Materials最新文献

With the rise in additive manufacturing in construction, particularly 3D printing using extrusion-based mortars, there is an increasing need to optimize material properties. This study compares the mechanical performance of mortar specimens produced by traditional casting and 3D printing, with a focus on flexural behavior. A high-durability mortar with very low chloride and sulfate content, which produces less CO₂ than standard Portland cement, was used. This study also explores the impact of varying water-cement (w/c) ratios to obtain a valid mix for both fabrication methods. The results show that the samples obtained by traditional processes and those produced through 3D printing exhibit distinctly different behaviors under bending stresses. In the case of the molded samples, the maximum stress ranged from 1.23 to 1.78 MPa, indicating good strength and uniformity within these materials. In contrast, the 3D-printed samples showed higher values but with greater variation, ranging between 2.77 and 3.76 MPa. This variation highlights the influence of the fabrication technique in 3D printing, which may contribute to either the superiority or limitations of these samples. In terms of deformation, molded specimens exhibited brittle failure with limited post-peak energy dissipation (0.11-0.22 kN.mm), whereas 3D-printed samples displayed a mixed brittle-ductile response and enhanced energy absorption (1.70-2.82 kN.mm). These findings suggest that traditionally obtained specimens are suitable for applications requiring predictable stiffness, while 3D-printed mortars are advantageous for applications demanding greater flexibility and energy absorption.

This study presents the first application of Machine Learning (ML) models to optimise Powder Bed Fusion using Laser Beam (PBF-LB) process parameters for H13 steel fabricated on a 350 °C preheated building platform. A total of 189 cylindrical specimens were produced for training and testing machine learning (ML) models using variable process parameters: laser power (250-350 W), scanning speed (1050-1300 mm/s), and hatch spacing (65-90 μm). Eight ML models were investigated: 1. Support Vector Regression (SVR), 2. Kernel Ridge Regression (KRR), 3. Stochastic Gradient Descent Regressor, 4. Random Forest Regressor (RFR), 5. Extreme Gradient Boosting (XGBoost), 6. Extreme Gradient Boosting with limited depth (XGBoost LD), 7. Extra Trees Regressor (ETR) and 8. Light Gradient Boosting Machine (LightGBM). All models were trained using the Fast Library for Automated Machine Learning & Tuning (FLAML) framework to predict the relative density of the fabricated samples. Among these, the XGBoost model achieved the highest predictive accuracy, with a coefficient of determination R2=0.977, mean absolute percentage error MAPE = 0.002, and mean absolute error MAE = 0.017. Experimental validation was conducted on 27 newly fabricated samples using ML predicted process parameters. Relative densities exceeding 99.6% of the theoretical value (7.76 g/cm³) for all models except XGBoost LD and KRR. The lowest MAE = 0.004 and the smallest difference between the ML-predicted and PBF-LB validated density were obtained for samples made with LightGBM-predicted parameters. Those samples exhibited a hardness of 604 ± 13 HV0.5, which increased to approximately 630 HV0.5 after tempering at 550 °C. The LightGBM optimised parameters were further applied to fabricate a part of a forging die incorporating internal through-cooling channels, demonstrating the efficacy of machine learning-guided optimisation in achieving dense, defect-free H13 components suitable for industrial applications.

This study employed machine learning to investigate the mechanical behavior of one-part geopolymer (OPG)-stabilized soil subjected to acid erosion. Based on the unconfined compressive strength (UCS) data of acid-eroded OPG-stabilized soil, eight machine learning models, namely, Adaptive Boosting (AdaBoost), Decision Tree (DT), Extra Trees (ET), Gradient Boosting (GB), Light Gradient Boosting Machine (LightGBM), Random Forest (RF), Support Vector Machine (SVM), and eXtreme Gradient Boosting (XGBoost), along with hyper-parameter optimization by Genetic Algorithm (GA), were used to predict the degradation of the UCS of OPG-stabilized soils under different durations of acid erosion. The results showed that GA-SVM (R² = 0.9960, MAE = 0.0289) and GA-XGBoost (R² = 0.9961, MAE = 0.0282) achieved the highest prediction accuracy. SHAP analysis further revealed that solution pH was the dominant factor influencing UCS, followed by the FA/GGBFS ratio, acid-erosion duration, and finally, acid type. The 2D PDP combined with SEM images showed that the microstructure of samples eroded by HNO₃ was marginally denser than that of samples eroded by H₂SO₄, yielding a slightly higher UCS. At an FA/GGBFS ratio of 0.25, abundant silica and hydration products formed a dense matrix and markedly improved acid resistance. Further increases in FA content reduced hydration products and caused a sharp drop in UCS. Extending the erosion period from 0 to 120 days and decreasing the pH from 4 to 2 enlarged the pore network and diminished hydration products, resulting in the greatest UCS reduction. The results of the study provide a new idea for applying the ML model in geoengineering to predict the UCS performance of geopolymer-stabilized soils under acidic erosion.

