Ability Of Bulk Metallic Glasses Research Articles

Identifying key features for predicting glass-forming ability of bulk metallic glasses via interpretable machine learning

Identifying key features for predicting glass-forming ability of bulk metallic glasses via interpretable machine learning

Relating the combinatorial materials chip mapping to the glass-forming ability of bulk metallic glasses via diffraction peak width

There is a gap between the cooling rate of thin film deposition (∼108 K/s) and bulk metallic glass (BMG) casting (∼103 K/s). In this work, we tried to corelate the glass-forming range (GFR) determined from the combinatorial materials chip (CMC) with the glass-forming ability (GFA) of BMG. Data from five CMCs and the reported critical casting diameter (Dmax) of BMGs were used as the dataset. It is found that the full-width at half maximum (FWHM) of the first sharp diffraction peak (FSDP) is a good indicator of the GFA of BMG as suggested by Liu et al. [1]. Strong positive correlations between the FWHM and Dmax are shown in the Zr-, Ag-, Cu-, Fe-, La-, and Mg-based BMG systems. Furthermore, the Pearson correlation coefficients of the symbolic regression (SR) models indicate possible similarity in the GFA natures of the Ag-Zr, Cu-Zr, Cu-Ag, and Mg-La pairs.

Glass-Forming Ability of Bulk Metallic Glass

Glass-Forming Ability of Bulk Metallic Glass

Overcoming the challenge of the data imbalance for prediction of the glass forming ability in bulk metallic glasses

Machine learning (ML) has been extensively studied in predicting the glass-forming ability of bulk metallic glasses (BMGs). Based on the current state of development of BMGs, the reported critical casting diameter (Dmax) data show an imbalance. Nevertheless, almost most of the current literature using ML to predicted Dmax has failed to consider this phenomenon, resulting in generally low prediction accuracy. Only a very small amount of literature deals with this issue only at the data level. In this work, an improved deep neural network (IDNN) model based on the DenseLoss function was proposed at the algorithmic level. The IDNN model has better generalization capability than the currently reported models by obtaining the highest R2 score of 0.841 in the test set. Our work highlights the importance of dealing with the Dmax imbalance problem to address the low accuracy of current ML models in predicting Dmax. This research work will have a significant impact on the discovery of novel BMGs.

Data-driven glass-forming ability criterion for bulk amorphous metals with data augmentation

• A data augmentation technique is employed for regression problems in the current work, and the machine learning model for predicting glass-forming ability of bulk metallic glasses are developed on the augmented training dataset. The data augmentation technique enhances the predictive accuracy greatly. • A new feature T g T x ( T l − T g ) ( T l − T x ) is proposed based on domain knowledge, and this feature has significant importance in predicting glass-forming ability of bulk metallic glasses. • The novel ML model gives the glass forming ability rules: the average atomic radius ranging from 140 pm to 165 pm, the value of T g T x ( T l − T g ) ( T l − T x ) higher than 2.5, the entropy of mixing higher than 10 J/K/mol, and the enthalpy of mixing ranging from −32 kJ/mol to −26 kJ/mol. A data augmentation technique is employed in the current work on a training dataset of 610 bulk metallic glasses (BMGs), which are randomly selected from 762 collected data. An ensemble machine learning (ML) model is developed on augmented training dataset and tested by the rest 152 data. The result shows that ML model has the ability to predict the maximal diameter D max of BMGs more accurate than all reported ML models. In addition, the novel ML model gives the glass forming ability (GFA) rules: average atomic radius ranging from 140 pm to 165 pm, the value of T g T x ( T l − T g ) ( T l − T x ) being higher than 2.5, the entropy of mixing being higher than 10 J/K/mol, and the enthalpy of mixing ranging from -32 kJ/mol to -26 kJ/mol. ML model is interpretative, thereby deepening the understanding of GFA.

