Boosting the discovery of 3D topological materials: mixing chemistry with physics via a two-step computational screening strategy

Boosting the discovery of 3D topological materials: mixing chemistry with physics via a two-step... Topological materials in crystal solids, including topological insulators (TIs), crystalline insulators (TCIs), Dirac semimetals (DSMs), topological Weyl semimetals (WSMs), topological Dirac or Weyl nodal line semimetals (NLSMs) and beyond [1–5], are mainly featured with topological, protected non-trivial surface states and their bulk phases are insulators or semimetals with the proper presence of Dirac cones, Weyl nodes or Dirac nodal lines around the Fermi level. Because of their potential for use in spintronics, quantum computers and novel physics, topological materials have attracted interest. Currently, the design of topological materials with appealing properties still heavily relies on a traditional trial-and-error method, despite the achievements that have been made by combining state-of-the-art first-principles calculations and advanced experiments. Among the 3D topological materials reported to date, two main features can be observed. The first is that completely filled bands are necessary (Fig. 1a). This feature ensures the insulating behavior of the bulk of TIs or the appearance of semimetals with the possible existence of bulk Dirac cones, Weyl nodesand Dirac nodal lines near the Fermi level in DSMs, WSMs, NLSMs and other semimetals. The second important prerequisite is the occurrence of the so-called band inversion [6] (Fig. 1b), which can be impacted by many factors [1–5], including chemical bonding, crystal field effects, spin-orbit coupling (SOC), strong electron correlations and even van der Waals forces. Figure 1. View largeDownload slide Schematic of a two-step computational screening strategy. The left panels denote the concepts of fully filling the bands (a) according to the electronegativity in the periodic table of elements and of analysing the band inversion (b) in association with the topological states. The right panels (c) show the cases of a famous TI (Bi2Se3 family) and DSM (Na3Bi family) from the first computational screen with electronegativity. Figure 1. View largeDownload slide Schematic of a two-step computational screening strategy. The left panels denote the concepts of fully filling the bands (a) according to the electronegativity in the periodic table of elements and of analysing the band inversion (b) in association with the topological states. The right panels (c) show the cases of a famous TI (Bi2Se3 family) and DSM (Na3Bi family) from the first computational screen with electronegativity. Combining these two common features—fully filled bands and band inversion—is a natural approach to developing high-throughput computational screening of topological materials in computational materials science. The first criterion of fully filled bands in materials can be addressed by utilizing the simple chemical concept of electronegativity, which is a property that describes the tendency of an atom to attract electrons (or electron density). For instance, the famous 3D TIs (Bi2Se3 family [7]) are recognized to form fully filled orbitals. As illustrated in Fig. 1c, the composition of Bi2Se3 satisfies the fully filled p6 orbitals because the Se atom has an electronegativity of 2.55, which is larger than that of Bi (2.02). Using the concept of electronegativity, Se has a Se2− configuration with fully filled p6 orbitals by obtaining extra electrons from Bi, which has a Bi3+ configuration. In contrast, Bi may exhibit a Bi3− configuration by obtaining three electrons to fully fill the 6p6 orbital. This requires that the electronegativity of the other counterparts in the materials is smaller than that of Bi. As illustrated in Fig. 1c, because the electronegativity of the alkali metals is smaller than that of Bi, it can form a fully filled 6p6 orbital in Na3Bi by transferring the Na 3s1 electron to Bi. This is true, and Na3Bi was the first experimental realization of DSMs [8,9] after our prediction [4] in collaboration with Professor Zhong Fang's group at the Institute of Physics, Chinese Academy of Sciences, which led to important progress, both theoretically and experimentally. Similarly to Na3Bi, the Sr3Bi2 family was theoretically demonstrated to be wide-gap TIs [10] because the alkali-earth metals have a smaller electronegativity value than Bi. Of course, it needs to be emphasized that the so-called fully filled bands not only denote the bands from atomic orbitals with full shell structure (such as 8-electron or 18-electron rules), but