The volume size factor (VSF) is a key descriptor associated with lattice distortion and related phenomena such as solid-solution strengthening. However, experimental VSF data remain scarce because of the difficulty and cost of preparing solid solutions, while first-principles calculations often exhibit systematic discrepancies from experiments in specific alloy systems. Here, we demonstrate that these discrepancies can be treated not as random errors but as learnable physical information. A transfer-learning framework was developed to refine first-principles VSF predictions using limited experimental data while preserving extrapolative reliability. First-principles calculations were performed for 1,998 binary solid-solution systems, producing one of the largest VSF datasets reported to date. Six transfer-learning strategies and six machine-learning models were systematically evaluated with respect to generalization and extrapolation performance. Conventional machine-learning approaches exhibited substantial degradation outside the experimental-data distribution, whereas transformation-based approaches maintained predictive reliability under extrapolation conditions. Pareto analysis identified optimal models balancing agreement with experiments and consistency with first-principles calculations, and their ensemble average was adopted to construct the final refined VSF dataset. Descriptor analysis revealed that discrepancies between first-principles and experimental VSFs are dominated by electronic-structure-related factors, particularly the period of the solute element, suggesting insufficient treatment of relativistic effects and strong electron correlation in conventional first-principles approaches. The proposed framework establishes a practical route for integrating first-principles calculations and experimental data toward reliable alloy design, and the refined dataset may serve as a useful resource for the design and exploration of advanced structural alloys, including high-entropy alloys, particularly for experimentally unexplored alloy systems.
This work presents a systematic experimental campaign characterising the elastic and strength properties of a novel class of hybrid lattice structures obtained by combining triply periodic minimal surfaces (TPMS), namely Schwarz Primitive (Scp) and Gyroid (Gyr), with polyhedron-based open-cell structures (POCS), namely Diamond (Dia) and Kelvin (Kel). Hybrid configurations were generated by varying the volume fraction of each constituent under both fully-filled (FULL) and partially hybridised (HALF) infill strategies, fabricated by fused filament fabrication using polylactic acid (PLA), and tested under uniaxial quasi-static compression to characterise the compressive modulus, yield strength, plateau stress, and specific energy absorption (SEA), with mechanical responses described through Gibson–Ashby power-law scaling relations. The effective compressive stiffness is primarily governed by the TPMS parent topology, with the POCS contribution becoming appreciable only at specific volume-fraction combinations. The Scp(uc48,vf5)+Dia(uc24,vf3)–FULL configuration approached the Hashin–Shtrikman upper bound at the corresponding relative density, while Schwarz Primitive–Kelvin hybrids exhibited elevated stiffness–density exponents. For energy absorption, Kelvin-based configurations yielded systematically higher SEA than their Diamond-based counterparts, with Gyr(uc48,vf1)+Kel(uc24,vf3)–FULL recording the highest quadratic SEA coefficient (A=138.0 J/g), while Kelvin-based Gyroid hybrids in HALF configurations achieved peak efficiencies of ηmax=91.54%. These findings demonstrate that the mechanical response of TPMS–POCS hybrids is governed by the topological interaction between parent geometries rather than by their individual properties alone, providing a quantitative basis for the design of architected materials requiring a compromise between stiffness, energy absorption, and deformation stability.
Fe41Co7Cr15Mo14C15B6Y2 amorphous alloy was fabricated by laser directed energy deposition (LDED) using laser powers of 550–800 W and scanning speeds of 1100–1500 mm/min to investigate the effect of thermal input on amorphous-phase retention and thermal stability. X-ray diffraction showed that all as-built samples were predominantly amorphous within the investigated processing window. Differential scanning calorimetry revealed a significant transition in crystallization behavior with increasing thermal input. At 550 W, the crystallization exotherms remained in the range of 600–700 °C, comparable to those of the feedstock powder. In contrast, at 700–800 W, the low-temperature exotherm was suppressed, and a dominant high-temperature exotherm appeared at approximately 780–820 °C, with its intensity increasing as the effective cooling rate decreased. Transmission electron microscopy combined with EDS mapping demonstrated that higher thermal input promoted elemental segregation and the formation of a limited number of nanoscale crystalline phases. This behavior is attributed to process-induced compositional modulation resulting from partial crystallization and solute redistribution during deposition. The modified amorphous matrix exhibited enhanced thermal stability, which effectively suppressed crystallization during subsequent thermal cycles. The results indicate that, within an appropriate processing window, a moderate reduction in the effective cooling rate can improve the thermal stability of additively manufactured Fe-based amorphous alloys, thereby promoting the retention of a higher amorphous fraction under cyclic thermal conditions.
