Experimental Efforts Research Articles

Combining High-Throughput Experiments and Active Learning to Characterize Deep Eutectic Solvents.

The high tunability of deep eutectic solvents (DESs) stems from the ease of changing their precursors and relative compositions. However, measuring the physicochemical properties across large composition and temperature ranges, necessary to properly design target-specific DESs, is tedious and error-prone and represents a bottleneck in the advancement and scalability of DES-based applications. As such, active learning (AL) methodologies based on Gaussian processes (GPs) were developed in this work to minimize the experimental effort necessary to characterize DESs. Owing to its importance for large-scale applications, the reduction of DES viscosity through the addition of a low-molecular-weight solvent was explored as a case study. A high-throughput experimental screening was initially performed on nine different ternary DESs. Then, GPs were successfully trained to predict DES viscosity from its composition and temperature, showcasing the ability of these stochastic, nonparametric models to accurately describe the physicochemical properties of complex mixtures. Finally, the ability of GPs to provide estimates of their own uncertainty was leveraged through an AL framework to minimize the number of data points necessary to obtain accurate viscosity modes. This led to a significant reduction in data requirements, with many systems requiring only five independent viscosity data points to be properly described.

Theoretical analysis of the reactions induced by interaction of Li6 with H3 and He3 nuclei

We determine cross sections and astrophysical S factors of the reactions generated in collisions between Li6 and H3 and between Li6 and He3. A microscopic three-cluster model is employed to study the dynamics of reactions occurring in the mirror nuclei Be9 and B9. In a previous study [], this model was successfully applied to investigate the resonance structure of Be9 and B9, as well as reactions induced by the interaction of deuterons with Li7 and Be7. A fairly good agreement between theoretical results and available experimental data was achieved. To the best of our knowledge, appropriate experimental data for the astrophysical S factors of the reactions generated by the interaction of Li6 with H3 and with He3 are currently unavailable. Thus, our results can serve as a guideline for future experimental efforts. Published by the American Physical Society 2024

Response Surface Design Models to Predict the Strength of Iron Tailings Stabilized with an Alkali-Activated Cement

Tailing storage facilities are very complex structures whose failure generally leads to catastrophic consequences in terms of casualties, serious environmental impacts on local biodiversity, and disruptions in the mineral supply. For this reason, structures at risk must be reinforced or decommissioned. One possible option is its reinforcement with compacted filtered tailings stabilized with binders. Alkali-activated binders provide a more sustainable solution than ordinary Portland cement but require an optimization of the tailing–binder mixture, which, in some cases, can lead to a substantial experimental effort. Statistical models have been used to reduce the number of those experiments, but a rational design methodology is still lacking. This methodology to define the right mixture for a required strength should consider both the mixture components and in situ conditions. In this paper, response surface methods were used to plan and interpret unconfined compression strength test results on an iron tailing stabilized with alkali-activated binders. It was concluded that the fly ash content was the most important parameter, followed by the liquid content and sodium hydroxide concentration. From the obtained results, several statistical models were defined and compared according to the definition of a strength prediction model based on a mixture index parameter. It was interesting to observe that models with the porosity cement index still provide reasonable adjustment even when different tailings’ water contents are considered.

Two photons everywhere.

We discuss two-photon physics, taking for illustration the particular but topical case of resonance fluorescence. We show that the basic concepts of interferences and correlations provide at the two-photon level an independent and drastically different picture than at the one-photon level, with landscapes of correlations that reveal various processes by spanning over all the possible frequencies at which the system can emit. Such landscapes typically present lines of photon bunching and circles of antibunching. The theoretical edifice to account for these features rests on two pillars: (i) a theory of frequency-resolved photon correlations and (ii) admixing classical and quantum fields. While experimental efforts have been to date concentrated on correlations between spectral peaks, strong correlations exist between photons emitted away from the peaks, which are accessible only through multi-photon observables. These could be exploited for both fundamental understanding of quantum-optical processes as well as applications by harnessing these unsuspected resources. This article is part of the theme issue 'Celebrating the 15th anniversary of the Royal Society Newton International Fellowship'.