also involve the bands from sub-group of atomic orbitals well separated by crystal field, such as fully filled dtg orbitals in cubic crystal field. Certainly, the first electronegativity screening strategy for the fully filled bands is not a final answer as to whether the screened materials are topological. The main purpose is to narrow the wide search down to a manageable window for potential candidates. This narrowing will heavily reduce the time-consuming aspect of computational design. By using key empirical knowledge—namely some heavy metallic elements, such as Bi, Pb, Tl and Hg, have a strong SOC and possibly induce the band inversion by interacting with the crystal field effects—in combination with the electronegativity considerations, the search can be more effective. It should be emphasized that the strong SOC is not the only important factor that causes the topological states. For instance, the topological states in TCIs (e.g. rock-salt SnTe [11] and SnSe [12,13]) is protected by the mirror symmetry of the lattice (rather than SOC) and one type of Dirac NLSMs requires the SOC strength to be as small as possible, such as pure alkali-earth, fully filled s-orbital beryllium [14]. Therefore, once the first screen is performed, the second screening strategy is extremely important to physically identify the topological states of a targeted material and to analyse the topologically protected surface states. The second screen needs to calculate the topological invariants (Z or Z2) [15]. For a lattice with inversion symmetry, the Z2 invariant can be derived by calculating the parities of all occupied bands at eight time-reversal invariant momenta (TRIMs) in the Brillouin zone (BZ) [16]. However, no matter whether the lattice has an inversion center, the Wilson loop method is useful to identify topological properties by elucidating the Wannier center evolution loop of each occupied band in the TRIMs-contained planes of the BZ [17,18]. The Z2 invariant is related to the summation of winding numbers on the cylinder surface along the loop between two TRIM points for all paired Wannier centers. The non-trivial topological property can be identified by a count of how many times (even or odd) the evolution lines cross an arbitrary reference line [18]. The ongoing efforts to discover topological materials involve developing a high-throughput computational code to design topological materials via combining two-step computational screens within the frameworks of the Big Data, Materials Informatics, and Materials Genome Initiative. The core challenge is to elucidate a high-throughput, fast, accurate algorithm to boost the predictive power for the topological states among the various realistic complexities of topological materials. FUNDING This work was supported by the Hundred Talents Project of the Chinese Academy of Sciences, the National Natural Science Foundation of China (51671193) and the Science Challenging Project (TZ2016004). References 1. Qi X-L , Zhang S-C . Rev Mod Phys 2011 ; 83 : 1057 . CrossRef Search ADS 2. Wan XG , Turner AM , Vishwanath A et al. Phys Rev B 2011 ; 83 : 205101 . CrossRef Search ADS 3. Weng HM , Fang C , Fang Z et al. Phys Rev X 2015 ; 5 : 011029 . 4. Wang ZJ , Sun Y , Chen X-Q et al. Phys Rev B 2012 ; 85 : 195320 . CrossRef Search ADS 5. Fang C , Weng HM , Dai X et al. Chin Phys B 2016 ; 25 : 117106 . CrossRef Search ADS 6. Haldane FDM Phys Rev Lett. 1998 ; 61 : 2015 . CrossRef Search ADS 7. Zhang HJ , Liu C-X , Qi X-L et al. Nat Phys 2009 ; 5 : 438 . CrossRef Search ADS 8. Liu ZK , Zhou B , Zhang Y et al. Science 2014 ; 343 , 864 – 7 . CrossRef Search ADS PubMed 9. Xiong J , Kushwaha SK , Liang T et al. Science 2015 ; 350 : 413 – 6 . CrossRef Search ADS PubMed 10. Li RH , Xie Q , Cheng XY et al. Phys Rev B 2015 ; 92 : 205130 . CrossRef Search ADS 11. Dziawa P , Kowalski BJ , Dybko K et al. Nat Mater 2012 ; 11 : 1023 . CrossRef Search ADS PubMed 12. Sun Y , Zhong ZC , Shirakawa T et al. Phys Rev B 2013 ; 88 : 235122 . CrossRef Search ADS 13. Wang ZY et al. Adv Mater 2015 ; 27 : 4150 . CrossRef Search ADS PubMed 14. Li RH , Ma H , Cheng XY et al. Phys Rev Lett 2016 ; 117 : 096401 . CrossRef Search ADS PubMed 15. Kane CL , Mele EJ . Phys Rev Lett 2005 ; 95 : 146802 . CrossRef Search ADS PubMed 16. Fu L , Kane CL . Phys Rev B 2006 ; 74 : 195312 . CrossRef Search ADS 17. Soluyanov AA , Vanderbilt D . Phys Rev B 2011 ; 83 : 035108 . CrossRef Search ADS 18. Yu R , Qi XL , Bernevig A et al. Phys Rev B 2011 ; 84 : 075119 . CrossRef Search ADS © The Author(s) 2017. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png National Science Review Oxford University Press

