We used the superconductors in the SuperCon database to construct element vectors and then perform unsupervised learning of their critical temperatures (T$_c$). Only the chemical composition of superconductors was used in this procedure. No physical predictors (neither experimental nor computational) of any kind were used. We achieved the coefficient of determination R$^2$$\simeq$0.93, which is comparable and in some cases higher then similar estimates using other artificial intelligence techniques. Based on this machine learning model, we predicted several new superconductors with high critical temperatures. We also discuss the factors that limit the learning process and suggest possible ways to overcome them.
Cuprates and iron-based superconductors are two classes of unconventional high Tc superconductors based on 3d transition elements. Recently, two principles, correspondence principle and magnetic selective pairing rule, have been emerged to unify their high Tc superconducting mechanisms. These principles strongly regulate electronic structures that can host high Tc superconductivity. Guided by these principles, here we propose high Tc superconducting candidates that are formed by cation-anion trigonal bipyramidal complexes with a d^7 filling configuration on the cation ions. Their superconducti…
In this work we used unsupervised machine learning methods in order to find possible clustering structures in superconducting materials data sets. We used the SuperCon database, as well as our own data sets complied from literature, in order to explore how machine learning algorithms groups superconductors. Both conventional clustering methods like k-means, hierarchical or Gaussian mixtures, as well as clustering methods based on artificial neural networks like self-organizing maps, were used. For dimensionality reduction and visualization t-SNE was found to be the best choice. Our results ind…
In this contribution to the J. Phys. memorial issue in honor of Sandro Massidda I reflect on a phenomenon Sandro had been a part of. While theoretical condensed matter physicists have made, over the years, exciting and most elegant contributions to the theory of superconductivity (which, in and by itself, is one of the most beautiful constructs in theoretical physics), some of them of utmost importance, they have had less success in predicting and explaining superconducting states and mechanisms in specific materials. More down-to-earth computational materials scientists, who often go by the m…
Two-dimensional (2D) superconductors have attracted great attention in recent years due to the possibility of new phenomena in lower dimensions. With many bulk transition metal carbides being well-known conventional superconductors, here we perform first-principles calculations to evaluate the possible superconductivity in a 2D monolayer Mo$_2$C. Three candidate structures (monolayer alpha-Mo$_2$C, 1T MXene-Mo$_2$C, and 2H MXene-Mo$_2$C) are considered and the most stable form is found to be the 2H MXene-Mo$_2$C. Electronic structure calculations indicate that both unpassivated and passivated …
We predict Co-based chalcogenides with a diamond-like structure can host unconventional high temperature superconductivity (high-$T_c$). The essential electronic physics in these materials stems from the Co layers with each layer being formed by vertex-shared CoA$_4$ (A=S,Se,Te) tetrahedra complexes, a material genome proposed recently by us to host potential unconventional high-$T_c$ close to a $d^7$ filling configuration in 3d transition metal compounds. We calculate the magnetic ground states of different transition metal compounds with this structure. It is found that (Mn,Fe,Co)-based comp…
Electronic states within superconducting vortices hold crucial information about paring mechanisms and topology. While scanning tunneling microscopy/spectroscopy(STM/S) can image the vortices, it is difficult to isolate the intrinsic electronic states from extrinsic effects like subsurface defects and disorders. We combine STM/S with unsupervised machine learning to develop a method for screening out the vortices pinned by embedded disorder in Fe-based superconductors. The approach provides an unbiased way to reveal intrinsic vortex-core states and may address puzzles on Majorana zero modes.