Virtual rapid prototyping of materials with deep learning: spatiotemporal stress fields prediction in ceramics employing convolutional neural networks and transfer learning

ABSTRACT Additive manufacturing offers a solution for producing advanced ceramics with complex geometries by enabling precise control over geometry, microstructure, and composition. By leveraging deep learning, rapid prototyping and evaluation of printed ceramic parts become feasible. This study employs convolutional neural networks and transfer learning to predict spatiotemporal fields in ceramics, using synthetic datasets generated from X-ray computed tomography (micro-CT) and finite element analysis. The novel approach integrates spatiotemporal factors into deep learning models, enhancing the prediction of stress and damage evolution, and ultimately providing deeper insights into material behaviour and performance. Additionally, we introduce a target loss training strategy, which focuses on achieving a specific accuracy level, thus reducing training time while maintaining high precision. The proposed deep learning framework achieves accurate predictions of spatiotemporal stress fields, and captures fracture initiation and propagation behaviours. This method facilitates rapid prototyping and advances material design and evaluation processes while significantly reducing experimental efforts.

An experimental and numerical effort on the vibration behavior of additively manufactured recycled polyethylene terephthalate glycol components

AbstractThe importance of recycling engineering components and thus obtaining low‐cost production solutions has become prominent in today's world. In this study, the mechanical and dynamic behaviors of three‐dimensional‐printed recycled polyethylene terephthalate glycol (RePET‐G) beams were investigated numerically and experimentally for the first time in the literature. Initially, the governing equations of the beams were determined according to the Bernoulli–Euler beam theory, and these equations were numerically solved using the differential quadrature method and ANSYS program. Subsequently, to validate the accuracy of the numerical models, the obtained natural frequencies were compared with experimental results. It was observed that the numerical results showed good agreement with the experimental results. Finally, the effects of beam length, infill rate, and building direction on the natural frequencies of RePET‐G beams were investigated. The outcomes showed that as the beam length changed, natural frequencies were significantly affected. Increasing the infill rate, especially for beams with vertical building direction, from 20% to 100% led to a slight decrease in the natural frequency values of the structure. Moreover, it was found that for beams with an infill rate of 100%, the natural frequency values obtained in the horizontal building direction were higher than those obtained in the vertical building direction.Highlights Printable recycled filaments have great potential for vibration applications. Sample length affects the first natural frequency value of RePETG parts. Differential quadrature and ANSYS methods can be utilized for the vibration. For the 3D‐printed samples, rising infill rate causes a natural frequency drop.

Effect of ion-specific water structures at metal surfaces on hydrogen production

Water structures at electrolyte/electrode interfaces play a crucial role in determining the selectivity and kinetics of electrochemical reactions. Despite extensive experimental and theoretical efforts, atomic-level details of ion-specific water structures on metal surfaces remain unclear. Here we show, using scanning tunneling microscopy and noncontact atomic force microscopy, that we can visualize water layers containing alkali metal cations on a charged Au(111) surface with atomic resolution. Our results reveal that Li+ cations are elevated from the surface, facilitating the formation of an ice-like water layer between the Li+ cations and the surface. In contrast, K+ and Cs+ cations are in direct contact with the surface. We observe that the water network structure transitions from a hexagonal arrangement with Li+ to a distorted hydrogen-bonding configuration with Cs+. These observations are consistent with surface-enhanced infrared absorption spectroscopy data and suggest that alkali metal cations significantly impact hydrogen evolution reaction kinetics and efficiency. Our findings provide insights into ion-specific water structures on metal surfaces and underscore the critical role of spectator ions in electrochemical processes.

