Achieving high thermoelectric figure of merit in polycrystalline SnSe

Subscriber access provided by UNIV OF DURHAM

Article

Achieving high thermoelectric figure of merit in polycrystalline SnSe via introducing Sn vacancies Wei Wei, Cheng Chang, Teng Yang, Jizi Liu, Huaichao Tang, Jian Zhang, Yusheng Li, Feng Xu, Zhidong Zhang, Jing-Feng Li, and Guodong Tang J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b11875 • Publication Date (Web): 15 Dec 2017 Downloaded from http://pubs.acs.org on December 15, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of the American Chemical Society is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Achieving high thermoelectric figure of merit in polycrystalline SnSe via introducing Sn vacancies Wei Wei,1† Cheng Chang,2† Teng Yang,3 Jizi Liu,1 Huaichao Tang,4 Jian Zhang,5 Yusheng Li,1 Feng Xu,1 Zhidong Zhang,3 Jing-Feng Li,4 Guodong Tang,1,∗

1

School of Materials Science and Engineering, Nanjing University of Science and

Technology, Nanjing 210094, China. 2

School of Material Science and Engineering, Beihang University, Beijing 10091,

China. 3

Shenyang National Laboratory for Materials Science, Institute of Metal Research,

Chinese Academy of Sciences, 72 Wenhua Road, Shenyang 110016, China. 4

State Key Laboratory of New Ceramics and Fine Processing, School of Materials

Science and Engineering, Tsinghua University, Beijing 100084, China. 5

Key Laboratory of Materials Physics, Institute of Solid State Physics, Chinese

Academy of Sciences, Hefei 230031, China.

*

Email: [email protected].

† These authors contributed equally to this work.

ACS Paragon Plus Environment 1

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT Thermoelectric power genetation technology has emerged as a clean ‘heat engine’ that can convert heat to electricity. Recently, the discovery of an ultrahigh thermoelectric figure of merit in SnSe crystals has drawn a great deal of attention. In view of their facile processing and scale-up applications, polycrystalline SnSe materials with ZT values comparable to those of the SnSe crystals are greatly desired. Here we achieve a record high ZT value ∼2.1 at 873 K in polycrystalline Sn1-xSe with Sn vacancies. We demonstrate that the carrier concentration increases by artificially introducing Sn vacancies, contributing significantly to the enhancements of electrical conductivity and thermoelectric power factor. The detailed analysis of the data in the light of first-principles calculations results indicates that the increased carrier concentration can be attributed to the Sn-vacancy-induced Fermi level downshift and the interplay between the vacancy states and valence bands. Furthermore, vacancies break translation symmetry and thus enhance phonon scattering, leading to extralow thermal conductivity. Such high ZT value ∼2.1 is achieved by synergistically optimizing both electrical- and thermal-transport properties of polycrystalline SnSe. The vast increase in ZT for polycrystalline SnSe may accelerate practical applications of this material in highly effective solid-state thermoelectric devices.


Page 2 of 24

Page 3 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Thermoelectric effects have potential applications in power generation and electronic cooling. The performance of a thermoelectric material is determined by the dimensionless figure of merit ZT, ZT = S2σT/κ, in which S, σ, κ, and T are respectively the Seebeck coefficient, electrical conductivity, thermal conductivity, and absolute temperature. Therefore, both a high power factor (S2σ) and a low thermal conductivity (κ) are desired for high thermoelectric performance materials. However, the complex interdependence of S, σ and κ makes it difficult in maximizing the ZT value. Several strategies including nanostructuring,1,2 electronic band structure engineering,3 energy filtering effect,4 and multiscale hierarchical architecturing5 have been proposed to improve ZT in the past decade. Recently, SnSe has received extensive attention owing to its excellent thermoelectric properties.6-13 It crystallizes in a highly anisotropic layered orthorhombic (Pnma group) crystal structure below 800 K and undergoes a structural transition from Pnma to Cmcm above 800 K.14-16 SnSe crystals are reported to show ultralow thermal conductivity due to giant phonon anharmonicity, contributing to a record ZT ∼ 2.6 at 923 K along the b axis.6 However, owing to demand for facile processing and scale-up applications, polycrystalline SnSe materials with ZT values comparable to those of the SnSe crystals are greatly desired.9, 17-24 To date, the high ZT values of SnSe crystals have not been reproduced in polycrystalline SnSe because polycrystalline samples show much lower carrier mobility due to their strong anisotropy, and a higher thermal conductivity because of the effects of Sn oxides.10, By carefully controlling the sample synthesis procedure, we recently demonstrated high performance ZT > 1.7 could be realized in polycrystalline SnSe through thermoelectric power factor while keeping a low lattice thermal conductivity via phase-separations and nanostructuring.9 However, compared to SnSe crystals, the thermoelectric performance in polycrystalline SnSe suggests that it is still a highly challenging to balance the nanostructuring effects on both the thermal and electrical



