اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدHigh-Efficient ab initio Bayesian Active Learning Method and Applications in Prediction of Two-dimensional Functional Materials
87
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Beyond the conventional trial-and-error method, machine learning offers a great opportunity to accelerate the discovery of functional materials, but still often suffers from difficulties such as limited materials data and unbalanced distribution of target property. Here, we propose the ab initio Bayesian active learning method that combines active learning and high-throughput ab initio calculations to accelerate prediction of desired functional materials with the ultrahigh efficiency and accuracy. We apply it as an instance to a large family (3,119) of two-dimensional hexagonal binary compounds with unbalanced materials property, and accurately screen out the materials with maximal electric polarization and proper photovoltaic band gaps, respectively, whereas the computational costs are significantly reduced by only calculating a few tenths of possible candidates in comparison to the random search. This approach shows enormous advantages for the cases with unbalanced distributions of target property. It can be readily applied to seek for a broad range of advanced materials.
قيم البحث
اقرأ أيضاً
Regression machine learning is widely applied to predict various materials. However, insufficient materials data usually leads to a poor performance. Here, we develop a new voting data-driven method that could generally improve the performance of reg
ression learning model for accurately predicting properties of materials. We apply it to investigate a large family (2135) of two-dimensional hexagonal binary compounds focusing on ferroelectric properties and find that the performance of the model for electric polarization is indeed greatly improved, where 38 stable ferroelectrics with out-of-plane polarization including 31 metals and 7 semiconductors are screened out. By an unsupervised learning, actionable information such as how the number and orbital radius of valence electrons, ionic polarizability, and electronegativity of constituent atoms affect polarization was extracted. Our voting data-driven method not only reduces the size of materials data for constructing a reliable learning model but also enables to make precise predictions for targeted functional materials.
We have developed an efficient and reliable methodology for crystal structure prediction, merging ab initio total-energy calculations and a specifically devised evolutionary algorithm. This method allows one to predict the most stable crystal structu
re and a number of low-energy metastable structures for a given compound at any P-T conditions without requiring any experimental input. Extremely high success rate has been observed in a few tens of tests done so far, including ionic, covalent, metallic, and molecular structures with up to 40 atoms in the unit cell. We have been able to resolve some important problems in high-pressure crystallography and report a number of new high-pressure crystal structures. Physical reasons for the success of this methodology are discussed.
We develop a theoretical and computational framework to study polarons in semiconductors and insulators from first principles. Our approach provides the formation energy, excitation energy, and wavefunction of both electron and hole polarons, and tak
es into account the coupling of the electron or hole to all phonons. An important feature of the present method is that it does not require supercell calculations, and relies exclusively on electron band structures, phonon dispersions, and electron-phonon matrix elements obtained from calculations in the crystal unit cell. Starting from the Kohn-Sham (KS) equations of density-functional theory, we formulate the polaron problem as a variational minimization, and we obtain a nonlinear eigenvalue problem in the basis of KS states and phonon eigenmodes. In our formalism the electronic component of the polaron is expressed as a coherent superposition of KS states, in close analogy with the solution of the Bethe-Salpeter equation for the calculation of excitons. We demonstrate the power of the methodology by studying polarons in LiF and Li2O2. We show that our method describes both small and large polarons, and seamlessly captures Frohlich-type polar electron-phonon coupling and non-Frohlich coupling to acoustic and optical phonons. To analyze in quantitative terms the electron-phonon coupling mechanisms leading to the formation of polarons, we introduce spectral decompositions similar to the Eliashberg spectral function. We validate our theory using both analytical results and direct calculations on large supercells. This study constitutes a first step toward complete ab initio many-body calculations of polarons in real materials.
We have developed a method that can analyze large random grain boundary (GB) models with the accuracy of density functional theory (DFT) calculations using active learning. It is assumed that the atomic energy is represented by the linear regression
of the atomic structural descriptor. The atomic energy is obtained through DFT calculations using a small cell extracted from a huge GB model, called replica DFT atomic energy. The uncertainty reduction (UR) approach in active learning is used to efficiently collect the training data for the atomic energy. In this approach, atomic energy is not required to search for candidate points; therefore, sequential DFT calculations are not required. This approach is suitable for massively parallel computers that can execute a large number of jobs simultaneously. In this study, we demonstrate the prediction of the atomic energy of a Fe random GB model containing one million atoms using the UR approach and show that the prediction error decreases more rapidly compared with random sampling. We conclude that the UR approach with replica DFT atomic energy is useful for modeling huge GBs and will be essential for modeling other structural defects.
We investigate, using a first-principles density-functional methodology, the nature of magnetism in monolayer $1T$-phase of tantalum disulfide ($1T$-TaS$_2$ ). Magnetism in the insulating phase of TaS$_2$ is a longstanding puzzle and has led to a var
iety of theoretical proposals including notably the realization of a two-dimensional quantum-spin-liquid phase. By means of non-collinear spin calculations, we derive $textit{ab initio}$ spin Hamiltonians including two-spin bilinear Heisenberg exchange, as well as biquadratic and four-spin ring-exchange couplings. We find that both quadratic and quartic interactions are consistently ferromagnetic, for all the functionals considered. Relativistic calculations predict substantial magnetocrystalline anisotropy. Altogether, our results suggest that this material may realize an easy-plane XXZ quantum ferromagnet with large anisotropy.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
