Method for assessing atomic sources of flicker noise in superconducting qubits

Flicker noise causes decoherence in Josephson junction-based superconducting qubits, thus limiting their practical potential as building blocks for quantum computers. This is due to limited length and complexity of executable algorithms, and increased dependency on error-correcting measures. Therefore, identifying and subsiding the atomic sources of flicker noise are of great importance to the development of this technology. We developed a method that combines ab initio DFT calculations and quantum dynamics to model charge transport across a Josephson junction, by which it is possible to more accurately assess different defects as sources of flicker noise. We demonstrate the use of our method in an investigation of various atomic defects, including vacancies, trapping, and substitutions, in an Al|Al2O3|Al Josephson junction. This demonstration both reveals weaknesses in previous attempts to pinpoint the atomic sources of flicker noise and highlights new candidates.

The Reaction of Hydrogen Halides with Tetrahydroborate Anion and Hexahydro-Closo-Hexaborate Dianion

The mechanism of the consecutive halogenation of the tetrahydroborate anion [BH4]− by hydrogen halides (HX, X = F, Cl, Br) and hexahydro-closo-hexaborate dianion [B6H6]2− by HCl via electrophile-induced nucleophilic substitution (EINS) was established by ab initio DFT calculations [M06/6-311++G(d,p) and wB97XD/6-311++G(d,p)] in acetonitrile (MeCN), taking into account non-specific solvent effects (SMD model). Successive substitution of H− by X− resulted in increased electron deficiency of borohydrides and changes in the character of boron atoms from nucleophilic to highly electrophilic. This, in turn, increased the tendency of the B–H bond to transfer a proton rather than a hydride ion. Thus, the regularities established suggested that it should be possible to carry out halogenation more selectively with the targeted synthesis of halogen derivatives with a low degree of substitution, by stabilization of H2 complex, or by carrying out a nucleophilic substitution of B–H bonds activated by interaction with Lewis acids (BL3).

Charge distribution functions for characterization of complex systems

A set of characteristics calculated within the scope of quantum chemistry methods may be assigned to local ones changing from atom to atom in complex systems. Simple averaging of the related values gives rather poor characteristics of the systems because various fractions of certain atoms can have different surrounding and, therefore, different characteristics, which may not correspond to the average one. The aim of this study is searching a more appropriate pathway to transform local characteristics, e.g., atomic charges, into nonlocal ones based on the distribution functions. The distribution functions of atomic charges (CDF) could be considered as a simple tool to analyze nonuniform complex systems since specificity of different fractions of atoms reflects in the CDF shape. As a whole, the approach accuracy and efficiency depend on the quality and appropriateness of molecular and cluster models used, as well as on the quantum chemical methods (ab initio, DFT, and semiempirical) and the basis sets used. Nanosystems with dozens of molecules (clusters, domains, nanodroplets), modelling a liquid phase or interfacial layers, and solid nanoparticles of almost real sizes (> 40 units, > 2 nm) may be considered as more appropriate models of real systems than the models with several molecules and small clusters (< 20 units, < 1 nm). This approach has been applied to a set of representatives of such various materials as activated carbon, porous and nanoparticulate silicas unmodified and modified interacting with nitrogen, methane, water, human serum albumin (HSA) binding doxorubicin molecules. This approach may give information useful upon the analysis of any complex system.

Predicting adsorption ability of adsorbents at arbitrary sites for pollutants using deep transfer learning

Accurately evaluating the adsorption ability of adsorbents for heavy metal ions (HMIs) and organic pollutants in water is critical for the design and preparation of emerging highly efficient adsorbents. However, predicting adsorption capabilities of adsorbents at arbitrary sites is challenging, with currently unavailable measuring technology for active sites and the corresponding activities. Here, we present an efficient artificial intelligence (AI) approach to predict the adsorption ability of adsorbents at arbitrary sites, as a case study of three HMIs (Pb(II), Hg(II), and Cd(II)) adsorbed on the surface of a representative two-dimensional graphitic-C3N4. We apply the deep neural network and transfer learning to predict the adsorption capabilities of three HMIs at arbitrary sites, with the predicted results of Cd(II) > Hg(II) > Pb(II) and the root-mean-squared errors less than 0.1 eV. The proposed AI method has the same prediction accuracy as the ab initio DFT calculation, but is millions of times faster than the DFT to predict adsorption abilities at arbitrary sites and only requires one-tenth of datasets compared to training from scratch. We further verify the adsorption capacity of g-C3N4 towards HMIs experimentally and obtain results consistent with the AI prediction. It indicates that the presented approach is capable of evaluating the adsorption ability of adsorbents efficiently, and can be further extended to other interdisciplines and industries for the adsorption of harmful elements in aqueous solution.