Experimental and modeling insights into mixing-limited reactive transport in heterogeneous porous media: Role of stagnant zones

The understanding of mixing-controlled reactive dynamics in heterogeneous porous media remains limited, presenting significant challenges for modeling subsurface contaminant transport processes and for designing cost-effective environmental remedial efforts. The complexity of accurately observing, measuring, and modeling mixing-limited reactive transport has led to inadequate exploration of these critical processes. This study investigates the mixing and reaction kinetics affected by stagnant zones, which are commonly found in alluvial aquifers-aquitards and fracture-matrix systems. By conducting experiments involving conservative and bimolecular reactive transport through porous media within translucent chambers filled with two sizes of glass beads and under varying flow rates, we explored the effects of grain size and hydrodynamic conditions. Using a high-resolution camera, we monitored the concentration changes of conservative and reactive tracers, with subsequent interpretation through three-dimensional numerical simulations. The outcomes revealed the emergence of distinct mixing interfaces within both mobile and stagnant zones, culminating in a bi-peaked plume formation. Notably, the mixing and reaction times in media containing stagnant zones were found to be approximately 10 times longer than in homogeneous media. These findings, through experimental and modeling efforts, advance our understanding of mixing-limited reactive transport phenomena within heterogeneous media, underscoring the significant role of stagnant zones—a topic previously underexplored.

Experimental investigation and micromechanical analysis of glass fiber reinforced polyamide 6

Achieving process stability in the thermoforming of fiber reinforced polymer materials (FRPs) for aerospace or automotive manufacturing is usually associated with a costly trial-and-error process, where experimental boundary conditions and other influencing factors, such as, for example, material composition, need to be adjusted over time. This is especially true when material phenomena on the microlevel, such as the crystallization kinetics of the polymer matrix or resulting stresses from temperature gradients, are the cause of the process instability. To reduce the experimental effort and reliably predict the material behavior during thermoforming, finite element simulation tools on multiple scales are a useful solution. Hereby, incorporating micromechanical phenomena into the model approaches is crucial for an accurate prediction by further reducing the deviation between simulation and experiment, in particular with regard to the underlying nonlinear material behavior. In this work, unit cell simulations on the microscale of a unidirectional glass fiber reinforced polymer (UD GFRP) are conducted to predict effective thermomechanical properties of a single material ply and ascertain the effect of individual ply constituents on the homogenized material behavior. The polymeric matrix material model used was identified in a prior publication with experimental data at various temperatures for polyamide 6 blends with varying degrees of crystallinities. Various randomization methods are tested to generate the unit cells and replicate the composites’ random fiber distribution, with a focus on process automation. The simulative results are successfully compared to an experimental study on glass fiber reinforced polyamide 6 tested at various temperatures, demonstrating the potential of the approach to reduce both time and cost required for material characterization. Finally, the unit cells are used to generate a database to predict untested load cases that will be used in future work to characterize a homogenized macroscopic material model.

Prediction of quantitative in-situ local corrosion via deep learning

A deep understanding of corrosion behavior is critical for improving steel durability and reliability. Typically, it requires significant experimental effort and extensive measurements over a long period of time. Therefore, exploring the time-dependent corrosion at an early stage to predict its progress at a later stage can effectively understand it. In this study, we quantitatively predicted later stages of local corrosion behavior using deep learning methods based on the early stage topographical information. Furthermore, we predicted and visualized the formation, growth, and accumulation of the particle-like oxides. Our proposed method can be extended to other types of corrosion-resistant electrochemical materials.

Modeling Polymer Microencapsulation Processes Using CFD and Population Balance Models

Computational fluid dynamics (CFD) modeling has emerged as a valuable tool for investigating complex processes like microencapsulation. This paper aims to validate the ability of CFD simulations to predict particle size distribution in a polymer microencapsulation process. The CFD modeling approach employed a Eulerian multiphase framework, incorporating a discrete population balance model to track the evolution of the droplet population. A realizable k-ε turbulence model and a multiple reference frame strategy were utilized to capture the system’s flow dynamics. The results reveal that while the CFD simulations align well with experimental data at higher agitation speeds (>10,000 rpm), discrepancies arise at lower speeds (<7500 rpm), indicating a challenge in accurately capturing turbulent viscous regimes. Despite these challenges, the CFD model demonstrates robust predictive capabilities for droplet formation and distribution in microencapsulation processes, validated by error margins within the acceptable limits. The validated model can be used as a reliable tool to guide experimental efforts and optimize process parameters, contributing to an enhanced understanding and control of microencapsulation processes.