This study examines the interplay between grain boundary and precipitation strengthening in Inconel 718 (IN718) by varying the solution heat treatment (SHT) temperature prior to standard double aging. Grain sizes ranging from 27 to 400 µm were produced by SHT at 968, 1066, 1200, and 1230°C, followed by aging at 720°C/8h and 620°C/8h. Microstructural characterization showed pronounced grain coarsening with increasing SHT temperature, consistent with the reduced δ-phase pinning effect at elevated temperatures. Precipitation strengthening appeared to be the major strengthening contribution under the investigated aged conditions, while grain boundary strengthening made a significant contribution in the finest-grained and aged condition. With increasing grain size, the grain boundary contribution decreased, whereas the γ″ contribution increased, which is consistent with increased Nb availability after partial δ-phase dissolution. At 1230°C, excessive grain growth was associated with a marked reduction in both strength and ductility. The calculated strengthening contributions were combined using the superposition method and compared with the experimentally measured yield strengths. The semi-quantitative model captured the experimental yield-strength trend, although the absolute values remain sensitive to fitted and simulation-derived input parameters. Overall, the results show that the highest strength is achieved when a fine grain structure is preserved together with effective γ′/γ″ precipitation during aging, rather than by applying the highest solution treatment temperature.
The use of iron as a recyclable metal fuel requires efficient, low-carbon reduction methods. This study examines the low-pressure non-thermal hydrogen plasma-assisted solid state reduction of iron oxide powders (45-63μm). Reduction experiments were carried out in the range of 200–500 °C and 0–30 W plasma power using a low-pressure capacitively-coupled radiofrequency (RF) plasma system with argon-hydrogen gas mixtures. The extent of reduction and phase composition were quantified by X-ray diffraction (XRD) and energy-dispersive X-ray spectroscopy (EDX), while surface morphology was characterized by scanning electron microscopy (SEM) and scanning transmission electron microscopy (STEM). Plasma-assisted reduction increased iron yield in comparison to thermal-only reduction experiments. Morphological analysis revealed bubble formation attributed to water vapor production, and plasma-induced surface smoothing consistent with an annealing effect. Subsequent re-oxidation tests performed in a thermogravimetric analyzer (TGA) indicate an increased kinetic barrier to oxidation, which may have implications for handling stability and high-temperature combustion. These results demonstrate that low-power non-thermal hydrogen plasma can enhance iron oxide reduction while modifying its surface properties, supporting its potential for sustainable iron fuel production.
Bi-Sb alloys are classic n-type cryogenic thermoelectric materials, yet research on p-type Bi-Sb alloys remains limited. Fabrication of thermoelectric devices therefore relies on other p-type counterparts for matching. Herein, we report a modification strategy for Bi-Sb alloys: Sn doping into the Bi0.88Sb0.12 matrix to endow the Bi-Sb alloy with p-type thermoelectric characteristics. Sn doping increases the hole concentration and optimizes electrical transport properties, yielding a peak thermoelectric figure of merit ZT of 0.10 at 200 K. Using the above material, we fabricated Bi-Sb-based thermoelectric modules by pairing Sn-doped p-type Bi0.88Sb0.12 with Te-doped n-type Bi0.88Sb0.12, where Nd2Fe14B permanent magnets are integrated into the n-type legs. Over a hot-side temperature range of 100–300 K, the device achieves a maximum temperature difference of 16.5 K at 250 K under zero magnetic field. When subjected to a 0.55 T magnetic field, this peak temperature difference rises to 19.3 K at 250 K. Sn doping enables p-type conduction in Bi-Sb alloys and alleviates structural mismatch when assembling devices with p-type materials of dissimilar matrices. This work offers a feasible new route for p-type Bi-Sb material synthesis as well as the construction and optimization of homojunction thermoelectric devices.