Boosting the discovery of 3D topological materials: mixing chemistry with physics via a two-step computational screening strategy

National Science Review , Volume Advance Article (3) – May 12, 2017

Loading next page...
 
/lp/ou_press/boosting-the-discovery-of-3d-topological-materials-mixing-chemistry-0FoDR5jBRe
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd.
ISSN
2095-5138
eISSN
2053-714X
D.O.I.
10.1093/nsr/nwx053
Publisher site
See Article on Publisher Site

Abstract

Topological materials in crystal solids, including topological insulators (TIs), crystalline insulators (TCIs), Dirac semimetals (DSMs), topological Weyl semimetals (WSMs), topological Dirac or Weyl nodal line semimetals (NLSMs) and beyond [1–5], are mainly featured with topological, protected non-trivial surface states and their bulk phases are insulators or semimetals with the proper presence of Dirac cones, Weyl nodes or Dirac nodal lines around the Fermi level. Because of their potential for use in spintronics, quantum computers and novel physics, topological materials have attracted interest. Currently, the design of topological materials with appealing properties still heavily relies on a traditional trial-and-error method, despite the achievements that have been made by combining state-of-the-art first-principles calculations and advanced experiments. Among the 3D topological materials reported to date, two main features can be observed. The first is that completely filled bands are necessary (Fig. 1a). This feature ensures the insulating behavior of the bulk of TIs or the appearance of semimetals with the possible existence of bulk Dirac cones, Weyl nodesand Dirac nodal lines near the Fermi level in DSMs, WSMs, NLSMs and other semimetals. The second important prerequisite is the occurrence of the so-called band inversion [6] (Fig. 1b), which can be impacted by many factors [1–5], including chemical bonding, crystal field effects, spin-orbit coupling (SOC), strong electron correlations and even van der Waals forces. Figure 1. View largeDownload slide Schematic of a two-step computational screening strategy. The left panels denote the concepts of fully filling the bands (a) according to the electronegativity in the periodic table of elements and of analysing the band inversion (b) in association with the topological states. The right panels (c) show the cases of a famous TI (Bi2Se3 family) and DSM (Na3Bi family) from the first computational screen with electronegativity. Figure 1. View largeDownload slide Schematic of a two-step computational screening strategy. The left panels denote the concepts of fully filling the bands (a) according to the electronegativity in the periodic table of elements and of analysing the band inversion (b) in association with the topological states. The right panels (c) show the cases of a famous TI (Bi2Se3 family) and DSM (Na3Bi family) from the first computational screen with electronegativity. Combining these two common features—fully filled bands and band inversion—is a natural approach to developing high-throughput computational screening of topological materials in computational materials science. The first criterion of fully filled bands in materials can be addressed by utilizing the simple chemical concept of electronegativity, which is a property that describes the tendency of an atom to attract electrons (or electron density). For instance, the famous 3D TIs (Bi2Se3 family [7]) are recognized to form fully filled orbitals. As illustrated in Fig. 1c, the composition of Bi2Se3 satisfies the fully filled p6 orbitals because the Se atom has an electronegativity of 2.55, which is larger than that of Bi (2.02). Using the concept of electronegativity, Se has a Se2− configuration with fully filled p6 orbitals by obtaining extra electrons from Bi, which has a Bi3+ configuration. In contrast, Bi may exhibit a Bi3− configuration by obtaining three electrons to fully fill the 6p6 orbital. This requires that the electronegativity of the other counterparts in the materials is smaller than that of Bi. As illustrated in Fig. 1c, because the electronegativity of the alkali metals is smaller than that of Bi, it can form a fully filled 6p6 orbital in Na3Bi by transferring the Na 3s1 electron to Bi. This is true, and Na3Bi was the first experimental realization of DSMs [8,9] after our prediction [4] in collaboration with Professor Zhong Fang's group at the Institute of Physics, Chinese Academy of Sciences, which led to important progress, both theoretically and experimentally. Similarly to Na3Bi, the Sr3Bi2 family was theoretically demonstrated to be wide-gap TIs [10] because the alkali-earth metals have a smaller electronegativity value than Bi. Of course, it needs to be emphasized that the so-called fully filled bands not only denote the bands