transport properties due to numerous interfaces. Vacancy resembles hole-doping that shifts Fermi level into the valence bands and simultaneously induces localized flat bands in the vicinity of the Fermi energy level, both of them can give rise to large electronic density of states, and thereby paves a venue for enhancing carrier concentration.28 Meanwhile, vacancies, which break translation symmetry, can act as phonon scattering center to mitigate lattice thermal conductivity. This motivates us to optimize thermoelectric transport behaviors by artificially introducing Sn vacancies in polycrystalline SnSe. Here, we report a successful strategy to achieve a high thermoelectric performance in polycrystalline Sn1-xSe. We demonstrate that the carrier concentration can be increased significantly, while low thermal conductivity is maintained via introducing Sn vacancies. The increase in carrier concentration leads to enhanced electrical conductivity and power factor. Vacancies act as efficient scattering sites to suppress the lattice thermal conductivity. As a consequence of this strategy, a record high performance with ZT value ∼2.1 is achieved in polycrystalline SnSe by synergistically optimizing its electrical and thermal transport properties. This work indicates that high thermoelectric performance could be achieved in low-cost, earth-abundant and environmentally-friendly polycrystalline SnSe. To the best of our knowledge, this is the first work that achieves ZT∼2.1 in binary polycrystalline compound without complex composition design.

EXPERIMENTAL AND COMPUTATIONAL METHOD SECTION A series of Sn1-xSe samples were synthesized using hydrothermal methods by using starting materials of SnCl2·2H2O (99%, Kelong, China) and Se powder (99.9%, Kelong, China). Here Sn vacancies were introduced by reducing the corresponding proportion of the starting raw material SnCl2·2H2O during the hydrothermal process. SnCl2·2H2O was first dissolved in deionized water after stirring for ∼10 min. Then NaOH was added into the solution and stirred for another 10 min. The Sn containing


Page 4 of 24

Page 5 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


solution was then transferred to a Teflon-lined stainless steel autoclave (100 ml capacity). After adding Se, the autoclave was sealed and then heated at 403 K for 36 h. Cooling it to room temperature, and then the black products were collected and for several times with absolute ethanol and deionized water. The powders were finally dried with a vacuum dryer at 333 K for 4 h. Spark plasma sintering (SPS) (HPD 10, FCT System GmbH) was conducted at 693 K for 7 min under a 50 MPa uniaxial pressure to obtain highly dense samples. The series of Sn vacancy-containing samples was denoted as Sample A, Sample B, and Sample C, as shown in Table 1. The powder X-ray diffraction patterns were performed using a Bruker D8 Advance instrument Cu Ka radiation and the scanning step size of 0.02°. Internal reference Si was used to calibrate the diffraction angles of Sn1-xSe compounds. Scanning electron microscppy (SEM) was performed by a FEI Quanta 250FEG instrument and the elements were performed by the inside EDS. Transmission electron microscopy (TEM) investigations were conducted with a FEI Tecnai 20 microscope operating at 200 kV accelerating voltage. Atomic-resolution high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) observations were conducted in aberration-corrected scanning transmission electron microscope (FEI Titan G2 60-300) operated at 300 kV. The samples were prepared in a focused ion-beam machine. Measurement of both the electrical and thermal transport properties of all samples was performed along the direction parallel to the SPS pressure direction. The Seebeck coefficient and electrical conductivity were measured on a commercial (Ulvac-Riko ZEM-3). The thermal diffusivity coefficient D was carried out on a laser flash apparatus (Netzsch LFA-457 instrument). The specific heat Cp was measured by differential scanning calorimetry. The density ρ was measured by the Archimedes method (Table S1). The thermal conductivity was thereby calculated according to the relation κ = DCpρ. The temperature-dependent Hall coefficient (RH) was measured by the van der Pauw method using a Hall measurement system (ResiTest 8340DC, Toyo, Japan). The carrier concentration (n) were calculated by using n = 1/eRH, and mobility