Prediction of glass forming ability of bulk metallic glasses based on convolutional neural network

Bulk metallic glasses (BMGs) have been widely used in different fields owing to their unique and excellent properties. In order to accelerate the development of BMGs, different feasible parameters or criteria of their glass forming ability (GFA) have been proposed. With the advent of the era of big data, machine learning (ML) methods provide novel insights into the study of BMGs. In this paper, we trained a convolutional neural network (CNN) model to investigate GFA of BMGs. A hundred alloying elements and their possible combinations were taken into account by mapping an alloying composition into a 10 × 10 feature graph. Compared with the other prediction methods of GFA in BMGs, the alloy composition is the only variable input without the requirement for various physical and chemical properties obtained through pre-experiments. The predictive ability of our proposed model is quantified by a training R2 of 0.9745 and a test R2 of 0.8137. This work suggests that ML approaches has great potential in guiding the design of new BMG materials.

Thermodynamically-guided machine learning modelling for predicting the glass-forming ability of bulk metallic glasses

Glass-forming ability (GFA) of bulk metallic glasses (BMGs) is a determinant parameter which has been significantly studied. GFA improvements could be achieved through trial-and-error experiments, as a tedious work, or by using developed predicting tools. Machine-Learning (ML) has been used as a promising method to predict the properties of BMGs by removing the barriers in the way of its alloy design. This article aims to develop a ML-based method for predicting the maximum critical diameter (Dmax) of BMGs as a factor of their glass-forming ability. The main result is that the random forest method can be used as a sustainable model (R2 = 92%) for predicting glass-forming ability. Also, adding characteristic temperatures to the model will increase the accuracy and efficiency of the developed model. Comparing the measured and predicted values of Dmax for a set of newly developed BMGs indicated that the model is reliable and can be truly used for predicting the GFA of BMGs.

Prediction of Glass Forming Ability of Bulk Metallic Glasses Using Machine Learning

Prediction of Glass Forming Ability of Bulk Metallic Glasses Using Machine Learning

Determination of glass forming ability of bulk metallic glasses based on machine learning

Nowadays, the development of new bulk metallic glasses (BMGs) is still subject to repeated testing. To address this challenging problem, this paper proposes the random forest (RF) regression model for predicting glass forming ability (GFA) based on the 810 datapoints of the metal alloy composition dataset (including Tg, Tx, Tl, and Dmax, where Tg is the glass transition temperature, Tx the onset crystallization temperature, Tl the liquidus temperature or the offset temperature of melting, Dmax is the critical diameter). Various types of feature parameters related to GFA were first screened to identify the optimal features. Grid search was then used to optimize hyperparameters of the machine-learning (ML). The research suggests that the random forest (RF) regression model's accuracy has been improved, and our proposed approach has great potential in predicting the formation of the BMGs. Furthermore, this study also suggests that both characteristic temperature (Tg, Tx, and Tl) and topological structure parameters play an important role in describing the glass formation of alloys.

Machine learning prediction of glass-forming ability in bulk metallic glasses

The critical casting diameter (Dmax) quantitatively represents glass-forming ability (GFA) of bulk metallic glasses (BMGs). The present work constructed a dataset of two subsets, L-GFA subset of 376 BMGs with 1 mm ≤Dmax < 5 mm and G-GFA subset of 319 BMGs with Dmax ≥ 5 mm. The sequential backward selector and exhaustive feature selector are introduced to select key features. The trained XGBoost classifier with four selected features is able to successfully classify the L-GFA and G-GFA BMGs. Furthermore, the trained XGBoost regression model with another four selected features predicts the Dmax of G-GFA samples with a cross-validated correlation coefficient of 0.8012. The correlation between features and Dmax will provide the guidance in the design and discovery of novel BMGs.

On Glass Forming Ability of Bulk Metallic Glasses by Relating the Internal Friction Peak Value

The internal friction (IF) behaviors of a series of LaCe-, Zr-, and La-based bulk metallic glasses (BMGs) were studied by a computer-controlled, conventional inverted torsion pendulum. The results indicate that with an increasing temperature, the IF also increases gradually in the supercooled liquid region, followed by a decrease caused by crystallization. BMGs with a good glass forming ability (GFA) usually possess a high IF peak value for an alloy system with the same constituent elements. Furthermore, the magnitude of the IF value (Qi−1) of the inflection point is an efficient criterion of GFA. The Qi−1 value is a valid criterion under the conditions of identical constituent elements and different element contents. However, Qi−1 and GFA have no relationship among different alloy systems.