from atomic orbitals with full shell structure (such as 8-electron or 18-electron rules), but also involve the bands from sub-group of atomic orbitals well separated by crystal field, such as fully filled dtg orbitals in cubic crystal field. Certainly, the first electronegativity screening strategy for the fully filled bands is not a final answer as to whether the screened materials are topological. The main purpose is to narrow the wide search down to a manageable window for potential candidates. This narrowing will heavily reduce the time-consuming aspect of computational design. By using key empirical knowledge—namely some heavy metallic elements, such as Bi, Pb, Tl and Hg, have a strong SOC and possibly induce the band inversion by interacting with the crystal field effects—in combination with the electronegativity considerations, the search can be more effective. It should be emphasized that the strong SOC is not the only important factor that causes the topological states. For instance, the topological states in TCIs (e.g. rock-salt SnTe [11] and SnSe [12,13]) is protected by the mirror symmetry of the lattice (rather than SOC) and one type of Dirac NLSMs requires the SOC strength to be as small as possible, such as pure alkali-earth, fully filled s-orbital beryllium [14]. Therefore, once the first screen is performed, the second screening strategy is extremely important to physically identify the topological states of a targeted material and to analyse the topologically protected surface states. The second screen needs to calculate the topological invariants (Z or Z2) [15]. For a lattice with inversion symmetry, the Z2 invariant can be derived by calculating the parities of all occupied bands at eight time-reversal invariant momenta (TRIMs) in the Brillouin zone (BZ) [16]. However, no matter whether the lattice has an inversion center, the Wilson loop method is useful to identify topological properties by elucidating the Wannier center evolution loop of each occupied band in the TRIMs-contained planes of the BZ [17,18]. The Z2 invariant is related to the summation of winding numbers on the cylinder surface along the loop between two TRIM points for all paired Wannier centers. The non-trivial topological property can be identified by a count of how many times (even or odd) the evolution lines cross an arbitrary reference line [18]. The ongoing efforts to discover topological materials involve developing a high-throughput computational code to design topological materials via combining two-step computational screens within the frameworks of the Big Data, Materials Informatics, and Materials Genome Initiative. The core challenge is to elucidate a high-throughput, fast, accurate algorithm to boost the predictive power for the topological states among the various realistic complexities of topological materials. FUNDING This work was supported by the Hundred Talents Project of the Chinese Academy of Sciences, the National Natural Science Foundation of China (51671193) and the Science Challenging Project (TZ2016004). References 1. Qi X-L , Zhang S-C . Rev Mod Phys 2011 ; 83 : 1057 . CrossRef Search ADS 2. Wan XG , Turner AM , Vishwanath A et al. Phys Rev B 2011 ; 83 : 205101 . CrossRef Search ADS 3. Weng HM , Fang C , Fang Z et al. Phys Rev X 2015 ; 5 : 011029 . 4. Wang ZJ , Sun Y , Chen X-Q et al. Phys Rev B 2012 ; 85 : 195320 . CrossRef Search ADS 5. Fang C , Weng HM , Dai X et al. Chin Phys B 2016 ; 25 : 117106 . CrossRef Search ADS 6. Haldane FDM Phys Rev Lett. 1998 ; 61 : 2015 . CrossRef Search ADS 7. Zhang HJ , Liu C-X , Qi X-L et al. Nat Phys 2009 ; 5 : 438 . CrossRef Search ADS 8. Liu ZK , Zhou B , Zhang Y et al. Science 2014 ; 343 , 864 – 7 . CrossRef Search ADS PubMed 9. Xiong J , Kushwaha SK , Liang T et al. Science 2015 ; 350 : 413 – 6 . CrossRef Search ADS PubMed 10. Li RH , Xie Q , Cheng XY et al. Phys Rev B 2015 ; 92 : 205130 . CrossRef Search ADS 11. Dziawa P , Kowalski BJ , Dybko K et al. Nat Mater 2012 ; 11 : 1023 . CrossRef Search ADS PubMed 12. Sun Y , Zhong ZC , Shirakawa T et al. Phys Rev B 2013 ; 88 : 235122 . CrossRef Search ADS 13. Wang ZY et al. Adv Mater 2015 ; 27 : 4150 . CrossRef Search ADS PubMed 14. Li RH , Ma H , Cheng XY et al. Phys Rev Lett 2016 ; 117 : 096401 . CrossRef Search ADS PubMed 15. Kane CL , Mele EJ . Phys Rev Lett 2005 ; 95 : 146802 . CrossRef Search ADS PubMed 16. Fu L , Kane CL . Phys Rev B 2006 ; 74 : 195312 . CrossRef Search ADS 17. Soluyanov AA , Vanderbilt D . Phys Rev B 2011 ; 83 : 035108 . CrossRef Search ADS 18. Yu R , Qi XL , Bernevig A et al. Phys Rev B 2011 ; 84 : 075119 . CrossRef Search ADS © The Author(s) 2017. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

National Science ReviewOxford University Press

Published: May 12, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off