Aerodynamic Characterization of a Hypersonic Projectile Using Acceleration-Based Measurements and Numerical Methods

Armor-piercing fin-stabilized projectiles stand out by their exceptionally high slenderness ratios and ground-level flight at super- and hypersonic speeds. As space constraints limit the integration of measurement equipment into such slender test models, nonintrusive measurement techniques become favorable. The present analysis demonstrates a renewed approach to the free-flight technique, which sets models into unconstrained flight through the test section. It enabled the evaluation of the quasi-steady drag, lift, and pitching moment characteristics, and, to some extent, also the dynamic pitch damping characteristics. An alternative to the free-flight technique is the free-oscillation technique, which limits the motion to a rotation around the center of gravity. The free-oscillation technique allowed a higher-fidelity analysis of the static and dynamic pitching moments. Both methods were based on the analysis of the accelerations derived from trajectory tracking. This acceleration-based approach enabled the evaluation of the highly nonlinear characteristics. The high slenderness of the investigated projectile leads to a significant contribution of the viscous forces to the overall drag. These viscous forces are sensitive to the laminar-turbulent boundary-layer state, as well as the thermal boundary conditions. The application of a high-resolution schlieren system enabled the assessment of the local laminar-turbulent boundary-layer state, while the use of low- and high-enthalpy testing facilities enabled the assessment of the aerothermal influence. Steady-state Reynolds-averaged Navier–Stokes simulations, which included laminar-turbulent transition modeling, were employed to replicate the results of the experimental efforts.

From docking to dynamics: Unveiling the potential non-peptide and non-covalent inhibitors of Mpro from natural products

MotivationThis study aims to investigate non-covalent and non-peptide inhibitors of Mpro, a crucial protein target, by employing a comprehensive approach that integrates molecular docking, molecular dynamics simulations, and top-hits activity predictions. The focus is on elucidating the non-covalent and non-peptide binding modes of potential inhibitors with Mpro. MethodsWe employed a semi-flexible molecular docking methodology, binding score and ADME screening, which are based on structure, to screen compounds from CMNPD and HERB in silico. These methodologies allowed us to find potential candidates depending on their binding values and interactions with the binding site of main protease. To further evaluate the stability of these interactions, we conducted molecular dynamics simulations and calculated binding energies. Ultimately, a top-hits activity prediction method was employed to prioritize compounds based on their predicted inhibitory potential. ResultsThrough a combination of binding energy calculations and activity predictions, we identified six potential inhibitor molecules exhibiting promising activity against Mpro. These compounds demonstrated favorable binding interactions and stability profiles, making them attractive candidates for further experimental validation and drug development efforts targeting Mpro.

Prediction of phase and tensile properties of selective laser melting manufactured high entropy alloys by machine learning

Selective laser melting (SLM) for high entropy alloys (HEAs) holds significant promise in commercial applications, and substantial experimental research efforts have been directed toward this domain. To take advantage of the reported experimental data of SLM manufactured (SLM-ed) HEAs and reduce unnecessary experimentation, this study incorporates machine learning (ML) techniques for the phase and tensile properties prediction of SLM-ed HEAs, thus presenting a novel avenue for accelerating the discovery of new SLM-ed HEAs. Through the adjustment of material descriptors and ML models, a model has been developed with an impressive accuracy of 81.58 % in distinguishing between face-centered cubic (FCC), body-centered cubic (BCC), and dual-phase (FCC+BCC) of SLM-ed HEAs. Additionally, optimized models have been devised for the prediction of tensile properties of SLM-ed HEAs, namely ultimate tensile strength (UTS) and yield strength (YS), achieving noteworthy MAPEs (mean absolute percentage errors) of 20.43 % and 20.25 %, respectively. Furthermore, several HEAs were fabricated using SLM, and the experimental outcomes exhibited a favorable alignment with the predicted results. These efforts carry significant implications in advancing the utilization of SLM for HEAs.