Splitting up entropy into vibrational and configurational contributions in bulk metallic glasses: A thermodynamic approach

We applied an efficient methodology to separate vibrational and configurational entropies in bulk metallic glasses by means of molecular dynamics simulation based on a combination of non-equilibrium adiabatic switching and reversible scaling methods. This approach involves calculating the vibrational free energy using the Einstein crystal as a reference for the solid phase and the recently proposed Uhlenbeck-Ford model for the fluid phase. This methodology has the advantage that it does not require a crystalline solid phase for separating the entropies. Therefore, in principle, it is applicable to any material, regardless of whether or not it has a crystalline phase. Using this methodology, we separate the vibrational and configurational entropies of two metallic glasses with different fragilities at zero external pressure, namely, Cu$_{50}$Zr$_{50}$ and Cu$_{46}$Zr$_{46}$Al$_{8}$. We find that the results for the former alloy are in quite reasonable agreement with recent experimental work by Smith \textit{et al.}[H. L. Smith \textit{et al.}, Nat. Phys. \textbf{13}, 900 (2017)]. We also find the configurational entropy of the glass containing Al to be 70\% larger than that of the other glass. Our results suggest that, although other factors may be at play, the configurational entropy can be used to investigate the effect of the addition of a minor-alloying element on the glass-forming ability of bulk metallic glasses.

Machine learning prediction of elastic properties and glass-forming ability of bulk metallic glasses

Abstract

The role of open spaces to glass-forming ability in bulk metallic glasses

Element-specific positron annihilation spectroscopy for Fe-B-Y-Nb bulk metallic glasses (BMGs) successfully sheds light on the role of interstitial open spaces in the amorphous matrix, as e.g., geometrically ideal open space, referred to as a Bernal hole, and large-scale nanovoids to glass-forming ability (GFA), yielding the following rules. Applying main constituent atoms in smaller sizes such as Fe, the Bernal holes get small without enhancing GFA since other elements cannot diffuse therein. On the contrary, main constituent atoms in larger sizes such as Ce form the large Bernal holes, which are feasible to accept other elements thus being responsible for GFA. The present findings is of pivotal importance in designing excellent BMGs with high GFA from the viewpoint of elemental combination of materials.

块体非晶合金的成分设计研究进展

高玻璃形成能力的非晶成分设计是非晶合金领域的研究难点，目前还没有一个可定量化、普适性的方法来确定潜在的块体非晶合金新成分。本文综述了多种准则用于预测具有高非晶形成能力的合金成分。 Amorphous component design of the glass forming ability of bulk metallic glasses (BMGs) is the key problem in the field of the amorphous alloy. There is no method that can be quantified and is universal to determine the potential new composition of bulk metallic glasses. In this paper, we reviewed a variety of criteria used to predict the alloy composition which has the high amorphous formation ability.

Beyond packing of hard spheres: The effects of core softness, non-additivity, intermediate-range repulsion, and many-body interactions on the glass-forming ability of bulk metallic glasses.

When a liquid is cooled well below its melting temperature at a rate that exceeds the critical cooling rate Rc, the crystalline state is bypassed and a metastable, amorphous glassy state forms instead. Rc (or the corresponding critical casting thickness dc) characterizes the glass-forming ability (GFA) of each material. While silica is an excellent glass-former with small Rc < 10(-2) K/s, pure metals and most alloys are typically poor glass-formers with large Rc > 10(10) K/s. Only in the past thirty years have bulk metallic glasses (BMGs) been identified with Rc approaching that for silica. Recent simulations have shown that simple, hard-sphere models are able to identify the atomic size ratio and number fraction regime where BMGs exist with critical cooling rates more than 13 orders of magnitude smaller than those for pure metals. However, there are a number of other features of interatomic potentials beyond hard-core interactions. How do these other features affect the glass-forming ability of BMGs? In this manuscript, we perform molecular dynamics simulations to determine how variations in the softness and non-additivity of the repulsive core and form of the interatomic pair potential at intermediate distances affect the GFA of binary alloys. These variations in the interatomic pair potential allow us to introduce geometric frustration and change the crystal phases that compete with glass formation. We also investigate the effect of tuning the strength of the many-body interactions from zero to the full embedded atom model on the GFA for pure metals. We then employ the full embedded atom model for binary BMGs and show that hard-core interactions play the dominant role in setting the GFA of alloys, while other features of the interatomic potential only change the GFA by one to two orders of magnitude. Despite their perturbative effect, understanding the detailed form of the intermetallic potential is important for designing BMGs with cm or greater casting thickness.

