Search

In this research, the adsorption performance of individual atoms on the surface of monolayer graphene surface was systematically investigated using machine learning methods to accelerate density functional theory. The adsorption behaviors of over thirty different atoms on the graphene surface were computationally analyzed. The adsorption energy and distance were extracted as the research targets, and the basic information of atoms (such as atomic radius, ionic radius, etc.) were used as the feature values to establish the dataset. Through feature engineering selection, the corresponding input feature values for the input-output relationship were determined. By comparing different models on the dataset using five-fold cross-validation, the mathematical model that best fits the dataset was identified. The optimal model was further fine-tuned by adjusting of the best mathematical ML model. Subsequently, we verified the accuracy of the established machine learning model. Finally, the precision of the machine learning model forecasts was verified by the method of comparing and contrasting machine learning results with density functional theory. The results suggest that elements such as Zr, Ti, Sc, and Si possess some potential in controlling the interfacial reaction of graphene/aluminum composites. By using machine learning to accelerate first-principles calculations, we have further expanded our choice of research methods and accelerated the pace of studying element–graphene interactions.

In order to provide guidance for furthering the balance of strength and toughness of AerMet 100 steel through tempering treatment, the effects of the tempering time on microstructure and mechanical properties are investigated. The microstructure evolution, especially M2C precipitates and austenite in AerMet 100 tempered at 482 °C for 1~20 h, was characterized, and its influences on the mechanical properties were studied. The tensile strength decreases gradually, the yield strength increases first and then decreases, and the fracture toughness KIC increases gradually with an increasing tempering time. The strength and toughness matching of AerMet 100 steel is achieved by tempering at 482 °C for 5~7 h. Without considering the martensitic size effect, the influence of the dislocation density on the tensile strength is more significant during tempering at 482 °C. The precipitation strengthening mechanism plays a dominant role in the yield strength when tempering for 5 h or less, and the combined influence of carbide coarsening and a sharp decrease in the dislocation density resulted in a significant decrease in tensile strength when tempering for 8 h or more. The fracture toughness KIC is primarily influenced by the reverted austenite, so that KIC increases gradually with the prolongation of the tempering time. However, a significant decrease in the dislocation density resulting from long-term tempering has a certain impact on KIC, giving rise to a decrease in the rising amplitude in KIC after tempering for 8 h or more.

In this paper, we studied the effects of a series of alloying atoms on the stability and micromechanical properties of aluminum alloy using a machine learning accelerated first-principles approach. In our preliminary work, high-throughput first-principles calculations were explored and the solution energy and theoretical stress of atomically doped aluminum substrates were extracted as basic data. By comparing five different algorithms, we found that the Catboost model had the lowest RMSE (0.24) and lowest MAPE (6.34), and this was used as the final prediction model to predict the solid solution strengthening of the aluminum matrix by the elements. Calculations show that alloying atoms such as K, Na, Y and Tl are difficult to dissolve in the aluminum matrix, whereas alloy atoms like Sc, Cu, B, Zr, Ni, Ti, Nb, V, Cr, Mn, Mo, and W exerted a strengthening influence. Theoretical studies on solid solutions and the strengthening effect of various alloy atoms in an aluminum matrix can offer theoretical guidance for the subsequent selection of suitable alloy elements. The theoretical investigation of alloy atoms in an aluminum matrix unveils the fundamental aspects of the solution strengthening effect, contributing significantly to the expedited development of new aluminum alloys.

Addressing the exceptional properties of aluminium alloy composites reinforced with graphene, this study presents an interpretable machine learning approach to aid in the rapid and efficient design of such materials. Initially, data on these composites were gathered and optimised in order to create a dataset of composition/process-property. Several machine learning algorithms were used to train various models. The SHAP method was used to interpret and select the best performing model, which happened to be the CatBoost model. The model achieved accurate predictions of hardness and tensile strength, with coefficients of determination of 0.9597 and 0.9882, respectively, and average relative errors of 6.02% and 5.01%, respectively. The results obtained from the SHAP method unveiled the correlation between the composition, process and properties of aluminium alloy composites reinforced with graphene. By comparing the predicted and experimental data in this study, all machine learning models exhibited prediction errors within 10%, confirming their ability to generalise. This study offers valuable insights and support for designing high-performance aluminium matrix composites reinforced with graphene and showcases the implementation of machine learning in materials science.

The FeNiCrAlCoCuTi alloy system has great advantages in mechanical properties such as high hardness and toughness. It has high performance potential and research value and the key in research is designing alloy compositions with target properties. The traditional method, experimental analysis, is highly inefficient to properly exploit the intrinsic relationship between material characteristics and properties for multi-component alloys, especially in investigating the whole composition space. In this work, we present a research way that uses first principles calculation to obtain the properties of multi-component alloys and uses machine learning to accelerate the research. The FeNiCrAlCoCuTi alloy system with its elastic properties is used as an example to demonstrate this process. We specifically design models for each output, all of which have RMSE values of less than 1.1, and confirm their effectiveness through experimental data in the literature, showing that the relative error is below 5%. Additionally, we perform an interpretable analysis on the models, exposing the underlying relationship between input features and output. By means of spatial transformation, we achieve the prediction of the full-component spatial performance from binary to multiple components. Taking the FeNiCrAlM (M = Co, Cu, Ti) quinary alloy system as an example, we design a single-phase BCC structure composed of an Fe0.23Cr0.23Al0.23Ni0.03Cu0.28 alloy with a Young’s modulus of 273.10 GPa, as well as a single-phase BCC structure composed of an Fe0.01Cr0.01Al0.01Ni0.44Co0.53 alloy with a shear modulus of 103.6 GPa. Through this research way, we use machine learning to accelerate the calculation, which greatly shortens research time and costs. This work overcomes the drawbacks of traditional experiments and directly obtains element compositions and composition intervals with excellent performance.

