FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2912055813> ?p ?o ?g. }

• W2912055813 endingPage "731" @default.
• W2912055813 startingPage "719" @default.
• W2912055813 abstract "•A H2 reduction process removes tin oxide films that invariably cover SnSe grains•Ultralow thermal conductivity innate in SnSe accessed in polycrystalline materials•ZT ∼2.5 of polycrystalline SnSe matches the record-high value of the single crystals Well prepared and properly handled single-crystal SnSe samples exhibit exceptional thermoelectric performance with a record-high thermoelectric figure of merit (ZT) of ∼2.5–2.7 due to the ultralow thermal conductivity. Achieving comparable thermoelectric performance to single crystals in polycrystalline SnSe samples has been the widely sought-after goal. These properties, however, have never been achievable in the polycrystalline samples. Particularly, serious debates have arisen on markedly higher apparent thermal conductivity for the latter. Here, we show that the underperformance of polycrystalline SnSe is mainly due to the presence of tin oxides in the sample. Our H2 reduction process removes the majority of tin oxide films that ubiquitously cover SnSe grains and eventually unveils the ultralow thermal conductivity intrinsic in this material. It also simultaneously enhances the hole mobility, electrical conductivity, and Seebeck coefficient, leading to an extraordinarily high ZT of ∼2.5 at 773 K. Tin selenide (SnSe) has emerged as a surprising new material with exceptional thermal transport and charge transport properties such as ultralow thermal conductivity, which give it a record-high thermoelectric figure of merit (ZT) of ∼2.5–2.7 at around 800 K. These properties, however, have been only observable in well-prepared and properly handled single-crystal samples. Polycrystalline SnSe samples have markedly inferior properties paradoxically with higher apparent thermal conductivity and much lower ZT values than single crystals. The high thermal conductivity in polycrystalline samples has been attributed to surface tin oxides. Based on this hypothesis, we have employed an oxide-removing strategy that involves a chemical reduction process at 613 K under a 4% H2/Ar atmosphere. This leads to an exceptionally low lattice thermal conductivity of ∼0.11 W m−1K−1 in polycrystalline hole-doped SnSe alloyed with 5% lead selenide, even lower than that of single crystals, and boosts the ZT to ∼2.5 at 773 K. Tin selenide (SnSe) has emerged as a surprising new material with exceptional thermal transport and charge transport properties such as ultralow thermal conductivity, which give it a record-high thermoelectric figure of merit (ZT) of ∼2.5–2.7 at around 800 K. These properties, however, have been only observable in well-prepared and properly handled single-crystal samples. Polycrystalline SnSe samples have markedly inferior properties paradoxically with higher apparent thermal conductivity and much lower ZT values than single crystals. The high thermal conductivity in polycrystalline samples has been attributed to surface tin oxides. Based on this hypothesis, we have employed an oxide-removing strategy that involves a chemical reduction process at 613 K under a 4% H2/Ar atmosphere. This leads to an exceptionally low lattice thermal conductivity of ∼0.11 W m−1K−1 in polycrystalline hole-doped SnSe alloyed with 5% lead selenide, even lower than that of single crystals, and boosts the ZT to ∼2.5 at 773 K. Harnessing the ubiquitous and huge amounts of waste heat energy, which constitute more than 70% of total energy produced, can contribute to better energy efficiency and environmental management.1Gingerich D.B. Mauter M.S. Quantity, quality, and availability of waste heat from united states thermal power generation.Environ. Sci. Technol. 2015; 49: 8297-8306Crossref PubMed Scopus (113) Google Scholar Thermoelectric (TE) technology based on semiconductors can be a clean and sustainable choice for such purposes.2Vineis C.J. Shakouri A. Majumdar A. Kanatzidis M.G. Nanostructured thermoelectrics: big efficiency gains from small features.Adv. Mater. 2010; 22: 3970-3980Crossref PubMed Scopus (1166) Google Scholar, 3Kanatzidis M.G. Discovery-synthesis, design, and prediction of chalcogenide phases.Inorg. Chem. 2017; 56: 3158-3173Crossref PubMed Scopus (121) Google Scholar, 4Xianli S. Ping W. Han L. Wei L. Yonggao Y. Peng L. Chuqi S. Changjun X. Wenyu Z. Pengcheng Z. et al.Thermoelectric materials: multi-scale microstructural thermoelectric materials: transport behavior, non-equilibrium preparation, and applications.Adv. Mater. 2017; 29: 1602013Crossref Scopus (211) Google Scholar, 5Lei Y. Zhi-Gang C. Dargush M.S. Jin Z. High performance thermoelectric materials: progress and their applications.Adv. Energy Mater. 2018; 8: 1701797Crossref Scopus (505) Google Scholar When a temperature gradient is applied, TE semiconductors create an electrical potential difference on account of the Seebeck effect. The performance of TE materials is typically expressed by a figure of merit (ZT) = S2σT/κtot,6Tan G. Zhao L.