Micro-lateral extrusion of Zr55Cu30Al10Ni5bulk metallic glass under low-frequency vibration loading

The effect of vibration on the micro-forming ability of Zr 55 Cu 30 Al 10 Ni 5 bulk metallic glass in its supercooled liquid region was studied. The experiment of micro-extrusion was carried out under different amplitude (38 ∼ 760 N) and different frequency (0.1 ∼ 2.0 Hz) at a fixed temperature of 723 K. The extrusion length was taken as a measure to characterize the micro-forming ability. Results reveal that the extrusion length of bulk metallic glass is effectively improved under vibration loading, and increases with increasing loading frequency and amplitude, whereas the frequency dependence is stronger. The viscosity of bulk metallic glass declines under vibration loading because of a larger free volume concentration and surface effect caused by vibration. This research indicates that the vibration forming is an effective method to enhance the micro-forming ability of bulk metallic glasses.

Insight on the process ability of bulk metallic glasses by thermo-mechanical analysis and dynamic mechanical analysis

Mechanical properties of glassy materials exhibit drastic changes in the supercooled liquid region (SLR). In the current research, mechanical properties in several typical bulk metallic glasses (Pd42.5Ni7.5Cu30P20, Zr56Co28Al16 and Zr58Nb3Cu16Ni13Al10) were measured using both dynamic mechanical analysis and thermo-mechanical analysis. The modifications of elastic modulus (determined by dynamic mechanical analysis) and viscosity (determined by thermo-mechanical analysis) are strongly connected. The experimental findings suggest that the elastic modulus of the bulk metallic glasses evaluated by dynamic mechanical analysis correlates with the length variations determined by thermo-mechanical analysis facilities. Evidences from the experiments demonstrate that the modifications are very important in the bulk metallic glass with a large supercooled region and a large fragility, (Pd42.5Ni7.5Cu30P20), very limited in the bulk metallic glass with a limited supercooled region and a small fragility (Zr56Co28Al16), intermediate in the third case (Zr58Nb3Cu16Ni13Al10). The onset of crystallization corresponds to the end of the deformation of the sample and to the onset of the hardness increase. Therefore, these two different techniques (dynamic mechanical analysis and thermo-mechanical analysis) can provide useful and insightful information on the process ability of these amorphous materials.

Ab initio study of Si doping effects in Pd–Ni–P bulk metallic glass

In order to understand the improved glass formation ability of bulk metallic glass due to Si doping, several structures were obtained by means of density functional theory. The results indicate that Si enters mainly as a substitute for P in the clusters, inducing a shoving effect in the P–P random network. The corresponding electronic properties indicate an enhanced cluster stabilization due to an electronic mechanism, which reduces the density of states at the Fermi level.

Connection between the packing efficiency of binary hard spheres and the glass-forming ability of bulk metallic glasses.

We perform molecular dynamics simulations to compress binary hard spheres into jammed packings as a function of the compression rate R, size ratio α, and number fraction x(S) of small particles to determine the connection between the glass-forming ability (GFA) and packing efficiency in bulk metallic glasses (BMGs). We define the GFA by measuring the critical compression rate R(c), below which jammed hard-sphere packings begin to form "random crystal" structures with defects. We find that for systems with α≳0.8 that do not demix, R(c) decreases strongly with Δϕ(J), as R(c)∼exp(-1/Δϕ(J)(2)), where Δϕ(J) is the difference between the average packing fraction of the amorphous packings and random crystal structures at R(c). Systems with α≲0.8 partially demix, which promotes crystallization, but we still find a strong correlation between R(c) and Δϕ(J). We show that known metal-metal BMGs occur in the regions of the α and x(S) parameter space with the lowest values of R(c) for binary hard spheres. Our results emphasize that maximizing GFA in binary systems involves two competing effects: minimizing α to increase packing efficiency, while maximizing α to prevent demixing.