Discovering Electrochemistry with an Electrochemistry-Informed Neural Network (ECINN).

Machine learning is increasingly integrated into chemistry research by guiding experimental procedures, correlating structure and function, interpreting large experimental datasets, to distill scientific insights that might be challenging with traditional methods. Such applications, however, largely focus on gaining insights via big data and/or big computation, while neglecting the valuable chemical prior knowledge dwelling in chemists' minds. In this paper, we introduce an Electrochemistry-Informed Neural Network (ECINN) by explicitly embedding electrochemistry priors including the Butler-Volmer (BV), Nernst and diffusion equations on the backbone of neural networks for multi-task discovery of electrochemistry parameters. We applied the ECINN to voltammetry experiments of F e 2 + / F e 3 + ${{\rm F}{{\rm e}}^{2+}/{\rm F}{{\rm e}}^{3+}}$ and R u N H 3 6 2 + / R u N H 3 6 3 + ${{\rm R}{\rm u}{\left({\rm N}{{\rm H}}_{3}\right)}_{6}^{2+{\rm \ }}/{\rm R}{\rm u}{\left({\rm N}{{\rm H}}_{3}\right)}_{6}^{3+{\rm \ }}}$ redox couples to discover electrode kinetics and mass transport parameters. Notably, ECINN seamlessly integrated mass transport with BV to analyze the entire voltammogram to infer transfer coefficients directly, so offering a new approach to Tafel analysis by outdating various mass transport correction methods. In addition, ECINN can help discover the nature of electron transfer and is shown to refute incorrect physics if imposed. This work encourages chemists to embed their domain knowledge into machine learning models to start a new paradigm of chemistry-informed machine learning for better accountability, interpretability, and generalization.

Metallurgical Preparation of Nb-Al and W-Al Intermetallic Compounds and Characterization of Their Microstructure and Phase Transformations by DTA Technique.

The possibilities of metallurgical preparation of 40Nb-60Al and 15W-85Al intermetallic compounds (in at.%) by plasma arc melting (PAM) and vacuum induction melting (VIM) were studied. Both methods allow easy preparation of Nb-Al alloys; however, significant evaporation of Al was observed during the melting, which affected the resulting chemical composition. The preparation of W-Al alloys was more problematic because there was no complete re-melting of W during PAM and VIM. However, the combination of PAM and VIM allowed the preparation of W-Al alloy without any non-melted parts. The microstructure of Nb-Al alloys consisted of Nb2Al and NbAl3 intermetallic phases, and W-Al alloys consisted mainly of needle-like WAl4 intermetallic phase and Al matrix. The effects of melting conditions on chemical composition, homogeneity, and microstructure were determined. Differential thermal analysis was used to determine melting and phase transformation temperatures of the prepared alloys.

Removing 65 Years of Approximation in Rotating Ring Disk Electrode Theory with Physics-Informed Neural Networks.

The rotating Ring Disk Electrode (RRDE), since its introduction in 1959 by Frumkin and Nekrasov, has become indispensable with diverse applications in electrochemistry, catalysis, and material science. The collection efficiency (N) is an important parameter extracted from the ring and disk currents of the RRDE, providing valuable information about reaction mechanism, kinetics, and pathways. The theoretical prediction of N is a challenging task: requiring solution of the complete convective diffusion mass transport equation with complex velocity profiles. Previous efforts, including by Albery and Bruckenstein who developed the most widely used analytical equations, heavily relied on approximations by removing radial diffusion and using approximate velocity profiles. 65 years after the introduction of RRDE, we employ a physics-informed neural network to solve the complete convective diffusion mass transport equation, to reveal the formerly neglected edge effects and velocity corrections on N, and to provide a guideline where conventional approximation is applicable.

Synthesis of FeSi-FeAl Composites from Separately Prepared FeSi and FeAl Alloys and Their Structure and Properties.

Composites consisting of iron aluminide and iron silicide phases were studied in this work. Powders of iron aluminide and iron silicide were prepared by mechanical alloying separately. Subsequently, they were blended in three different proportions and sintered by the SPS method under various conditions. After sintering, the composites are composed of FeAl and amounts of other silicides (Fe5Si3 and Fe3Si). Ternary Fe-Al-Si phases were not determined, even though their presence was predicted by DFT calculations. This disagreement was explained by steric factors, i.e., by differences in the space lattice of the present phases. Hardness and tribological properties were measured on composites with various weight ratios of iron aluminide and iron silicide. The results show that sintered silicides with the matrix composed of iron aluminide reach comparable hardness to tool steels. The composites with higher mass ratios of iron aluminide than silicide have higher hardness and better tribological properties.