Unlocking inherent values of manufacturing metadata through digital characteristics (DC) alignment

Data form the backbone of manufacturing sciences, initiating a revolutionary transformation in our understanding of manufacturing processes by unravelling complex scientific patterns embedded within them. Digital characteristics (DC) is defined as a strategic framework mapping the manufacturing metadata and integrates essential information across the entire spectrum spanning from the design, manufacturing, and application phases of manufactured products. By carrying these inherent distinctive features, DC serves as the ‘DNA’ for every manufacturing process. Through enormous experimental and simulation efforts, a digital characteristics space (DCS) was established to provide access to the up-to-date and information-rich DC repository containing over 140 manufacturing processes. In digital manufacturing, sensing networks play a pivotal role in metadata acquisition, contributing nearly 2000 petabytes of metadata annually. However, an overwhelming majority-nearly 100 %-of the data collected through sensing networks can be categorised as ‘fragmental data’, encompassing only a few (e.g., 1–2) essential pieces of information. Moreover, the current absence of efficient metadata identification methods presents an emerging and critical need to enable industry to unlock the full potential of manufacturing metadata. To this end, the authors of the present paper developed a physics-based alignment filter, considering DCS as an alignment reference similar to the ‘GenBank’. Specifically, the origins of naturally unattributed fragmental data were identified with an overall probability exceeding 82 % with a minimum length of 10 data points. The probability increased to 99 % when aligning the fragmental data with length of 100 data points. This was realised by comparing the thermo-mechanical DC of fragmental data with their counterparts stored in the DCS. Subsequently, we analysed the distinct DC of this identified manufacturing process to facilitate digitally-enhanced research. This study introduces a pioneering methodology developed to extract latent values embedded in manufacturing metadata derived from unattributed fragmental data. By revolutionising insights into advanced manufacturing sciences, our work provides an enabling approach for identifying and leveraging fragmental data sourced from sensing networks. This empowers the exploration of manufacturing metadata, promising transformative implications for the field.

Quark core-gluon model for heavy hybrid baryons

Besides the ordinary hadrons, quantum chromodynamics (QCD) allows the existence of states in which excitations of the gluonic field can play the role of valence particles, either alone in a glueball, or coupled to quarks in a hybrid. So, hybrid baryons, made of three quarks and a gluon, can exist. Till now, there is no experimental evidence for such exotic hadrons but experimental efforts are being made to search for them at CEBAF Large Acceptance Spectrometer. In this work, a hybrid baryon is considered as a two-body system composed of a color octet three-quark core and a gluon, interacting via a QCD-inspired interaction. A semirelativistic potential model is built in which the dominant interaction is a potential simulating the flux tube confinement, and the Casimir scaling is assumed to link interactions between triplet and octet color sources. This picture is similar to the quark-diquark description for baryons. It is chosen in order to take properly into account the helicity of the gluon. Only cccg and bbbg states are considered because the strong mass asymmetry between the quark core and the gluon is expected to favor the formation of the core. As the results for heavy hybrid baryons seem relevant, we consider this paper as a proof of concept which can be extended for the study of light hybrid baryons. Published by the American Physical Society 2024

Disorder-Dependent Li Diffusion in Li6PS5Cl Investigated by Machine-Learning Potential.

Solid-state electrolytes with argyrodite structures, such as Li6PS5Cl, have attracted considerable attention due to their superior safety compared to liquid electrolytes and higher ionic conductivity than other solid electrolytes. Although experimental efforts have been made to enhance conductivity by controlling the degree of disorder, the underlying diffusion mechanism is not yet fully understood. Moreover, existing theoretical analyses based on ab initio molecular dynamics (MD) simulations have limitations in addressing various types of disorder at room temperature. In this study, we directly investigate Li-ion diffusion in Li6PS5Cl at 300 K using large-scale, long-term MD simulations empowered by machine-learning potentials (MLPs). To ensure the convergence of conductivity values within an error range of 10%, we employ a 25 ns simulation using a 5 × 5 × 5 supercell containing 6500 atoms. The computed Li-ion conductivity, activation energies, and equilibrium site occupancies align well with experimental observations. Notably, Li-ion conductivity peaks when Cl ions occupy 25% of the 4c sites rather than at 50% where the disorder is maximized. In addition, Li-ion diffusion shows non-Arrhenius behavior, leading to different activation energies at high temperatures (>400 K). These phenomena are explained by the interplay between inter- and intracage jumps. By elucidation of the key factors affecting Li-ion diffusion in Li6PS5Cl, this work paves the way for optimizing ionic conductivity in the argyrodite family.