Page 6 of 24

(µ) were obtained by the relationship of µ = RH/ρ, The uncertainties for the electrical conductivity and the Seebeck coefficient measurements are both equal to 5%. The uncertainty for the total thermal conductivity is about 12% (comprising uncertainties 5% for the specific heat, 2% for the density, and 5% for the thermal diffusivity). The combined uncertainty for determination of ZT is approximately 20%. The structures with vacancies are constructed based on the 2× 2× 2 supercell. We calculated the electronic band structure by using ab. initio. density functional theory (DFT)

as

implemented

in

the

Quantum

Espresso

package.

The

Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional29 within the general gradient approximation (GGA) and PAW potentials were used. The Brillouin zone (BZ) of the 2× 2× 2 SnSe supercell was sampled by the k mesh of 3× 3× 3 Monkhorst-Pack grid30 for total energy calculations. The electronic kinetic energy cutoff of 40 Ry for the plane-wave basis was taken and the total energy difference of no more than 10−10 eV between two consecutive self-consistency iterations as the convergence criterion. The vacancy fractions (number of Sn vacancies divided by the number of Sn atoms in the corresponding unit cell of pristine SnSe) of 3.13% (one vacancy per 32 Sn atoms) is chosen for the calculation convenience to approximately represent the experimental 5% Sn vacancy content.

RESULTS AND DISCUSSION Fig. 1a shows X-ray diffraction (XRD) patterns of Sn1-xSe samples. All samples show the single-phase Pnma orthorhombic structure without impurity peaks. To clearly show the effect of Sn vacancies on the lattice, we perform careful XRD measurements by focusing on a narrow angular range. It can be seen from Fig. 1b that the (110), (103) and (111) diffraction peaks shift from a low angle to a high angle with the increase of Sn vacancies, revealing that lattice parameters are reduced by the introduction of Sn vacancies. Rietveld refinement of the XRD pattern was performed. The lattice parameters extracted from XRD refinement demonstrate that the lattice parameters decrease with increasing Sn vacancy content (see Table S2). Chemical compositions analysis by electron probe micro analysis (EPMA) indicate the existence of a large number of Sn vacancies in these samples. To designate the samples, we use the chemical compositions derived from EPMA (see Table 1).


Page 7 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


To provide the evidence of Sn vacancies, the microstructure analysis of the Sn1-xSe samples are performed using atomic-resolution high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM). HAADF-STEM images obtained for the Sn0.95Se sample are presented as a typical example. As shown in Figs. 2a and 2b, many nanodomains (∼5 nm) are submerged in the matrix. HAADF-STEM image (Fig. 2b) reveals that there is no clear boundary between the matrix and nanodomain. The contrast gradually changes from light to dark crossing the diffuse boundary from the matrix to the interior of the nanodomain while the crystal structure is not changed. The intensity fluctuations indicate that high density of vacancies exist in the Sn0.95Se material.31,32 It is obvious that the large quantity of vacancies tend to aggregate at many small local areas, creating abundant dark nanodomains (Figs. 2a and 2b) dispersed within grains. Atomic-resolution HAADF-STEM and quantitative analysis were employed to detect Sn vacancies. The contrast of HAADF-STEM image is proportionally relevant to the atomic number Z1.7~2 and the specimen thickness.31,32 If the composition and specimen thickness are the same, the contrast depends on the atomic number in the column. If there are less atoms in a column or some atoms are replaced by vacancy, the contrast of this column should be weaker. Fig. 2c presents high-quality experimental HAADF-STEM image viewed along the b-axis of Sn0.95Se, enabling the examination of the individual atomic columns. Owing to the Z contrast, Sn and Se atomic columns can be directly distinguished by their brightness (Sn atomic columns are brighter than Se atomic columns).13 The alternating brightness is shown quantitatively by the intensity trace (Fig. 2d) in Fig. 2c. At the positions indicated by arrows, the Sn-atom contrast is weaker than that in the neighboring Sn position, demonstrating the presence of Sn vacancies in the atomic column parallel to the electron beam.13,33,34 However, no Se-vacancy is detected in the intensity profile (Fig. 2d). Polycrystalline SnSe sample shows anisotropic transport properties because of layered structure of SnSe and the uniaxial pressure of SPS process. Previous studies suggested that a higher ZT is obtained along the pressing direction.9,24 Therefore, in this work, the thermoelectric properties of polycrystalline Sn1-xSe are investigated along the pressing direction. Figure 3a shows the temperature dependence of electrical conductivity (σ) for