Low-density thermal insulation materials tend to settle during operation or under small loads. Resistance to loads and settling is ensured by increasing the density of thermal insulation materials several times. This increases the weight of the material and the structure and production costs. In this work, using various technological processes, corrugated textile sheets and thermal insulation materials were produced from textile fabric. The development of such materials as effective thermal insulation materials for building insulation has not yet been studied. The corrugation of textile sheets enabled the thermal insulation material to exhibit good mechanical and deformation properties without increasing its density or thermal conductivity. The density of the specimens of the thermal insulation material made from corrugated sheets ranged from 76.8 to 51.9 kg/m³, and the thermal conductivity ranged from 0.0535 to 0.0385 W/(m·K). The most significant influences on density and thermal conductivity were the wave size and the adhesive layer. The density of unglued sheets of the same composition ranged from 51.3 to 29.8 kg/m³, and the thermal conductivity ranged from 0.0530 to 0.0371 W/(m·K). The highest compressive and bending strengths were observed in thermal insulation materials prepared from finely corrugated sheets. Their compressive stress at 10% deformation was 17.3 kPa, and their bending strength was -157 kPa. In comparison, the compressive stress of thermal insulation materials of the same density as non-corrugated sheets was only 0.686 kPa and, in the case of bending strength, 9.90 kPa. The obtained results show that the application of materials engineering principles allows us to improve the performance characteristics of materials.

Al-Al₄C₃ composites exhibit promising mechanical properties including high specific strength, high specific stiffness. However, high reinforcement contents often promote brittle behavior, making it necessary to understand the mechanisms governing their limited toughness. In this work, a microstructural and mechanical study was carried out to evaluate the energy storage capacity in Al-Al₄C₃ composites fabricated by mechanical milling followed by heat treatment using X-ray diffraction (XRD) and Convolutional Multiple Whole Profile (CMWP) fitting method, the microstructural parameters governing the initial stored energy after fabrication were determined: dislocation density (ρ), dislocation character (q), and effective outer cut-off radius (R_e). Compression tests were carried out to quantify the elastic energy stored during loading (Es). The energy absorption efficiency (EAE) in the elastic region of the stress-strain curve was evaluated with respect to the elastic energy density per unit volume stored (Ee), obtained from microstructural parameters (ρ, q, and Re) present in the samples after fabrication and determined by XRD. A predictive model is proposed that expresses Es as a function of Ee and q, where the parameter q is critical for achieving quantitative agreement between both energy states. In general, samples with high EAE exhibited microstructures dominated by screw-character dislocations. High-resolution transmission electron microscopy (HRTEM) analyses revealed graphite regions near Al₄C₃ nanorods-formed during prolonged sintering-which, together with the thermal mismatch between Al and graphite during cooling, promote the formation of screw dislocations, their dissociation into extended partials, and the development of stacking faults. These mechanisms enhance the redistribution of stored energy and contribute to improved toughness of the composite.

Cement-based materials are central to modern infrastructure construction [...].

