Accurate prediction of band gap of materials using stacking machine learning model

New materials are an important driving force for social development.In recent years, attention on boron-containing materials is growing in the fields of new energy and catalysis, and calls for compelling need for in-depth study of their structure-property relationship to contribute to the research and development of boron-containing materials.In this work, on the aware of the shift of materials research from traditional trial-and-error paradigm to data-driven research paradigm, we explored the application of ten important machine learning algorithms in the prediction of band gaps of boron-containing materials through feature selection (Pearson correlation analysis), grid search-based optimization (Model optimal parameters), and feature importance analysis (interpretability analysis of the model).The results show that the band gap prediction model using the Random Forest algorithm outperforms the other models with a 84% prediction accuracy, and the total magnetization of boron-containing materials is identified to significantly correlate negatively with the band gap, i.e. the smaller the total magnetization of the material, the larger its band gap.The advantage of the Random Forest algorithm over other models is that it is better able to capture correlations between features.For example, the linear model is unable to detect the importance of the total magnetization of boron-containing materials from the material features, thus leading to a lower model prediction performance.This work shows that machine learning methods can be used to guide the design of boron-containing materials with specific band gap.Meanwhile, the results also show that, as an integrated learning model, Random Forest has good learning ability and stable prediction performance, and can be applied to the prediction of band gap and other material properties of other types of material systems, accelerating the design and optimization process of material properties, and is of great scientific significance for the rapid screening and high-performance prediction of new functional materials.

Explainable machine learning for predicting the band gaps of ABX3 perovskites

Prognostic models for diffuse large B-cell lymphoma (DLBCL), such as the International Prognostic Index (IPI) are widely used in clinical practice. The models are typically developed with simplicity in mind and thus do not exploit the full potential of detailed clinical data. This study investigated whether nationwide lymphoma registries containing clinical data and machine learning techniques could prove to be useful for building modern prognostic tools. This study was based on nationwide lymphoma registries from Denmark and Sweden, which include large amounts of clinicopathologic data. Using the Danish DLBCL cohort, a stacking approach was used to build a new prognostic model that leverages the strengths of different survival models. To compare the performance of the stacking approach with established prognostic models, cross-validation was used to estimate the concordance index (C-index), time-varying area under the curve, and integrated Brier score. Finally, the generalizability was tested by applying the new model to the Swedish cohort. In total, 2,759 and 2,414 patients were included from the Danish and Swedish cohorts, respectively. In the Danish cohort, the stacking approach led to the lowest integrated Brier score, indicating that the survival curves obtained from the stacking model fitted the observed survival the best. The C-index and time-varying area under the curve indicated that the stacked model (C-index: Denmark [DK], 0.756; Sweden [SE], 0.744) had good discriminative capabilities compared with the other considered prognostic models (IPI: DK, 0.662; SE, 0.661; and National Comprehensive Cancer Network-IPI: DK, 0.681; SE, 0.681). Furthermore, these results were reproducible in the independent Swedish cohort. A new prognostic model based on machine learning techniques was developed and was shown to significantly outperform established prognostic indices for DLBCL. The model is available at https://lymphomapredictor.org .

Undiscovered perovskite materials for applications in capturing solar lights are explored through the implementation of data science. In particular, 15000 perovskite materials data is analyzed where visualization of the data reveals hidden trends and clustering of data. Random forest classification within machine learning is used in order to predict the band gap of perovskite materials where 18 physical descriptors are revealed to determine the band gap. With trained random forest, 9328 perovskite materials with potential for applications in solar cell materials are predicted. The selected Li and Na based perovskite materials within predicted 9328 perovskite materials are evaluated with first principle calculations where 11 undiscovered Li(Na) based perovskite materials fall into the ideal band gap and formation energy ranges for solar cell applications. Thus, the implementation of data science accelerates the discovery of hidden perovskite materials and the approach can be applied to the materials science in general for searching undiscovered materials.