The thermal stability and phase evolution of a CoCrFeMnNi high-entropy alloy were investigated using free-standing amorphous films. The structural and chemical evolution of the alloy at different temperatures was investigated using in situ annealing in a transmission electron microscope. The amorphous structure was found to be remarkably stable up to 300°C. Above this temperature, the microstructure evolved into a dual-phase configuration consisting of FCC and BCC phases, which remained stable up to 550°C. The alloy remained chemically homogeneous throughout the transformation. At ∼700°C, concurrent formation of an L10-ordered phase and Cr-Mn-rich regions resulted in chemical inhomogeneity. However, the σ-phase, often observed in the bulk transformation of crystalline alloys, was not observed throughout the thermal treatment. These findings provide an amorphous-to-crystalline pathway to stabilize multiphase microstructures that are difficult to access via conventional/bulk processing routes.
Magnetostrictive smart composites offer a promising route for developing intelligent fastening components that can self-sense internal stress. In this study, we develop and evaluate magnetostrictive composite bolts that sense internal axial force via changes in magnetic flux density through the inverse magnetostrictive effect. The bolts are fabricated by embedding FeCo wires in epoxy resin and then applying tensile and torsional prestresses to the bolts during fabrication to enhance their magnetic response. Fastening tests are conducted at various tightening torques to record axial strain and magnetic flux density. The effect of an external bias magnetic field during fastening is also examined. Theoretical and finite element analyses are performed, and predictions are compared with measurements to identify sensitivity-enhancing design conditions. Experiments demonstrate a monotonic relation between the tightening torque, axial strain, and magnetic flux density change, indicating that axial force can be estimated from magnetic measurements. Sensitivity increases with the applied bias field, with the response being amplified further by the introduced prestress. Fabrication prestress with an appropriate operational bias may support axial force sensing in magnetostrictive composite bolts. The findings offer initial guidance for the in situ preload monitoring of bolted joints.
Ge-rich Ge-Sb-Te (GGST) alloys are of high interest for industrial production of complementary metal-oxide-semiconductor integrated phase change random access memories (PCRAM) of high performance. BEOL-integrated GGST-based PCRAM constitute a relevant solution for low-power on-chip non-volatile memory production and in-memory computing, addressing new challenges of automotive and artificial intelligence applications for example. During cycling, in the low-resistive SET state, the crystallized GGST alloy is mainly made of nano-grains of two phases: diamond Ge and a ternary rock-salt (RS) Ge-Sb-Te (GST) phase. The RS-GST phase is metastable and generally assumed to exhibit the stoichiometry Ge2Sb2Te5 corresponding to the stable hexagonal Ge2Sb2Te5 compound. However, recent works suggest that this metastable RS-GST phase is not stoichiometric and can accept higher Ge contents, meaning that the Ge content of RS-GST could vary depending on the Ge excess level in the GGST alloy compared to the Ge2Sb2Te5 reference stoichiometry. Consequently, the performance of GGST-based PCRAMs could vary with Ge composition of GGST alloy, suggesting the existence of an ideal GGST alloy composition showing a good compromise between crystallization temperature, programming current density, resistance, threshold voltage drift… In the present work, the structure and composition of crystallized GGST films elaborated through Ge, Sb, and Te co-sputtering were studied by X-ray diffraction, high resolution transmission electron microscopy, and atom probe tomography. The metastable RS-GST phase crystallized in GGST films exhibiting a Ge excess between 23% and 42% is found to be able to incorporate a large amount of Ge, of about 50 at% and sometimes beyond.
The concept of high-entropy materials (HEMs) features the integration of multiple elements into a single phase, unlocking vast compositional freedom and yielding new functionalities. While oxides, sulphides, and nitrides have been widely explored, high-entropy iodides (HEIs) remain vastly unknown despite the rich optoelectronic properties of iodide perovskites. Here, we introduce low-dimensional perovskite derivative high-entropy iodides based on the MA3Bi2I9 structure, synthesised via mechanochemical routes with systematic multi-cation substitution (Sn, Ag, Bi, Cu, Sb, Ca, Na, and In). X-ray diffraction confirms the formation of single-phase HEI, with phase purity depending strongly on elemental choice. The optical band gap is robust against compositional complexity while allowing the fine-tuning within 2.0–2.1 eV, indicating that electronic properties can be delicately adjusted without structural destabilisation. In contrast, mechanical properties vary almost by an order of magnitude, with Young’s moduli spanning from 1.5 to 9.6 GPa, establishing a direct link between cation chemistry and lattice rigidity. These results demonstrate that high-entropy iodides are synthetically accessible, structurally stable, and provide a new platform for the fine-tuning of optical properties. At the same time, mechanical responses are widely adjustable by design, altogether opening previously unexplored compositional space for halide perovskite derivatives in optoelectronic and energy applications.
