Elsevier

Materials Today Physics

Volume 20, September 2021, 100497
Materials Today Physics

Lone-pair engineering: Achieving ultralow lattice thermal conductivity and enhanced thermoelectric performance in Al-doped GeTe-based alloys

https://doi.org/10.1016/j.mtphys.2021.100497Get rights and content

Highlights

  • Al-induced lone-pair distortion is demonstrated.

  • Al-doing leads to the reduction of phonon velocity and increase of anharmonicity.

  • An ultralow κlat of 0.21 Wm-1 K−1 at 773 K can be obtained.

  • A peak ZT of 2.21 at 773 K and average ZT of 1.51 can be achieved.

Abstract

The lone pair in Ⅳ-Ⅵ compounds plays a significant role on tuning the phonon and electron transport for excellent thermoelectric performance. Here, we theoretically reveal a unique lone-pair distortion feature in Al-doped GeTe and further experimentally demonstrate the reduction of lattice thermal conductivity by Al doping. Due to the reduction of phonon velocity and enhanced anharmonicity, together with other phonon scattering mechanisms of point defects and stacking faults, an ultralow lattice thermal conductivity of ∼0.21Wm-1 K−1 at 773 K can be obtained in Al-doped GeTe. Besides, owing to the distorted distribution of Al-3s lone-pair electrons, the Al also show unusual p-type doping behavior in GeTe. By systemically combining the doping effects of Al, Sb and Pb, we can obtain a high figure of merit (ZT) of 2.21 at 773 K and a high average ZT of 1.51 within 300–773 K in the Ge0.74Al0.02Pb0.1Sb0.1Te. This study demonstrates the impact of lone-pair distortion on enhancing the ZT of GeTe, which also paves a new path way for developing other high-ZT thermoelectric materials.

Introduction

Thermoelectricity provides a sustainable and ecofriendly energy resource by directly converting heat to electricity [1,2]. The performance of thermoelectric (TE) materials at a given temperature (T) is usually evaluated by a dimensionless figure of merit (ZT) [3], ZT = S2Tσ/κ, where S, σ and κ are the Seebeck coefficient, electrical conductivity and thermal conductivity, respectively. The κ generally consists of the lattice thermal conductivity (κlat) and electronic thermal conductivity (κe) [3], where the κe is related to the σ, T and Lorentz number L, κe=LσT. Because the S, σ and κe all show dependence on the carrier density (n), optimizing the n by doping or alloying external elements is a routine way to improve the power factor PF (S2σ) and ZT [3,4]. Besides, many band-engineering and phonon-engineering strategies have been introduced to enhance the ZT by traditional doping and alloying methods [[5], [6], [7], [8], [9], [10]]. However, it is still a very tough work to explore novel mechanisms of dopants for enhancing ZT.

Among the big family of TE materials [3,5,6], GeTe-based alloys are one kind of the best mid-temperature TE materials so far [[11], [12], [13], [14], [15], [16], [17]], demonstrating peak ZT > 2 by doping or alloying with appropriate elements [[18], [19], [20], [21], [22], [23]]. The pristine GeTe usually have high Ge vacancy (VGe) concentration due to the low formation energy of VGe [11,12], and the VA-group elements of Sb and Bi are the mostly used dopants for reducing the high hole-carrier density in GeTe [21,[23], [24], [25], [26], [27], [28], [29]]. The Sb-doped and Bi-doped samples are the best single-dopant GeTe-based TE materials having peak ZT > 1.5, such as single-crystal Ge0.92Sb0.08Te [30], polycrystalline Ge0.9Sb0.1Te [31,32], (GeTe)17Sb2Te3 [22], Ge0.93Bi0.07Te [33], Ge0.95Bi0.05Te1.025 [34] and Ge0.99Bi0.05Te [23]. To further improve the ZT values, other doping elements together with Sb or Bi are usually co-doped into GeTe for tuning the band structures and phonon scatterings, such as Pb-Sb [35,36], Pb-Bi [37,38], Mg-Bi [39], In-Sb [21], Cd-Bi [20], Cu-Sb [40], Cr-Bi [41], Cr-Sb [42] and Zn-Sb [43], Se-Sb [44], achieving enhanced peak ZT > 2. Exploring other doping elements with novel effects is still significant for further improving the ZT of GeTe-based alloys.