Sn1-xSe samples. For comparison, the data for polycrystalline SnSe + 1% PbSe are also listed.9 It is observed that for all the samples,σshows the same temperature-dependence trend. Semiconducting transport behavior is first observed from 300 K to 500 K but then switches to metallic-like behavior from 500 K up to 650 K. Further heating leads to an increases of σ until 850 K which can be ascribed to the thermal carrier excitations.6,9 Obviously, the samples with Sn vacancies exhibit a significantly enhanced σ compared to that of pure SnSe and polycrystalline SnSe + 1% PbSe. σ for Sn0.95Se reaches the maximum value of 68 S·cm-1 at 873 K. High-temperature Hall measurements are performed to further investigate the electrical transport behavior in Sn1-xSe samples. The temperature dependence of carrier concentration (n) and carrier mobility (µ) from 300 to 650 K are shown in Figs. 3b and 3c. Compared to the pure SnSe, n is greatly enhanced by the introduction of Sn vacancies (Fig. 3b). At 300 K, n of all Sn1-xSe samples is on the order of 1019 cm-3. For example, the room-temperature n of Sn0.95Se reaches as high as 1.07 × 1019 cm-3, which is over one order of magnitude larger than that of reported polycrystalline SnSe (~4×1017 holes cm−3)12 and crystals (~4×1017 holes cm−3).6 Furthermore, this value is even about two times bigger than the high-performance polycrystalline SnSe + 1% PbSe (6.1 × 1018 cm-3).9 In Sn1-xSe samples, Sn vacancies can be considered as acceptors that increase the hole carrier concentration and then lead to the high electrical conductivity. Sn vacancies should lead to a remarkable increase of hole concentration, however, the measured carrier concentration is much lower than expected. Elemental mapping results indicate existence of Se-precipitates (See Fig. S2). Precipitation of Se leads to a significantly lower Sn vacancies and carrier concentration. The temperature dependence of µ is presented in Fig. 3c. It is observed that Sn1-xSe samples show reduced µ compared to that of pure SnSe because of enhanced point defect scattering induced by Sn vacancies. µ decreases further with increasing Sn vacancies. Furthermore, µ of Sn1-xSe samples exhibits a stronger increasing tendency with temperature from 300 to 500 K than that of pure SnSe. A


Page 8 of 24

Page 9 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


barrier-like scattering that arises from the defects induced by Sn vacancies results in the rapid growing µ at low temperatures.19 The electronic band structures of SnSe are calculated based on ab initio density functional theory. The atomic structures with the vacancy are given in the supplementary material Fig. S3. We constructed the atomic structure of Sn-vacancies-containing SnSe based on the minimization of thermodynamic Gibbs free energy by maximizing the vibrational entropy due to vacancies and optimized atomic coordinations by using conjugate gradient method. To avoid the artificial effects due to dangling bonds and defect states, we relax the atomic structure until all the atomic forces are smaller than 0.001 eV/Ang. Fig. 4 shows the electronic band structure of the samples with Sn vacancies compared to that of pure SnSe. It reveals that the pure SnSe is a semiconductor with an indirect band gap, consistent with the previous calculations.6 With Sn vacancies, SnSe becomes a p-type semiconductor with a high level of hole doping. A large value of electronic density of states in the Fermi energy level is found, giving rise to the increase of n. This result is consistent with the vacancy-induced increase of carrier concentration shown in Fig. 3b. Electronic band gap around 0.6 eV remains from 0% to 5% of Sn vacancies above the valence band, and this value agrees with the band gap value extracted from the temperature dependence of the experimental σ (~ e-Eg/kBT, Eg is the band gap) between 700 K and 823 K in Fig. 3a. Worth pointing out is that at 5% Sn vacancy (in blue in Fig. 4) there is a flat defect band ( < 0.1 eV above EF along ΓX and partly along ΓM, which is equivalent to temperature below 1000K), suggesting that it is likely to observe a change on temperature dependence of electrical transport properties below 1000 K. Indeed, as observed in our experiment as shown in Fig. 3a, σ increases firstly with temperature below 450 K and then decreases with temperature till 650 K, and finally increases again with temperature from 650 K. This changing behavior of σ can be understood from the calculated band structure of SnSe with Sn vacancy. Electron is thermally excited from valence bands to the acceptor defect band, showing a