The prediction of transition temperatures can be regarded in several ways, either as an exacting test of theory, or as a tool for identifying theoretical rules for defining new homology models. Popular "first principle" methods for predicting transition temperatures in conventional crystalline superconductors have failed for cuprate HTSC, as have parameterized models based on CuO2 planes (with or without apical oxygen). Following a path suggested by Bayesian probability, we find that the glassy, self-organized dopant network percolative model is so successful that it defines a new homology cla…
Hydrogen-rich superhydrides are believed to be very promising high-T$_c$ superconductors, with experimentally observed critical temperatures near room temperature, as shown in recently discovered lanthanide superhydrides at very high pressures, e.g. LaH$_{10}$ at 170 GPa and CeH$_9$ at 150 GPa. With the motivation of discovering new hydrogen-rich high-T$_c$ superconductors at lowest possible pressure, quantitative theoretical predictions are needed. In these promising compounds, superconductivity is mediated by the highly energetic lattice vibrations associated with hydrogen and their interpla…
A wide variety of superconducting oxides are used to test a general model of high pressure induced transition temperature (T c) changes. The T c 's vary from a low of 24 K to a high of 164 K. Although the model is capable of predicting both increases and decreases in T c with pressure, only superconductors that exhibit an increase are considered at this time. Predictions are made of the maximum T^ cP theo for 15 super-conductors as a function of their compressibilities. The theoretical results generally agree well with experiment. This model of T c as a function of pressure is derived from a r…
We develop a multi-step workflow for the discovery of conventional superconductors, starting with a Bardeen Cooper Schrieffer inspired pre-screening of 1736 materials with high Debye temperature and electronic density of states. Next, we perform electron-phonon coupling calculations for 1058 of them to establish a large and systematic database of BCS superconducting properties. Using the McMillan-Allen-Dynes formula, we identify 105 dynamically stable materials with transition temperatures, Tc>5 K. Additionally, we analyze trends in our dataset and individual materials including MoN, VC, VTe, …
Improving the accuracy of circuit quantization using the electromagnetic properties of superconductors
Seoul National University（首尔大学）；Korea Research Institute of Standards and Science（韩国标准科学研究院）；Quantum Devices Labs, Korea Advanced Nano Fab Center（韩国先进纳米制造中心量子器件实验室）；Quantum Design Korea（量子设计韩国公司）；SKKU Adavanced Institute of Nanotechnology, Sungkyunkwan University（成均馆大学SKKU纳米技术高级研究所）
Predicting the transition temperature, Tc, of a superconductor from Periodic Table normal state properties is regarded as one of the grand challenges of superconductivity. By studying the correlations of Periodic Table properties with known superconductors, it is possible to estimate their transition temperatures. Starting from the isotope effect and correlations of superconductivity with electronegativity (\Chi), valence electron count per atom (Ne), atomic number(Z) and formula weight (Fw), we derive an empirical formula for estimating Tc that includes an unknown parameter,(Ko). With average…
Magnesium diboride superconducting wires give the largest critical current density (Jc) when produced with iron sheath. Because iron is ferromagnetic, it is expected to improve the field dependence of Jc by shielding of the external field for low magnetic fields. However, transport and magnetic measurements of Jc reveal that Jc in MgB2/Fe is improved far beyond the effect of simple magnetic shielding. The transport measurements in external field show that Jc initially decreases with the field. This is followed by an increase for intermediate fields and again a decrease for high fields, resembl…
We systematically explore the exquisiteness of Bardeen-Cooper-Schriefer(BCS) theory where the BCS-type electron-phonon interaction is unambiguously reinforced as the only viable superglue in cuprate superconductors because phonon-induced scattering is effectively nil for Cooper pairs (in its original form), and also phonons are never required to Bose-condense. Here, we prove that (i) the Cooper-pair binding energy can be strengthened to obtain high superconductor transition temperature (Tsc) and (ii) the existence of a generalized electron-phonon potential operator that can induce the finite-t…
Tree Models Machine Learning to Identify Liquid Metal based Alloy Superconductor
Technical Institute of Physics and Chemistry, Chinese Academy of Sciences（中国科学院理化技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Tsinghua University（清华大学）
Machine learning techniques have been shown to be effective to recognize different phases of matter and produce phase diagrams in the parameter space interested, while they usually require prior labeled data to perform well. Here, we propose a machine learning procedure, mainly in an unsupervised manner, which can first identify topological/non-topological phases and then refine the locations of phase boundaries. By following this proposed procedure, we expand our previous work on the one-dimensional $p$-wave superconductor [Phys. Rev. B 102, 054512 (2020)] and further on the Su-Schrieffer-Hee…
The recent advances in machine learning algorithms have boosted the application of these techniques to the field of condensed matter physics, in order e.g. to classify the phases of matter at equilibrium or to predict the real-time dynamics of a large class of physical models. Typically in these works, a machine learning algorithm is trained and tested on data coming from the same physical model. Here we demonstrate that unsupervised and supervised machine learning techniques are able to predict phases of a non-exactly solvable model when trained on data of a solvable model. In particular, we …
Much research in recent years has focused on using empirical machine learning approaches to extract useful insights on the structure-property relationships of superconductor material. Notably, these approaches are bringing extreme benefits when superconductivity data often come from costly and arduously experimental work. However, this assessment cannot be based solely on an open black-box machine learning, which is not fully interpretable, because it can be counter-intuitive to understand why the model may give an appropriate response to a set of input data for superconductivity characteristi…