This paper presents simulation, design and fabrication of surface acoustic wave (SAW) and Lamb wave micro-phononic devices. To illustrate the band gap formation and possible applications, numerical simulations on the air/silicon phononic crystals (PCs) were conducted which include: surface acoustic waves (SAWs) in a half-space and Lamb waves in a plate. With a lattice constant of 20 µm and high filling fraction, complete band gaps in the range of hundred MHz were found. Based on the band gap properties, numerical simulations on the point defects, waveguides and cavities in PC plates were then performed and discussed. On the experimental side, the associated micro acoustic resonators for both SAW and Lamb waves are demonstrated. Results on the fabrication and measurements of the silicon based micro-PC devices in the hundred MHz ranges are encouraging and may find potential applications in the areas of wireless filters.

The band gap and mechanical control ability of two-dimensional materials largely determine the application value of two-dimensional devices in optical and electronic properties, so the bandgap controllability of two-dimensional materials broadens the application range of multi-functional devices. In the layered van der Waals (vdW) material AgInP2S6, the band gap can be adjusted by the number of layers and flexible strain, and the few layers AgInP2S6 have discrete band gap values, which are also relevant for optoelectronic applications. In the strain range of up to 2.7% applied, the band gap can be adjusted, and the film is relatively stable under strain. We further analyzed the physical mechanism of flexible strain band gap regulation and found that strain-regulation reduced the band gap and increased the chemical bond length. These studies open up new opportunities for the future development of vdW material photoelectric resonators represented by AgInP2S6, and have important reference value.

Environmental forces can produce undesired vibrations in mechanical structures that can limit the precision of mechanical equipment and cause mechanical failure. Piezoelectric-shunt damping is an attractive technique for controlling the vibrating structures, which is reliable, economical and light-weight. Phononic crystal is an internal component whose elastic constant, density and sound velocity change periodically. When the elastic wave passes through a phononic crystal, special dispersion curve is formed due to the interaction of periodic arrangement materials. In order to study the electromagnetic oscillation band gap of the piezoelectric phononic crystal with LCR shunt network at torsional and flexural vibration, we propose a new phononic piezoelectric beam, which is composed of aluminum and epoxy resin. When the piezoelectric patch is strained, the electrical energy is dissipated as current flows through an external LCR shunting circuit. By combining the piezoelectric effect with the mechanical vibration of the smart material, the equivalent additional stress of piezoelectric patches is deduced. Moreover, coupling the energy band theory of phononic crystal with the effect of electromagnetic oscillation, we calculate the band gap characteristics of torsional and flexural vibration of intelligent material. Using the transfer matrix method and Bloch theorem for periodic boundary conditions, the band gap of the phononic beam can be calculated. With the increase of resistance, the amplitude attenuation of the band gap decreases. However, it can expand the frequency range. The inherent frequency of the electromagnetic oscillation is 1/[2π√L(C + CP)]. The sum of capacitance and inherent capacitance is the total capacitance of the shunting circuit. Therefore, the frequency of the electromagnetic oscillation decreases with the increases of the capacitance and inductance. The amplitude attenuation of the band gap increases with the increase of the inductance and decreases greatly with the rise of the capacitance. Three main differences between the LCR shunt networks and traditional circuits are found. First, the band gaps of the phononic piezoelectric smart material are composed of Bragg band gaps and local resonant band gaps. The former one is due to the mismatch between aluminum and epoxy resin, which makes the elastic waves have no corresponding vibration modes at certain frequencies. The latter one is from the effect of electromagnetic oscillation in LCR shunt networks, which consume the energy by resistor. Second, by tuning the resistance, capacitance and inductance, we can change the singularity position and stress magnitude of equivalent additional force curve. The amplitude attenuations of locally resonant band gaps and electromechanical coupling coefficient will be changed. Third, both locations and widths of the band gaps can be tuned by simply varying the value of negative capacitance of the shunting networks without needing to modify the configuration of the structure. Therefore, it provides a new idea for controlling the vibration and reducing the noise of the phononic piezoelectric smart material.