temperature dependence of semiconducting σ. As the acceptor band is saturated with temperature, σ shows a metallic behavior due to phonon scattering. With temperature further increasing, semiconducting σappears again due to thermal excitation of carriers over intrinsic band gap. It is encouraging to see that the electrical transport properties derived from the calculated band structure are consistent with the experimental measurements. Fig. 5a presents the temperature dependence of Seebeck coefficient (S) for Sn1-xSe sample. The positive S suggests p-type conducting and reaches the maximum value at around 650 K. The peak of S indicates onset of bipolar conduction. S decrease as the temperature enters the intrinsic range of these materials. Minority carriers (electrons) are then produced owing to heat-excited holes transferring from valence band to conduction band, adversely affecting S in the intrinsic temperature range. The increase of S at T > 823 K is a signature of the phase transition from Pnma to Cmcm phase.9,14,15 It is shown that S reduces with the increase of Sn vacancies and all Sn vacancies containing samples have lower S values than that of pure SnSe, which can be attributed to the enhanced n confirmed by Hall measurements. It is worth noting that S values in Sn1-xSe samples are larger than that of SnSe + 1% PbSe.9 The temperature dependence of power factor (PF) is indicated in Fig. 5b. Upon the introduction of Sn vacancies, the PF is significantly enhanced due to the quite high σ. The maximum PF value of 7.77 µWcm-1K-2 is obtained at 873 K in the Sn0.95Se sample, which is more than two times larger than that of pure SnSe. Compared to SnSe + 1% PbSe, the samples with Sn vacancies exhibit considerably large PF due to enhanced σ and S. Figure 5c shows the variation of total thermal conductivity (κT) of Sn1-xSe samples as a function of temperature. A general trend found for these materials is that

κT drops with increasing temperature. Low κT is achieved in all samples with Sn vacancies, which is significantly suppressed compared to that of other reported Ag-doped17, Zn-doped22 and alkali-ion doped SnSe materials.19 The lattice thermal


Page 10 of 24

Page 11 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


conductivity (κL) is calculated based on κL= κT − κe, in which the carrier thermal conductivity component κe is estimated based on the Wiedemann–Franz law (κe = LTσ). The Lorenz number was derived from Fermi energy with a simple one-band model and only acoustic phonon scattering considered, as illustrated in supplementary Fig. S4.35,36 The results shown in Fig. 5d suggest that the phonon contribution prevails over the electronic part. It is important to note that κL of all vacancy-containing samples is depressed compared to that of pure SnSe. The suppression of κT can mainly be attributed to enhanced phonon scattering. A very recent study reveals that the ultralow thermal conductivity in reported SnSe crystals can be partly attributed to Sn vacancies and Se interstitials.13 In our samples, Sn vacancies, which introduce strong phonon scattering by both missing atoms and missing interatomic linkages, significantly diminish lattice thermal conductivity.37 In the meanwhile, Se secondary phase further suppress the κL due to enhanced interface scattering. Therefore, a significant reduction of κL is obtained in polycrystalline Sn1-xSe. Figure 6 presents the temperature dependence of ZT for Sn1-xSe samples. It is found that ZT increases with increasing temperature and is significantly enhanced as compared with pure SnSe and SnSe + 1% PbSe. An extra-high ZT ∼ 2.1 at 873 K is achieved in a binary compound Sn0.95Se. To the best of our knowledge, this is a record ZT value in bulk SnSe polycrystals that outperforms most of state-of-the-art p-type thermoelectric materials.2,9,38-40 In Sn1-xSe samples, Sn vacancies further increase phonon scattering, significantly suppressing lattice thermal conductivity. More importantly, Sn vacancies induce significantly enhanced σ and PF due to the increased n. As a result, a high ZT value is achieved by synergistically optimizing electrical- and thermal-transport properties by introducing Sn vacancies. Moreover, we find good experimental repeatability for this high ZT, which is evidenced by the reproducible results from the measurements on several samples independently prepared (Fig. S5).