As the origin of the structure-property relationship, chemical bonding has been frequently used to understand the transport of phonons and electrons in TE materials [[45], [46], [47], [48]]. The so-called lone pair is formally from the s-valence electron pair (s2), which shows great promise for reducing κlat [[48], [49], [50], [51], [52], [53], [54], [55]]. The s2 lone pair behaves differently depending on the local coordination number [45]. It can be stereochemically “quenched” in the octahedral environment with high structural symmetry, or be stereochemically active to lead strong lattice distortion [45]. The stereochemically active lone-pair electrons usually can amplify the anharmonicity [45,[48], [49], [50], [51],[53], [54], [55]], which is an important factor for low κlat. For the IV-VI compounds (i.e. PbTe, GeTe, SnSe), the lone pairs have been widely used to explain their unique band structures and low κlat for excellent TE properties [45,56]. Waghmare et al. considered the active cation s2 lone pair as the driving force for the structural distortion in those IV-VI compounds [57], such as the Peierls distortion in rhombohedral GeTe [56]. Because the valence shell of Ge-4s24p2 has a pair of s electrons, the lone pair has also been considered as a key feature of bonding in GeTe [43,56], although Waghmare et al. earlier revealed the poor mixing of s and p states on the cation site of GeTe [57]. Therefore, adjusting the lone-pair features in GeTe can provide a new path way to enhance the ZT of GeTe, which however has rarely been considered before.

Herein, we introduce the Al dopant to tune the lone-pair features of GeTe for enhancing their TE performance. The Al-induced distortion of lone pair can reduce the phonon velocity and enhance the anharmonicity, which together with other phonon scattering mechanisms leads to an ultralow κlat of ∼0.21 Wm-1 K−1 at 773 K in our samples. Besides, the Al shows unusual p-type doping behavior in GeTe due to the localized lone-pair electrons of Al-3s, which can be used to optimize the carrier density for higher PF. Thus, owing to the obtained ultralow κlat and optimized PF by Al doping and Pb-Sb co-alloying, a high peak ZT of ∼2.21 at 773 K and high average ZT of ∼1.51 within 300–773 K can be obtained in Ge0.74Al0.02Pb0.1Sb0.1Te.

Section snippets

Analysis of chemical bonding and band structures

To understand the influence of Al doping on the electrical and thermal properties of GeTe, we have performed the density-functional-theory (DFT) calculations. As the Peierls distortion in rhombohedral GeTe can be considered as a very slight change of the lattice angle in cubic GeTe [11,56], we approximately use the cubic supercell model for the DFT calculations of the electronic properties to simplify the theoretical analysis. Fig. 1a and 1b displays the electron localization function (ELF)

Conclusion

In summary, we introduce a lone-pair-distortion strategy to enhance the TE performance of GeTe-based alloys. Due to the Al-induced lone-pair effects, the acoustic phonon branches are flattened for low phonon velocity and the anharmonicity is also amplified for enhancing the phonon-phonon scatterings. Combining the Al-induced lone-pair effects with other synergetic phonon-scattering mechanisms of stacking faults and point defects, an ultralow κlat of ∼0.21 Wm-1 K−1 at 773 K can be obtained in

Experimental section

Synthesis process: Raw materials of Ge, Al, Pb, Sb and Te with 99.99% purity were used to synthesize Ge0.9-xAlxSb0.1Te, Ge0.76-yAlyPb0.1Sb0.1Te and Ge1-zAlzTe. The quartz tubes were coated with carbon in advance to avoid the reaction of Al with the wall of the quartz tubes. After weighing, materials were sealed in carbon-coated quartz tubes in vacuum (∼1⨯10−3 Pa or better). Subsequently, raw samples were heated to 1050 °C in 4 h, kept at 1050 °C for 20 h, and then cooled to 600 °C in 4 h and

Authors contributions

Y. Dou: Investigation, Data Curation, Writing- Original draft; J. Li: Methodology, Resources, Formal analysis, Writing - review & editing; Y. Xie: Investigation, Visualization, Validation Writing- Original draft; X. Wu: Writing- Original draft, Resources; L. Hu: Methodology, Resources; F. Liu: Methodology, Resources; W. Ao: Methodology, Resources; Y. Liu: Writing- Original draft; C. Zhang: Supervision, Conceptualization, Methodology, Writing- Original draft, Writing- Review & Editing,