Biodegradable Fe-based alloys have emerged as potential candidates for temporary orthopedic implants due to their favorable mechanical performance and their higher degradation rate compared to pure iron. However, their fabrication by laser powder bed fusion (LPBF) typically employs pre-alloyed spherical powders produced by gas atomization, which increases the cost of manufacture. In this work, a robust, versatile, and affordable powder preparation route is proposed for the LPBF fabrication of biodegradable austenitic FeMnC alloys using irregular elemental particles. The approach starts from mechanical mixing of Fe, Mn, and C powders followed by a thermal treatment and milling to improve the flowability and promote elemental diffusion. Five conditions were evaluated to investigate how laser power and scanning velocity influence the properties of the printed samples, such as microstructure, porosity, hardness and corrosion behavior. The printed Fe12Mn1.2C alloy exhibited a fully austenitic microstructure, with pore size and interconnection strongly dependent on the chosen values of laser power and scanning velocity. Microhardness values around 350 HV were obtained, which are higher than those typically reported for conventionally processed Hadfield-type steels. Electrochemical analyses demonstrated that samples with higher porosity exhibited an accelerated degradation rate (∼2 mmpy), governed mainly by charge-transfer processes and enhanced by diffusion-assisted pathways. Overall, this research demonstrates the viability of the proposed powder preparation method to use irregular elemental powders for LPBF processing and highlights the potential of this cost-effective route for the manufacture of biodegradable FeMnC alloys.
To develop cast HEAs for sliding components with balanced deformability and damage tolerance, AlxCoCrNi1.5 alloys (x = 0, 0.5, 0.75, 1.0) were fabricated by vacuum arc melting, and their microstructures and dry-sliding wear properties were characterized via multiple experiments combined with molecular dynamics simulations. Rising Al content triggers an FCC → FCC+BCC/B2 transition. The A0.75 alloy forms a hierarchical structure: 2–5 μm NiAl-rich BCC/B2 zones dispersed in the FCC matrix, plus nanoscale L12 precipitates inside FCC grains. This structure yields the lowest friction coefficient (0.25) and wear rate (1.9 × 10–4 mm3/(N·m)), with worn surfaces dominated by shallow grooves and fine debris, in contrast to severe delamination in A0 and brittle spalling in A1. Simulations verify Al-promoted Ni–Al ordering and suppressed localized deformation in A0.75. Moderate Al addition is thus confirmed effective to balance hardness and wear damage resistance.
To achieve synergistic optimization of strength and ductility in Ti-Cu alloys, this study investigates Ti-5Cu alloy. By introducing high strain energy storage through 60% cold rolling and performing recrystallization annealing within the 600–750 °C range, the effects of annealing temperature on phase composition, microstructural evolution, and mechanical properties are systematically examined. Results indicate that increasing annealing temperature significantly enhances recrystallization fraction while progressively reducing dislocation density and intra-grain orientation gradient. Concurrently, alloy strength and hardness gradually decrease, whereas plasticity markedly improves, reaching optimal values at 700 °C. Notably, A700 and A750 specimens exhibit a nonlinear stress-strain curve deflection during initial tensile loading, characterized by a decrease followed by a rebound in slope. Multiscale microscopic analysis and molecular dynamics simulations indicate that this behavior stems from a progressive yielding process induced by a heterogeneous structure in which recrystallized soft regions coexist with residual hard regions; dislocation accumulation at the interface and the development of back stress significantly enhance the work-hardening capacity. Accumulation of dislocations at interfaces and the establishment of back stresses significantly enhanced work-hardening capacity.