CONCLUDING REMARKS By facile and cost-effective hydrothermal synthesis method, we have succeeded in fabricating high-performance polycrystalline SnSe. It is found that when Sn vacancies are introduced into the SnSe matrix, both σ and PF are significantly enhanced in polycrystalline Sn1-xSe samples. Meanwhile, Sn vacancies lead to a significant suppression of lattice thermal conductivity due to the strong phonon scattering caused both by missing atoms and missing interatomic linkages. With low thermal conductivity and highly enhanced PF, high thermoelectric performance is realized with ZT ∼ 2.1, which is a record ZT for polycrystalline SnSe. This work indicates that artificially introducing vacancies is a promising strategy to boost thermoelectric performance through simultaneous optimization of electrical and thermal transport properties.

ASSOCIATED CONTENT Supporting information Lorenz number calculation in details, Sample density (Table S1), Lattice parameters (Table S2), Low-magnification TEM image (Figure S1), Atomic structure of SnSe with 5% and 10.5% Sn vacancy (Figure S2), Lorenz number (Figure S3), The reproducibility of thermoelectric properties (Figure S4). This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION *Corresponding Authors: [email protected].

ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (No. U1732153, 51571007 and 51601094), Natural Science Foundation of Jiangsu Province (No. BK20161495), the Priority Academic Program Development of Jiangsu


Page 12 of 24

Page 13 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Higher Education Institutions, and the Fundamental Research Funds for the Central Universities (No. 30917011206). HAADF-STEM, TEM and SEM experiments were performed at the Materials Characterization and Research Center of Nanjing University of Science and Technology.

REFERENCES



(1) Hsu, K. F.; Loo, S.; Guo, F.; Chen, W.; Dyck, J. S.; Uher, C.; Hogan, T.; Polychroniadis, E. K.; Kanatzidis, M. G. Science 2004, 303, 818. (2) Poudel, B.; Hao, Q.; Ma, Y.; Lan, Y.; Minnich, A.; Yu, B.; Yan, X.; Wang, D.; Muto, A.; Vashaee, D.; Chen, X.; Liu, J.; Dresselhaus, MS.; Chen, G.; Ren, Z. Science 2008, 320, 634. (3) Pei, Y. Z.; Shi, X.; Lalonde, A.; Wang, H.; Chen, L.; Snyder, G. J. Nature 2011, 473, 66. (4) Heremans, J. P.; Thrush, C. M.; Morelli, D. T. Phys. Rev. B 2004, 70, 115334. (5) Biswas, K.; He, J. Q.; Blum, I. D.; Wu, C. I.; Hogan, T. P.; Seidman, D. N.; Dravid, V. P.; Kanatzidis, M. G. Nature 2012, 489, 414. (6) Zhao, L. D.; Lo, S. H.; Zhang, Y. S.; Sun, H.; Tan, G. J.; Uher, C.; Wolverton, C.; Dravid, V. P.; Kanatzidis, M. G. Nature 2014, 508, 373. (7) Zhao, L. D.; Tan, G.; Hao, S.; He, J.; Pei, Y.; Chi, H.; Gong, S. K.; Xu, H. B.; Dravid, V. P.; Uher, C.; Snyder, G. J.; Wolverton, C.; Kanatzidis, M. G. Science 2016, 351, 141. (8) Peng, K. L.; Lu, X.; Zhan, H.; Hui, S.; Tang, X.; Wang, G.; Dai, J.; Uher, C.; Wang, G. Y.; Zhou, X. Y. Energy Environ. Sci. 2016, 9, 454. (9) Tang, G. D.; Wei, W.; Zhang, J.; Li, Y. S.; Wang, X.; Xu, G. Z.; Chang, C.; Wang, Z. H.; Du, Y. W.; Zhao, L. D. J. Am. Chem. Soc. 2016, 138, 13647. (10) Zhao, L. D.; Chang, C.; Tan, G.; Kanatzidis, M. G. Energy Environ. Sci. 2016，9, 3044. (11) Carrete, J.; Mingo, N.; Curtarolo, S. Appl. Phys. Lett. 2014, 105, 101907. (12) Sassi, S.; Candolfi, C.; Vaney, J. B.; Ohorodniichuk, V.; Masschelein, P.; Dauscher, A.; Lenoir, B. Appl. Phys. Lett. 2014, 104, 212105. (13) Wu, D.; He, D. S.; Wu, L. J.; Zhao, L. D.; Li, W.; Wu, M. H.; Huang, L.; Zhu, Y. M.; Kanatzidis, M. G.; He, J. Q. Nano Energy 2017, 35, 321. (14) Chattopadhyay, T.; Pannetier, J.; Vonschnering, H. G. J. Phys. Chem. Solids 1986, 47, 879.