Mechanical, Thermal, and Fire Properties of Composite Materials Based on Gypsum and PCM.

One of the solutions for overheating the interior in the summer without increasing energy consumption is the integration of phase change material (PCM) into interior plasters. However, adding PCM to plasters deteriorates their properties and thus their usability. The aim of this paper is to determine how the microencapsulated PCM affects the mechanical, thermal, and fire properties of plasters and how much PCM can be added to the plaster. Two sets of samples were prepared: in set S, part of the aggregate was replaced by PCM; and in set R, only PCM was added. The bulk density, flexural strength, compressive strength, tensile strength perpendicular to the surface, thermal conductivity coefficient, specific heat capacity, melting, and solidification temperatures and enthalpy were measured. A single-flame source fire test and a gross heat of combustion fire test were performed to determine the reaction to the fire class. The results show that with an increasing proportion of PCM, the strength of the samples of set R decreased more significantly than it did with the samples of set S. It was found that only up to about 10% PCM could be added to set R, while up to 30% PCM could be added to set S.

The Effect of Trace Oxygen Addition on the Interface Behavior of Low-Alloy Steel.

This work aims to assess the effect of an oxygen content graded in minimal quantities, on the order of hundreds of ppms, on the determination of surface tension of low-alloy FeCOCr and FeCONi steels in contact with a corundum substrate. Oxygen, as a surface-active element, was segregated at the surface where it interacted with the major components of the alloys, leading to a reduction in surface tension. The sessile drop method was used for wetting tests in the temperature range from steel liquidus temperatures to 1600 °C under nonoxidizing conditions. The effect of oxygen on surface tension and wetting angles was verified by statistical analysis using the Kruskal-Wallis test, which supported the results stating that the values of these quantities decreased with increasing oxygen content. Furthermore, liquidus temperatures, which are of practical importance, were determined by the optical and DTA methods and then compared with theoretically calculated temperature values. It turned out that the increased chromium content causes difficulties in determining surface tension up to 1550 °C due to the formation of a thin Cr2O3 layer. In addition, SEM and XRD analyses accompanied by calculations in the FactSage oxide database were performed to better understand the wetting mechanism.

Blending Powder Process for Recycling Sintered Nd-Fe-B Magnets.

The wide application of Nd-Fe-B permanent magnets, in addition to rare-earth metal resource constraints, creates the necessity of the development of efficient technologies for recycling sintered Nd-Fe-B permanent magnets. In the present study, a magnet-to-magnet recycling process is considered. As starting materials, magnets of different grades were used, which were processed by hydrogen decrepitation and blending the powder with NdHx. Composition inhomogeneity in the Nd2Fe14B-based magnetic phase grains in the recycled magnets and the existence of a core-shell structure consisting of a Nd-rich (Dy-depleted) core and Nd-depleted (Dy-enriched) shell are demonstrated. The formation of this structure results from the grain boundary diffusion process of Dy that occurs during the sintering of magnets prepared from a mixture of Dy-free (N42) and Dy-containing magnets. The increase in the coercive force of the N42 magnet was shown to be 52%. The simultaneous retention of the remanence, and even its increase, were observed and explained by the improved isolation of the main magnetic phase grains as well as their alignment.

Crystallization and Quantification of Crystalline and Non-Crystalline Phases in Kaolin-Based Cordierites.

Kaolin is most often used as traditional raw material in ceramic industry. The purpose of the study was to obtain understanding of the structural and chemical variability of cordierite ceramics influenced by chemical and mineralogical properties of six raw kaolins taken from different localities when they are applied in ceramics mixtures with vermiculite and sintered up to 1300 °C. The X-ray diffraction and simultaneous thermogravimetric and differential thermal analysis were used to identify and characterize crystalline mineral phases and the course of reactions during the heating. The percentages of the crystalline and non-crystalline phases were newly determined by recalculation of the bulk chemical analyses of kaolins and cordierite ceramics using Chemical Quantitative Mineral Analysis (CQMA) method. Varying amounts of minerals in kaolins: kaolinite from 73.3 to 85.0, muscovite from 4.2 to 9.9, and quartz from 6.0 to 19.5 (mass %) affected amount of cordierite/indialite from 75.2 to 85.1, enstatite from 5.8 to 8.9 (when are calculated as their maximal possible percentages), and non-crystalline phases from 8.8 to 15.1 (mass %) in cordierite ceramics. Regression analysis predicted high relationship between quantity of: (a) kaolinite in kaolins and crystalline cordierite and (b) quartz in kaolins and non-crystalline phases in the ceramics. The migration of potassium from muscovite into the cordierite structure, melting point and crystallization of cordierite/indialite phases and pore size variability in relation to impurity of kaolins are documented and discussed.