the MOEA optimization, solutions are obtained that cover the entire phase space of the modulation wavelengths, but only a subset of these solutions corresponding to the target phase levels of the modulation library (Nl) is needed.

To characterize the performance of the multi-wavelength metalens, eight-layer meta-atom libraries were used to generate two metasurfaces. Each metasurface implements the hyperbolic phase profile from Equation 1 at its resonant wavelength, determined using a simple phase lookup table. To remain within the optimal angular bandwidth of the meta-atom libraries, a lens phase profile with a numerical aperture (NA) of 0.11 was used. The multilayer device was then simulated using the finite-difference time-domain (FDTD) method. Due to limited computational resources, the metalens diameter was chosen to be 125 μm. The metalens layers are separated by a distance of 2.34 μm. This distance was chosen to position the metalenses in their far-field (within the cladding medium), thereby avoiding inter-layer coupling effects. Thus, each layer could be modeled separately: the field obtained above the first layer was used as an excitation source for the second layer.

In both cases, the far field exhibits sharp focusing with a negligible change in focal length, indicating that the off-resonance response of the layers has no significant effect on the resulting resonant phase profile. The main difference is the appearance of increased intensity maxima located closer to the lens. This is due to the diffraction effect of the off-resonance layers, which have a small phase deviation. They impose a weak radial periodicity on the Fresnel zones of the lens, leading to the scattering of some of the energy to higher orders. The obtained results demonstrate that the layers operate independently, and therefore the system is resilient to misalignment: the required alignment accuracy between the layers for optimal performance is determined by the scale of the entire optical system, not the lattice period of the meta-atoms.

The shape of the PSF for both wavelengths is also virtually identical between the single-layer and full multilayer devices. The main difference is the reduction in focusing efficiency. Here, efficiency is defined, using the generally accepted approach, as the ratio of the optical power passing through an aperture with a diameter three times larger than the full width at half maximum (FWHM) of the PSF to the total power incident on the lens. A decrease in efficiency is also associated with off-resonance diffraction.

Nevertheless, the multilayer device still provides highly efficient polarization-independent focusing: 65% and 56% for incident fields at 2000 and 2340 nm, respectively. It should also be noted that due to the limited diameter of the metalens (125 μm), diffraction-limited focusing through an ideal lens with such NA has a maximum efficiency of 86%. Thus, the relative focusing efficiencies are 76% and 65%. Furthermore, larger centimeter-scale metalenses are expected to exhibit improved performance, as diffraction effects are reduced and the grating size increases, making LPA more accurate over larger lens areas.

The obtained results demonstrate focusing that is close to diffraction-limited, while the modulation transfer function (MTF) is almost completely identical to that of an ideal lens at both wavelengths. A multilayer metalens was found to provide near-perfect focusing performance in numerical simulations, with the main limiting factor being a decrease in overall efficiency. During development, reverse engineering of the meta-atom geometry was applied, taking into account technological constraints. This resulted in the creation of libraries with 8 phase levels and the design of a metalens operating at 2000 and 2340 nm, with a numerical aperture of NA = 0.11. The multilayer device was simulated using the FDTD method and demonstrated near-perfect focusing quality (MTF close to the diffraction limit), with absolute focusing efficiencies of approximately 65% and 56%, respectively, which corresponds to ~76% and ~65%, respectively, relative to an ideal lens of the same size. The scientists note that the potential application of this approach is quite impressive. Such multi-wavelength, polarization-insensitive metalenses can find their place in compact optical systems, sensors, spectroscopy, portable optics, and other applications where small size and high functionality are important.