Page 14 of 24

Page 15 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(15) Peters, M. J.; Mcneil, L. E. Phys. Rev. B 1990, 41, 5893. (16) Asfandiyar; Wei, T. R.; Li, Z. L.; Sun, F. H.; Pan, Y.; Wu, C. F.; Farooq, M. U.; Tang, H. C.; Li, F.; Li, B.; Li, J. F. Sci. Rep. 2017, 7, 43262. (17) Chen, C.; Wang, H.; Chen, Y. Y.; Day, T.; Snyder, G. J. J. Mater. Chem. A 2014, 2, 11171. (18) Zhang, Q.; Chere, E. K.; Sun, J. Y.; Cao, F.; Dahal, K.; Chen, S.; Chen, G.; Ren, Z. F. Adv. Energy Mater. 2015, 5, 1500360. (19) Wei, T. R.; Tan, G. J.; Wu, C. F.; Chang, C.; Zhao, L. D.; Li, J. F.; Snyder, G. J.; Kanatzidis, M. G. J. Am. Chem. Soc. 2016, 138, 8875. (20) Li, Y. L.; Shi, X.; Ren, D.; Chen, J.; Chen, L. Energies 2015, 8, 6275. (21) Leng, H. Q.; Zhou, M.; Zhao, J.; Han, Y. M.; Li, L. F. RSC Adv. 2016, 6, 9112. (22) Li, J. C.; Li, D.; Qin, X. Y.; Zhang, J. Scripta Mater. 2017, 126, 6. (23) Chang, C.; Tan, Q.; Pei, Y. L.; Xiao, Y.; Zhang, X.; Chen, Y. X.; Zheng, L.; Gong, S. K.; Li, J. F.; He, J. Q.; Zhao, L. D. RSC Adv. 2016, 6, 98216. (24) Tang, G. D.; Wen, Q.; Yang, T.; Cao, Y.; Wei, W.; Wang, Z. H.; Zhang, Z. D.; Li, Y. S. RSC Adv. 2017, 7, 8258. (25) Zhang, B.; Peng, K.; Sha, X.; Li, A.; Zhou, X.; Chen, Y.; Deng, Q.; Yang, D.; Ma, E.; Han, X. Microsc. Microanal. 2017, 23, 173. (26) Tan, G.; Zhao, L. D.; Kanatzidis, M. G. Chem. Rev. 2016, 116, 12123. (27) Chen, Y. X.; Ge, Z. H.; Yin, M. J.; Feng, D.; Huang, X. Q.; Zhao, W. Y.; He, J. Q. Adv. Funct. Mater. 2016, 26, 6836. (28) Kaxiras, E. “Atomic and electronic structure of solids.” Cambridge University Press, Cambridge, England 2003. (29) Perdew, J. P.; Zunger, A. Phys. Rev. B 1981, 23, 5048. (30) Monkhorst, H. J.; Pack, J. D. Phys. Rev. B 1976, 13, 5188. (31) Hillyard, S.; Silcox, J. Ultramicroscopy 1995, 58, 6. (32) Nellist, P. D.; Pennycook, S. J. Ultramicroscopy 1999, 78, 111. (33) Jia, C. L.; Lentzen, M.; Urban, K. Science 2003, 299, 870.