Evolutionary Probability and Stacked Regressions Enable Data-Driven Protein Engineering with Minimized Experimental Effort.

Protein engineering through directed evolution and (semi)rational approaches is routinely applied to optimize protein properties for a broad range of applications in industry and academia. The multitude of possible variants, combined with limited screening throughput, hampers efficient protein engineering. Data-driven strategies have emerged as a powerful tool to model the protein fitness landscape that can be explored in silico, significantly accelerating protein engineering campaigns. However, such methods require a certain amount of data, which often cannot be provided, to generate a reliable model of the fitness landscape. Here, we introduce MERGE, a method that combines direct coupling analysis (DCA) and machine learning (ML). MERGE enables data-driven protein engineering when only limited data are available for training, typically ranging from 50 to 500 labeled sequences. Our method demonstrates remarkable performance in predicting a protein's fitness value and rank based on its sequence across diverse proteins and properties. Notably, MERGE outperforms state-of-the-art methods when only small data sets are available for modeling, requiring fewer computational resources, and proving particularly promising for protein engineers who have access to limited amounts of data.

Josephson effect in a fractal geometry

The Josephson effect is a hallmark signature of the superconducting state, which, however, has been sparsely explored in non-crystalline superconducting materials. Motivated by this, we consider a Josephson junction consisting of two superconductors with a fractal metallic interlayer, which is patterned as a Sierpiński carpet by removing atomic sites in a self-similar and scale-invariant manner. We here show that the fractal geometry has direct observable consequences on the Josephson effect. In particular, we demonstrate that the form of the supercurrent–magnetic field relation as the fractal generation number increases can be directly related to the self-similar fractal geometry of the normal metallic layer. Furthermore, the maxima of the corresponding diffraction pattern directly encode the self-repeating fractal structure in the course of fractal generation, implying that the corresponding magnetic length directly probes the shortest length scale in the given fractal generation. Our results should motivate future experimental efforts to verify these predictions in designer quantum materials and motivate future pursuits regarding fractal-based SQUID devices.

Flow-by membraneless electrolyzer designs: A macroporous flow dividing mesh enhances maximum allowable electrode length

The membraneless electrolyzer design promises a low-cost and robust electrolyzer technology, eliminating the disadvantages associated with the membranes/diaphragms in conventional electrolyzers. Flow-by membraneless electrolyzers exploit the Segré–Silberberg effect, where the electrolyte flow between parallel face-to-face cathode and anode forbids the evolving hydrogen and oxygen bubbles to cross over to the other side, while still allowing ionic currents between the electrodes to pass. The removal of the membrane from traditional electrolyzers, and instead exploiting the electrolyte flow itself to function as a gas separator also imposes certain requirements, namely: 1) upward laminar flow and, 2) vertically aligned electrodes. Given the upper limit of the laminar flow regime (Reynolds number, Re ∼ 1800), the admissible length of both vertically aligned electrodes is constrained by the production volume of H2 and O2 at both electrodes. Beyond a certain production rate the evolving gas plume increases in thickness until it reaches the central line dividing the channel between the electrodes. From that point onwards, flow mediated separation of both gases becomes practically impossible. In this work the design constraints of membraneless electrolyzers are investigated by combined multiphysics modeling and mass-balance analysis. Next, a macroporous flow dividing mesh is introduced in the design that allows seamless ionic flow between the electrodes while facilitating a higher electrolyte velocity in the laminar regime. This in turn enables to increase the maximum electrode length (or height) by >50 %. The model based analysis provides important guidelines for further development of membraneless electrolyzers, significantly reducing future experimental optimization efforts.