In this paper, the interfacial behavior and the atom diffusion behavior of an Al4Si alloy were systematically investigated by means of first-principles calculations. The K-points and cutoff energy of the computational system were determined by convergence tests, and the surface energies for five different surfaces of Al4Si alloys were investigated. Among the five surfaces investigated for Al4Si, it was found that the (111) surface was the surface with the lowest surface energy. Subsequently, we investigated the interfacial stability of the (111) surface and found that there were two types of interfaces, the Al/Al interface and the Al/Si interface. The fracture energies and theoretical strengths of the two interfaces were calculated; the results show that the Al/Al interface had the highest interfacial strength, and the calculation of their electronic results explained the above phenomenon. Subsequently, we investigated the diffusion and migration behavior of Si atoms in the alloy system, mainly in the form of vacancies. We considered the diffusion of Si atoms in vacancies of Al and Si atoms, respectively; the results showed that Si atoms are more susceptible to diffusive migration to Al atomic vacancies than to Si atomic vacancies. The results of the calculations on the micromechanics of aluminum alloys, as well as the diffusion migration behavior, provide a theoretical basis for the further development of new aluminum alloys.

High-entropy alloys have gained widespread concern in response to the increased requirements for future high-temperature structural superalloys. By combining phase-diagram calculations with microhardness, compression behavior measurements at room temperature, and elevated temperature conditions, the very important role of the Cr element on the microstructure and properties is deeply revealed, which provides candidates materials for future high-temperature alloy applications. The increment of Cr favors the regulation of the two-phase fraction and distribution. The thermodynamic calculations illustrate that the density and melting point of the HEAs showed an increasing trend with the increase of the Cr content. The typical worm-like microstructure of the Cr0.6 alloy with a dual BCC structure was detected. Meanwhile, on the one hand, the increment of the Cr elements results in a considerable optimization of the mechanical properties of the alloy in terms of strength and ductility at room temperature. The corresponding compressive strength and plasticity of Cr0.6 alloy at room temperature are 3524 MPa and 43.3%. On the other hand, the high-temperature mechanical properties of the alloy are greatly enhanced. At 1000 °C, the yield strength of the Cr0.6 alloy is about 25 MPa higher than that of the Cr0.4 alloy. The superior mechanical properties are attributed to the pronounced work-hardening response, and the work-hardening behavior of Cr-containing HEAs was systematically analyzed by employing the modified Ludwik model. The higher content of Cr helps the resistance of the local deformation response, improving the nonuniform strain and promoting the balance of strength and ductility of the alloys.

Graphene has attracted significant interest due to its unique properties. Herein, we built an adsorption structure selection workflow based on a density functional theory (DFT) calculation and machine learning to provide a guide for the interfacial properties of graphene. There are two main parts in our workflow. One main part is a DFT calculation routine to generate a dataset automatically. This part includes adatom random selection, modeling adsorption structures automatically, and a calculation of adsorption properties. It provides the dataset for the second main part in our workflow, which is a machine learning model. The inputs are atomic characteristics selected by feature engineering, and the network features are optimized by a genetic algorithm. The mean percentage error of our model was below 35%. Our routine is a general DFT calculation accelerating routine, which could be applied to many other problems. An attempt on graphene/magnesium composites design was carried out. Our predicting results match well with the interfacial properties calculated by DFT. This indicated that our routine presents an option for quick-design graphene-reinforced metal matrix composites.

Graphene has become an ideal reinforcement for reinforced metal matrix composites due to its excellent mechanical properties. However, the theory of graphene reinforcement in graphene/aluminum matrix composites is not yet well developed. In this paper, the effect of different temperatures on the mechanical properties of the metal matrix is investigated using a classical molecular dynamics approach, and the effects of the configuration and distribution of graphene in the metal matrix on the mechanical properties of the composites are also described in detail. It is shown that in the case of a monolayer graphene-reinforced aluminum matrix, the simulated stretching process does not break the graphene as the strain increases, but rather, the graphene and the aluminum matrix have a shearing behavior, and thus, the graphene “pulls out" from the aluminum matrix. In the parallel stretching direction, the tensile stress tends to increase with the increase of the graphene area ratio. In the vertical stretching direction, the tensile stress tends to decrease as the percentage of graphene area increases. In the parallel stretching direction, the tensile stress of the system tends to decrease as the angle between graphene and the stretching direction increases. It is important to investigate the effect of a different graphene distribution in the aluminum matrix on the mechanical properties of the composites for the design of high-strength graphene/metal matrix composites.

Precipitation hardening stainless steels have attracted extensive interest due to their distinguished mechanical properties. However, it is necessary to further uncover the internal quantitative relationship from the traditional standpoint based on the statistical perspective. In this review, we summarize the latest research progress on the relationships among the composition, microstructure, and properties of precipitation hardened stainless steels. First, the influence of general chemical composition and its fluctuation on the microstructure and properties of PHSS are elaborated. Then, the microstructure and properties under a typical heat treatment regime are discussed, including the precipitation of B2-NiAl particles, Cu-rich clusters, Ni3Ti precipitates, and other co-existing precipitates in PHSS and the hierarchical microstructural features are presented. Next, the microstructure and properties after the selective laser melting fabricating process which act as an emerging technology compared to conventional manufacturing techniques are also enlightened. Thereafter, the development of multi-scale simulation and machine learning (ML) in material design is illustrated with typical examples and the great concerns in PHSS research are presented, with a focus on the precipitation techniques, effect of composition, and microstructure. Finally, promising directions for future precipitation hardening stainless steel development combined with multi-scale simulation and ML methods are prospected, offering extensive insight into the innovation of novel precipitation hardening stainless steels.