(34) Jiao, X. C.; Chen, Z. W.; Li, X. D.; Sun, Y. F.; Gao, S.; Yan, W. S.; Wang, C. M.; Zhang, Q.; Lin, Y.; Luo, Y.; Xie, Y. J. Am. Chem. Soc. 2017, 139, 7586. (35) Zhao, L. D.; Lo, S. H.; He, J.; Li, H.; Biswas, K.; Andorulakis, J.; Wu, C. I.; Hogan, T. P.; Chung, D. Y.; Dravid, V. P.; Kanatzidis, M. G. J. Am. Chem. Soc. 2011, 133, 20476. (36) Liu, Y.; Zhao, L. D.; Liu, Y. C.; Lan, J. L.; Xu, W.; Li, F.; Zhang, B. P.; Berardan, D.; Dragoe, N.; Lin, Y. H.; Nan, C. W.; Li, J. F.; Zhu, H. M. J. Am. Chem. Soc. 2011, 133, 20112. (37) Li, Z.; Xiao, C.; Fan, S.; Deng, Y.; Zhang, W.; Ye, B.; Xie, Y. J. Am. Chem. Soc., 2015, 137, 6587. (38) Zhao, H. Z.; Sui, J. H.; Tang, Z. J.; Lan, Y. C.; Jie, Q.; Kraemer, D.; McEnaney, K.; Guloy, A.; Chen, G.; Ren, Z. F. Nano Energy 2014, 7, 97. (39) Kim, S. I.; Lee, K. H.; Mun, H. A.; Kim, H. S.; Hwang, S. W.; Roh, J. W.; Yang, D. J.; Shin, W. H.; Li, X. S.; Lee, Y. H.; Snyder, G. J.; Kim, S. W. Science 2015, 348, 109. (40) Liu, Y.; Zhao, L. D.; Zhu, Y. C.; Liu, Y. C.; Li, F.; Yu, M. J.; Liu, D. B.; Xu, W.; Lin, Y. H.; Nan, C. W. Adv. Energy Mater. 2016, 6, 1502423.


Page 16 of 24

Page 17 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 1: (a) XRD patterns of polycrystalline Sn1-xSe samples, standard peaks of SnSe were given for comparison. (b) XRD patterns measured by focusing on a narrow angular range. Si powders were used as a reference.



Fig. 2: HAADF-STEM images of Sn0.95Se along the b-axis: image of grain interior; (b) vacancy domains;

Page 18 of 24

(a) low-magnification

(c) atomic-resolution image showing

the projection of atomic columns; (d) intensity line sanning profine along the green boxed atomic layer, taken directly from the digital readout of the detector, The arrows indicate Sn poor column.


Page 19 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 3: Temperature dependence of (a) electrical conductivity (σ), (b) carrier concentration (n), (c) carrier mobility (µ) for polycrystalline Sn1-xSe.



Fig. 4: Electronic band structure of pure SnSe (black) compared to that of SnSe with 5% (blue) Sn vacancy. Brillouine zone with high-symmetry points is shown in the bottom.


Page 20 of 24

Page 21 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 5: Temperature dependence of (a) Seebeck coefficient (S), (b) power factor (PF), c) total thermal conductivity (κT), (d) lattice thermal conductivity (κL) for polycrystalline Sn1-xSe. Data for high performance phase-separated polycrystalline SnSe + 1% PbSe9 is also shown for comparison.



Fig. 6: Temperature dependence of ZT for polycrystalline Sn1-xSe.


Page 22 of 24

Page 23 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 1: Chemical compositions as determined by EPMA for Sn vacancies containing Sn1-xSe samples.

Sample A

Sample B

Sample C

Sn0.95Se

Sn0.925Se

Sn0.895Se



TOC


Page 24 of 24

Achieving high thermoelectric figure of merit in polycrystalline SnSe

Recommend Documents