Wide band gap oxides, such as ZnO, SnO 2 and ZrO 2 , are functional materials with a wide range of applications in several important technological areas such as those including lighting, transparent electronics, sensors, catalysis and biolabeling. Recently, doping and co-doping of oxides with lanthanides have attracted a strong interest for lighting purposes, especially among them nanophosphors for bioassays. Tailoring the crystalline materials physical properties for such applications often requires a well-controlled incorporation of dopants in the material lattice and a comprehensive understanding of their role in the oxides matrices. These undoped or intentionally doped oxides have band gap energies exceeding 3.3 eV at room temperature and are known to exhibit optically active centers that span from the ultraviolet to the near infrared region. Typically, by using photon energy excitation above the materials band gap, high quality undoped materials display narrow emission lines near the band edge due to free and bound-exciton recombination, as well as shallow donor-acceptor recombination pairs. Additionally, broad emission bands are often observed in these wide band gap hosts, hampering some of the desired physical properties for further applications. Recognizing and understanding the role of the dopant-related defects when deliberately introduced in the oxide hosts, as well as their influence on the samples luminescence properties, constitutes a matter of exploitation by the scientific community worldwide. In this work, we investigate the luminescence properties of undoped and lanthanide doped oxide materials grown by laser assisted techniques. Laser assisted flow deposition (LAFD) and pulse laser ablation in liquids (PLAL) were used for the growth of ZnO, SnO 2 and yttria stabilized ZrO 2 (YSZ) micro and nanocrystals with different morphologies, respectively. Regarding the YSZ host, trivalent lanthanide ions were optically activated by in-situ doping and co-doping. The influence of the defect energy states on the optical properties of the different undoped and doped metal oxide hosts is investigated under ultraviolet and infrared excitation by means of photoluminescence and photoluminescence excitation.

The crystal structure of Tl2TeS3 (orthorhombic, Pnma, No. 62, Z = 4, Pearson code oP24, a = 815.5(2) pm, b = 1352.6(6) pm, c = 596.48(15) pm, 600 unique reflections with Io > 2σ(Io), R1 = 0.0541, wR2 = 0.0997, GOF = 0.891, T = 298(2) K) in a new structure type has been determined. It can be understood as a LaI2-like arrangement of thallium cations and thiotellurite anions TeS32-. With a direct band gap of 1.61 eV and an indirect band gap of 0.9 eV, Tl2TeS3 is a new material that might be useful for thermoelectrical applications. The materials band gap is significantly lower compared to similar alkali metal systems. In addition, other than alkali metal thiotellurites, Tl2TeS3 is not sensitive toward the atmosphere (especially moisture). A synthetic strategy for improved thermoelectrical materials in the future might be a further reduction of the band gap by introduction of transition metals in such Tl/Te/S systems. Furthermore, density functional studies indicate that effects leading to structural distortions due to the thallium 6s2 lone pair are less important in sulfur compounds than in the analogous oxygen compounds.

Ultrafast laser processing of zirconia/alumina nanocomposite ceramics, the current gold standard material for ceramic bearing components in orthopedics, was investigated. Instead of considering the substrate as a homogeneous material, as commonly assumed in laser micromachining, the damage behavior of different phases around the laser ablation threshold upon ultrafast laser irradiation was investigated. Under appropriate experimental conditions, the zirconia phase was selectively ablated while the alumina phase remained intact. The origin of this selective ablation behavior and its relationship with the material band gaps were discussed. Due to the nonlinear absorption mechanisms under ultrafast laser irradiation, the zirconia phase, with its band gap of 5.8 eV, can absorb more laser energy than the alumina phase which has a larger band gap of 8.8 eV. The negligible heat diffusion length ensures that the absorbed laser energy remains confined in the individual phases, leading to the selective ablation of zirconia phase under the given laser fluence. Based on this observation, an ultrafast laser selective phase removal method which can be used to modify the surface composition of nanocomposite materials consisting of phases with different band gaps was proposed.

