Session R67: Predictive Discoveries of Novel Two-Dimensional (2D) Materials Through Complementary High-Throughput Approaches

Topological Regions in a 'Map of Materials Properties' of Two-dimensional Materials(*)

We performed extensive density-functional theory calculations of the geometries, band structures, and Z2 topological invariants of 2D honeycomb materials. Using a recently developed artificial-intelligence approach (SISSO, Ref. 1) we identify the actuators (‘material’s genes’) that govern the topological transition. We also draw a 'map of materials properties' which identifies the regions of metals, trivial insulators, and topological insulators. The latter region contains several million newly predicted topologically nontrivial alloyed materials.

1) R. Ouyang, S. Curtarolo, E. Ahmetcik, M. Scheffler, and L.M. Ghiringhelli, SISSO: a compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates. Phys. Rev. Mat. 2, 083802 (2018).

(*) Work performed in collaboration with C. M. Acosta, G. Cao, Ch. Carbogno, A. Fazzio, L. M. Ghiringhelli, H. Liu, R. Ouyang. Z. Zhang.   

Novel two-dimensional materials from high-throughput computational exfoliation

We have performed an extensive high-throughput screening of known inorganic materials, to identify those that could be exfoliated into novel two-dimensional monolayers [1]. The screening protocol first identifies bulk materials that appear layered according to a simple and robust chemical definition of bonding, determining then for all of these the binding energies of the respective monolayers, their electronic state (metallic vs insulating), magnetic configuration (ferro-,ferri- or antiferro-magnetic), and phonon dispersions (to evaluate mechanically stability). Such protocol identifies a portfolio of close to 2,000 inorganic materials that appear either easily or potentially exfoliable. With further data ingestion, this initial portfolio has recently almost doubled, providing an extensive pool to investigate promising properties. First focus has been on the determination of the effective masses and mobilities (from the full solution of the Boltzmann transport equation) for electronic applications; of topological invariants; of superconductivity and charge-density waves; and of photocatalytic parameters for water splitting. Thanks to the use of the AiiDA (http://aiida.net) materials' informatics platform, all the high-throughput calculations can be performed and streamlined in fully searchable and reproducible ways, they are stored in a database with the full provenance tree of all parent and children calculations, and can be shared with the community at large in the form of raw or curated data via the Materials Cloud (http://www.materialscloud.org) dissemination portal.

[1] Nicolas Mounet, Marco Gibertini, Philippe Schwaller, Davide Campi, Andrius Merkys, Antimo Marrazzo, Thibault Sohier, Ivano Eligio Castelli, Andrea Cepellotti, Giovanni Pizzi and Nicola Marzari, Nat. Nanotechnol. 13, 246 (2018).   

Traditional & Exotic Semiconductors in the Two-Dimensional Limit

Recent years have witnessed an amazing expansion in the family of 2D materials, which has led to many exotic physical properties distinctly different from their bulk counterparts. An example is the traditional II-VI compound semiconductors in their monolayer limit [1], for which the experiments have been catching up rapidly. Besides the monolayer honeycomb (MLHC) hBN-like structure, a double layer honeycomb (DLHC) structure was also predicted to be stable over a large portion of ordinary semiconductors [2]. It is becoming clear that when the thickness of a solid is below certain critical value, irrespective of its commonly-observed bulk form, a universal van der Waals stacking may happen, leading to unexpected physical properties. Taking GaAs, a well-known covalent semiconductor, as an example, non-trivial topological properties emerge as a result of the stacking of DLHCs [2], as well as the formation of an excitonic insulator (EI) where the exciton binding energy magically becomes larger than its band gap [3]. In strongly-correlated 2D systems such as MX₂ where M = Ni, Co and X = Cl, Br, on the other hand, half excitonic insulator can also form where one spin channel is an ordinary semiconductor while the other spin channel is an EI, leading to Bose-Einstein condensation [4].
[1] H. Zheng, et al., Phys. Rev. B 92, 115307 (2015).
[2] M. C. Lucking, et al., Phys. Rev. Lett. 120, 086101 (2018).
[3] Z. Jiang, Y. Li, S. B. Zhang, and W. Duan, Phys. Rev. B 98, 081408(R) (2018).
[4] Z. Jiang, Y. Li, W. Duan, and S. B. Zhang, Phys. Rev. Lett. 122, 236402 (2019).   

Opto-electronic excitations of 2D materials: Unraveling similarities through fingerprints(*)

2D materials exhibit peculiar optoelectronic properties [1] that can be further tuned by stacking, functionalization, or creating interfaces [2-4]. Analysis based on many-body approaches provide us insight into their intriguing features. Making use of this information and the large data pool of the NOMAD Repository [5], we apply data-analytics tools to create a pool of fingerprints and find descriptors for particular properties. A novel approach, based on these fingerprints, will be demonstrated to unravel hidden similarities between materials.

[1] S. Haastrup et al., The Computational 2D Materials Database: High-Throughput Modeling and Discovery of Atomically Thin Crystals, 2D Materials 5, 042002 (2018).
[2] K. W. Lau, C. Cocchi, and C. Draxl, Electronic and optical excitations of two-dimensional ZrS2 and HfS2 and their heterostructure, Phys. Rev. Materials 3, 074001 (2019).
[3] W. Aggoune, C. Cocchi, D. Nabok, K. Rezouali, M. Belkhir, and C. Draxl, Dimensionality of excitons in stacked van der Waals materials: The example of hexagonal boron nitride, Phys. Rev. B 97, 241114(R) (2018).
[4] W. Aggoune, C. Cocchi, D. Nabok, K. Rezouali, M. Belkhir, and C. Draxl, Enhanced Light-Matter Interaction in Graphene/h-BN van der Waals Heterostructures, J. Chem. Phys. Lett. 8, 1464 (2017).
[5] https://nomad-repository.eu

(*) Work performed in collaboration with Martin Kuban and Santiago Rigamonti


  

Epitaxial tellurene, phosphorene and MoTe2 derivatives: from theoretical prediction to experimental observations

The family of 2D materials has been expanding continuously showing the diverse structural, symmetrical and electronic properties. Presently, there are two major categories of 2D materials experimentally obtainable: one is by exfoliation from bulk crystals (e.g., graphene) and the other is by epitaxial growth (e.g., silicene), and the latter is more versatile capable of synthesizing materials that would otherwise be not present naturally. In this presentation, some of our synergistic research efforts between theory, modeling, and experiments in the discoveries and further characterizations of a few members of the 2D materials family will be given. These include layered tellurium [1,2], metal-phosphorene network consisted of blue-phosphorene units linked by Au atoms [3,4], and a variational hexagonal phase of molybdenum telluride [5]. Each of these are achieved by molecular-beam epitaxy and their structural and electronic properties are characterized by scanning transmission electron microscopy, scanning tunneling microscopy and scanning tunneling spectroscopy. These materials show potentials of various applications.

[1] Zhili Zhu et al., Phys. Rev. Lett. 118, 046101 (2017)
[2] Jinglei Chen et al., Nanoscale 9, 15945 (2017)
[3] Jinpeng Xu et al., Phys. Rev. Mater. 1, 061002(R) (2017)
[4] Hao Tian et al., Matter, doi.org/10.1016/j.matt.2019.08.001
[5] Junqiu Zhang et al., unpublished