Data availability

The raw/processed data required to reproduce these findings can be reached from the corresponding author through e-mail.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (grant number: 21805196, 52071218), Natural Science Foundation of Guangdong Province, China (grant number: 2018A030310416), Foundation for Distinguished Young Talents in Higher Education of Guangdong, China (grant number: 2017KQNCX178), Shenzhen Science and Technology Innovation Commission (grant number: 20200731215211001; 20200814110413001 JCYJ20180305124020928), and Shenzhen Clean Energy Research Institute and the

References (75)

  • W. Liu et al.

    Mater. Today Phys.

    (2017)
  • E. Nshimyimana et al.

    Sci. Bull.

    (2018)
  • D. Wu et al.

    J. Alloys Compd.

    (2019)
  • J. Li et al.

    Joule

    (2018)
  • L. Xie et al.

    Nano Energy

    (2020)
  • M. Dutta et al.

    Chem. Sci.

    (2019)
  • C. Chang et al.

    Mater. Today Phys.

    (2018)
  • L. Wang et al.

    Mater. Today Phys.

    (2021)
  • A. Togo et al.

    Scripta Mater.

    (2015)
  • L.E. Bell

    Science

    (2008)
  • J. He et al.

    Science

    (2017)
  • G.J. Snyder et al.

    Nat. Mater.

    (2008)
  • T. Zhu et al.

    Adv. Mater.

    (2017)
  • X.L. Shi et al.

    Chem. Rev.

    (2020)
  • Y. Pei et al.

    Adv. Mater.

    (2012)
  • J. Yang et al.

    Adv. Energy Mater.

    (2013)
  • L. Yang et al.

    Adv. Energy Mater.

    (2018)
  • Z. Chen et al.

    Adv. Mater.

    (2018)
  • M. Hong et al.

    Adv. Mater.

    (2019)
  • W.D. Liu et al.

    Adv. Energy Mater.

    (2020)
  • D. Wu et al.

    J. Am. Chem. Soc.

    (2014)
  • S. Perumal et al.

    J. Mater. Chem. C

    (2016)
  • Z. Liu et al.

    Proc. Natl. Acad. Sci. U.S.A.

    (2018)
  • T. Xing et al.

    Natl. Sci. Rev.

    (2019)
  • Z. Liu et al.

    Adv. Energy Mater.

    (2020)
  • J. Li et al.

    ACS Appl. Mater. Interfaces

    (2019)
  • Y. Gelbstein et al.

    Adv. Energy Mater.

    (2013)
  • M. Hong et al.

    Adv. Energy Mater.

    (2018)
  • M. Hong et al.

    Adv. Mater.

    (2018)
  • X. Xu et al.

    Adv. Sci.

    (2018)
  • J.F. Dong et al.

    Energy Environ. Sci.

    (2019)
  • P. Li et al.

    Adv. Funct. Mater.

    (2020)
  • S. Perumal et al.

    Chem. Mater.

    (2017)
  • Z. Zheng et al.

    J. Am. Chem. Soc.

    (2018)
  • J. Li et al.

    NPG Asia Mater.

    (2017)
  • Y.F. Tsai et al.

    Adv. Mater.

    (2021)
  • R.K. Vankayala et al.

    Adv. Sci.

    (2020)
  • Cited by (23)

    • Suppressing lone-pair expression endows room-temperature cubic structure and high thermoelectric performance in GeTe-based materials

      2022, Materials Today Physics
      Citation Excerpt :

      Pristine GeTe has a large number of intrinsic Ge vacancies, resulting in an extremely high carrier concentration of 1021 cm−3 at room temperature [23,24], which far exceeds the optimal carrier concentration, about 1020 cm−3 [23,24], and leads to a relatively low ZT. Therefore, reducing the carrier concentration through aliovalent ion doping or alloying is an effective way to enhance the thermoelectric performance of GeTe [24–31]. Similar to PbTe, GeTe has multivalence bands and intrinsically strong anharmonicity [23,32], ensuring a large PF and a small κL.

    View all citing articles on Scopus
    1

    Author contributed equally to this work.

    View full text