Background. Apatite-related modified calcium phosphates have increasing interest for use in orthopedics as bone substitutes. In addition, partial substitution of phosphate by vanadate-group in the anionic sublattice opens up wide opportunities in the development of catalysts for organic synthesis, as well as materials with special optical properties. The aim of the work are the synthesis of apatite-related calcium phosphate-vanadates and the study of the effect of partial substitution of phosphate by the vanadate-anion in the structure and the sorption of zinc cations from an aqueous solution on the band gap of materials. Methods. The samples were synthesized from aqueous solutions of the system NH4+-Сa2+-PO43--NO3--VO43- with molar ratios Сa2+:PO43-:-VO43- = 10: (6-х): х, heated to 500°С and used for sorption of Zn2+ cations from the aqueous solution with subsequent heating to 500°С for 2 hours. The methods of powder X-ray diffraction, FTIR and electron spectroscopy were used for their characterization. Results. According to the X-ray diffraction data, the synthesized samples are monophasic and belong to the hexagonal system, space group P63/m (apatite-type structure), and the calculated parameters of the lattice increase as the vanadate content in their composition increases. FTIR spectroscopy data confirm the presence of two types of anions (РО4 and VO4) in the composition of the synthesized phases. A decrease in the band gap width was established as the degree of substitution of phosphate anion by vanadate increased to 50%, as well as upon sorption of Zn2+ cations onto the surface of synthesized nanoparticles of vanadate-containing hydroxyapatites and their heating to 500°С. Сonclusions. The obtained results can be used in the future in the development of materials with special optical properties or catalysts for organic transformations based on apatite-related calcium phosphate-vanadates.

With the proliferation of directional and horizontal wells, modeling the mechanics of cuttings transport from its downhole origin to the shale shaker is essential in drilling operations since problems observed in vertical trajectories tend to become more complex when encountered in deviated sections. A good example is the efficacy of hole cleaning operations, a straightforward affair in vertical wells. However, when the well's orientation is inclined, it quickly becomes a complex interplay of fluid rheology, cutting bed features, wellbore properties, and in-situ drilling hardware. Inefficient hole cleaning can lead to problems such as reduced rate of penetration (ROP), stuck pipe due to cuttings buildup, high torque, bit balling, excessive equivalent circulating density, and poor cementing. This is why efficient cuttings removal remains a crucial strategy in optimizing the economics of drilling operations. Using experimental data from Yu et al. (2007) at the University of Tulsa, this work set out to model cuttings concentration from fluid rheology and drilling parameters using several machine learning (ML) techniques accessible as open-source packages in the Python environment. Fluid density, yield point, plastic viscosity, flow rate, temperature, inclination, hole eccentricity, pipe rotation, and ROP were all controllable parameters in this experiment. Our exploratory analysis sought to understand relationships among these drilling parameters and the degree of correlation between these features and the objective. Pearson's coefficient and the Gini impurity coefficient values suggest that fluid density, flow rate, and pipe rotation were the focal variables influencing our predictions of the concentration of cuttings. The investigated algorithms included ridge regression, support vector machines, several ensemble approaches, and neural networks. Several base models were created in this study using default parameters to establish baseline performances that hyperparameter tuning approaches such as RandomizedSearchCV and GridSearchCV could improve. Mean absolute error (MAE) and correlation coefficient (R) score served as performance metrics to evaluate model performance, with lower MAE and higher R values indicating superior performance. The Random Forest, Gradient Boosting, Adaptive Boosting, and a stacked regression model of the former three models showcase the workflow implemented in this study since they exhibited the most reliable performance across the training and test datasets. When applied to the entire dataset, the stacked model had the best performance, with an MAE of 2.13% cuttings concentration and a correlation coefficient of 0.946, demonstrating the ability of machine learning models to infer wellbore cuttings concentration reliably. Compared to the performance of empirical and machine learning models in the literature surrounding this work, this demonstrates an improvement in our ability to model downhole hole cleaning efficiency vis-a-vis cutting concentrations. Predictions from the model can help the drilling engineer make informed decisions about drilling fluid programs by allowing for a quick and accurate evaluation of hole cleaning conditions.

Theoretical investigation of Cu-containing materials with different valence structure types: BaCu2S2, Li2CuSb, and LiCuS

Predicting bond strength of corroded reinforced concrete after high-temperature exposure: A stacking model and feature selection

Predicting Post-Heart Transplant Composite Renal Outcome Risk in Adults: A Machine Learning Decision Tool