Inter-pass forging with different degrees of deformation during WAAM of Inconel 718 specimens (single-stage, three passes; two-stage, six passes) was investigated. Macrostructural analysis of the specimens showed that inter-pass forging led to a recrystallized structure. Alternation of layers with different grain shapes (columnar and equiaxed) is observed throughout the height of the specimens. Increasing the number of passes improves the mechanical properties of the material (tensile strength, yield strength, microhardness). A finite element model of inter-pass forging was developed to determine the effect of inter-pass surface deformation during WAAM on the residual stress-strain state. The non-stationary formulation was replaced with a quasi-static one. Johnson-Cook material constants were obtained for the deposited Inconel 718 material, including the effect of forging. Verification of the mathematical model was performed using a wall (specimen 2) deposited with single-stage forging. The deviation between the simulation results and the experiment did not exceed 15%. It was found that the sequence and number of passes significantly affect residual strain and displacements but have little effect on residual stress. Numerical modeling showed that the depth of plastic deformation exceeds the melting depth when depositing the subsequent layer, ensuring the preservation and accumulation of the inter-pass forging effect throughout the deposition process.

Mesoporous carbons based on silica hard templates were used to investigate physical processes in confined pores. Nitrogen adsorption, scanning electron microscopy, and scattered X-ray analyses revealed two classes of materials: carbons with moderate and highly developed mesoporosity. The pore structure was strongly dependent on pore expanders which proved essential for generating open, accessible architectures. All carbons exhibited a basic, graphitic surface (pH_PZC = 8.4-10.9), enriched in electron-donating oxygen functionalities. Differential scanning calorimetry studies of confined water showed that melting point depression follows the Gibbs-Thomson relationship, confirming the strong dependence of phase transitions on pore size and water-surface interactions. Adsorption experiments using methylene blue demonstrated that capacity is governed by surface area, pore volume, and pore size distribution. For carbon with the largest average pore size, adsorption of various dyes revealed that uptake decreases with increasing molecular size, whereas affinity depends strongly on electrostatic interactions. Kinetic studies indicated that carbons with larger mesopores exhibit the fastest adsorption, and that large, complex dye molecules undergo significant diffusion limitations. Overall, the results show that the interplay between pore structure, adsorbate size, and surface chemistry influences both the equilibrium uptake and adsorption kinetics in mesoporous carbon materials.

The CoCrMo alloys are progressively utilized as biomaterials. This research is dedicated to studying the consequence of (1, 3, and 5) wt% nano-TiO₂ addition on the porosity, microstructure, microhardness, and wear behavior of pre-alloyed CoCrMo powder produced by powder metallurgy (PM). Microstructural features were examined using SEM, SEM mapping, and XRD. Wear behavior was assessed through pin-on-disk tests performed under dry sliding conditions at varying loads and durations. Porosity increased with the addition of nano-TiO₂, from 15.26 at 0 wt% reaching 25.12% at 5 wt%, while density decreased from 7.16 to 6.33 g/cm³. Microhardness exhibited a slight improvement, attaining 348 HV at 5 wt%. SEM and XRD analyses confirmed partial particle separation after sintering and identified the TiO₂ reinforcement as rutile. Wear tests revealed that adding 1 wt% nano-TiO₂ enhanced wear resistance, whereas extended sliding durations resulted in increased wear rates. Adhesive wear was the predominant mechanism, accompanied by limited abrasive wear, oxidation, and plastic deformation.

Accurately predicting the ultimate strain of fiber-reinforced polymer (FRP)-confined concrete columns is essential for the widespread application of FRP in strengthening reinforced concrete (RC) columns. This study comprehensively investigates the performance of ensemble machine learning (ML) models in estimating the ultimate strain of FRP-confined concrete (FRP-CC) columns. A dataset of 547 test results of the ultimate strain of FRP-CC columns was collected from the literature for training and testing ML models. The four best single ML models were used to develop ensemble models employing voting, stacking and bagging techniques. The performance of the ensemble models was compared with 10 single ML and 11 empirical strain models. The study results revealed that the single ML models yielded good agreement between the estimated ultimate strain and the test results, with the best single ML models being the K-Star, k-Nearest Neighbor (k-NN) and Decision Table (DT) models. The three best ensemble models, a stacking-based ensemble model comprising K-Star, k-NN and DT models; a stacking-based ensemble model comprising K-Star and k-NN models and a voting-based ensemble model comprising K-Star and k-NN models, achieved higher estimation accuracy than the best single ML model in estimating the strain capacity of FRP-CC columns.