Machine Learning to Instruct Single Crystal Growth by Flux Method

Growth of high-quality single crystals is of great significance for research of condensed matter physics. The exploration of suitable growing conditions for single crystals is expensive and time-consuming, especially for ternary compounds because of the lack of ternary phase diagram. Here we use machine learning(ML) trained on our experimental data to predict and instruct the growth. Four kinds of ML methods, including support vector machine(SVM), decision tree, random forest and gradient boosting decision tree, are adopted. The SVM method is relatively stable and works well, with an accuracy of 81% in predicting experimental results. By comparison,the accuracy of laboratory reaches 36%. The decision tree model is also used to reveal which features will take critical roles in growing processes.

Direct observation of the scaling relation between density of states and pairing gap in a dirty superconductor

Theories and experiments on dirty superconductors are complex but important in terms of both theoretical fundamentals and practical applications.These activities are even more challenging when magnetic fields are present because the field distribution,electron density of states,and superconducting pairing potentials become nonuniform.Here,we present tunneling microspectroscopic experiments on NbC single crystals and demonstrate that NbC is a homogeneous dirty superconductor.When applying magnetic fields to the samples,we found that the zero-energy local density of states and the pairing energy gap followed the explicit scaling relation proposed by de Gennes for homogeneous dirty superconductors in high magnetic fields.More significantly,our experimental findings indicate that the validity of the scaling relation extends to magnetic field strengths far below the upper critical field,calling for a new nonperturbative understanding of this fundamental property in dirty superconductors.On the practical side,we used the observed scaling relation to derive a simple and straightforward experimental scheme for estimating the superconducting coherence length of a dirty superconductor in magnetic fields.

Realization of low-energy type-Ⅱ Dirac fermions in(Ir_(1-x)Pt_x)Te_2 superconductors

Topological Dirac semimetals(DSMs) present a kind of topologically nontrivial quantum state of matter, which has massless Dirac fermions in the bulk and topologically protected states on certain surfaces. In superconducting DSMs, the effects of their nontrivial topology on superconducting pairing could realize topological superconductivity in the bulk or on the surface. As superconducting pairing takes place at the Fermi level E_F, to make the effects possible, the Dirac points should lie in the vicinity of E_F so that the topological electronic states can participate in the superconducting paring. Here,we show using angle-resolved photoelectron spectroscopy that in a series of(Ir_(1-x)Pt_x)Te_2 compounds, the type-Ⅱ Dirac points reside around E_F in the superconducting region, in which the bulk superconductivity has a maximum T_c of ～ 3 K.The realization of the coexistence of bulk superconductivity and low-energy Dirac fermions in(Ir_(1-x)Pt_x)Te_2 paves the way for studying the effects of the nontrivial topology in DSMs on the superconducting state.