-D. Kanatzidis M.G. Rationally designing high-performance bulk thermoelectric materials.Chem. Rev. 2016; 116: 12123-12149Crossref PubMed Scopus (1305) Google Scholar where S is the Seebeck coefficient, σ is the electrical conductivity, T is the absolute temperature, and κtot is the total thermal conductivity that is the sum of the electrical (κele) and lattice thermal conductivity (κlat). The search for increasingly efficient TE materials in the past decade has been intense, systematic, and multipronged and has succeeded in identifying several new materials with unprecedented performance. Representative examples are PbTe involving endotaxial nanostructures2Vineis C.J. Shakouri A. Majumdar A. Kanatzidis M.G. Nanostructured thermoelectrics: big efficiency gains from small features.Adv. Mater. 2010; 22: 3970-3980Crossref PubMed Scopus (1166) Google Scholar, 7Kanatzidis M.G. Nanostructured thermoelectrics: the new paradigm?†.Chem. Mater. 2010; 22: 648-659Crossref Scopus (945) Google Scholar, 8Lee Y. Lo S.H. Chen C. Sun H. Chung D.Y. Chasapis T.C. Uher C. Dravid V.P. Kanatzidis M.G. Contrasting role of antimony and bismuth dopants on the thermoelectric performance of lead selenide.Nat. Commun. 2014; 5: 3640Crossref PubMed Scopus (91) Google Scholar, 9Park K. Ahn K. Cha J. Lee S. Chae S.I. Cho S.P. Ryee S. Im J. Lee J. Park S.D. et al.Extraordinary off-stoichiometric bismuth telluride for enhanced n-type thermoelectric power factor.J. Am. Chem. Soc. 2016; 138: 14458-14468Crossref PubMed Scopus (71) Google Scholar, 10Lee Y.K. Ahn K. Cha J. Zhou C. Kim H.S. Choi G. Chae S.I. Park J.H. Cho S.P. Park S.H. et al.Enhancing p-type thermoelectric performances of polycrystalline SnSe via tuning phase transition temperature.J. Am. Chem. Soc. 2017; 139: 10887-10896Crossref PubMed Scopus (91) Google Scholar or hierarchical architecture,11Girard S.N. He J. Zhou X. Shoemaker D. Jaworski C.M. Uher C. Dravid V.P. Heremans J.P. Kanatzidis M.G. High performance Na-doped PbTe–PbS thermoelectric materials: electronic density of states modification and shape-controlled nanostructures.J. Am. Chem. Soc. 2011; 133: 16588-16597Crossref PubMed Scopus (298) Google Scholar, 12Biswas K. He J. Blum I.D. Wu C.I. Hogan T.P. Seidman D.N. Dravid V.P. Kanatzidis M.G. High-performance bulk thermoelectrics with all-scale hierarchical architectures.Nature. 2012; 489: 414-418Crossref PubMed Scopus (3279) Google Scholar, 13Zhao L.-D. Dravid V.P. Kanatzidis M.G. The panoscopic approach to high performance thermoelectrics.Energy Environ. Sci. 2014; 7: 251-268Crossref Google Scholar, 14Korkosz R.J. Chasapis T.C. Lo S.H. Doak J.W. Kim Y.J. Wu C.I. Hatzikraniotis E. Hogan T.P. Seidman D.N. Wolverton C. et al.High ZT in p-type (PbTe)1–2x (PbSe)x(PbS)x thermoelectric materials.J. Am. Chem. Soc. 2014; 136: 3225-3237Crossref PubMed Scopus (201) Google Scholar multiple-filled skutterudites,15Shi X. Yang J. Salvador J.R. Chi M. Cho J.Y. Wang H. Bai S. Yang J. Zhang W. Chen L. Multiple-filled skutterudites: high thermoelectric figure of merit through separately optimizing electrical and thermal transports.J. Am. Chem. Soc. 2011; 133: 7837-7846Crossref PubMed Scopus (1142) Google Scholar, 16Tang Y. Gibbs Z.M. Agapito L.A. Li G. Kim H.S. Nardelli M.B. Curtarolo S. Snyder G.J. Convergence of multi-valley bands as the electronic origin of high thermoelectric performance in CoSb3 skutterudites.Nat. Mater. 2015; 14: 1223-1228Crossref PubMed Scopus (486) Google Scholar, 17Zhao W. Liu Z. Sun Z. Zhang Q. Wei P. Mu X. Zhou H. Li C. Ma S. He D. et al.Superparamagnetic enhancement of thermoelectric performance.Nature. 2017; 549: 247-251Crossref PubMed Scopus (369) Google Scholar and half-Heusler compounds.18Fu C. Bai S. Liu Y. Tang Y. Chen L. Zhao X. Zhu T. Realizing high figure of merit in heavy-band p-type half-Heusler thermoelectric materials.Nat. Commun. 2015; 6: 8144Crossref PubMed Scopus (741) Google Scholar, 19Zeier W.G. Schmitt J. Hautier G. Aydemir U. Gibbs Z.M. Felser C. Snyder G.J. Engineering half-Heusler thermoelectric materials using Zintl chemistry.Nat. Rev. Mater. 2016; 1: 16032Crossref Scopus (257) Google Scholar One remarkable discovery, however, is tin selenide (SnSe), which consists of Earth’s-crust-abundant elements. Previously ignored, when grown as a single crystal, pristine p-type SnSe shows a ZT of 2.6 at 923 K along the b axis,20Zhao L.D. Lo S.H. Zhang Y. Sun H. Tan G. Uher C. Wolverton C. Dravid V.P. Kanatzidis M.G. Ultralow thermal conductivity and high thermoelectric figure of merit in SnSe crystals.Nature. 2014; 508: 373-377Crossref PubMed Scopus (3398) Google Scholar and Br-doped n-type crystals exhibit a ZT of 2.8 at 773 K along the a axis.21Chang C. Wu M. He D. Pei Y. Wu C.F. Wu X. Yu H. Zhu F. Wang K. Chen Y. et al.3D charge and 2D phonon transports leading to high out-of-plane ZT in n-type SnSe crystals.Science. 2018; 360: 778-783Crossref PubMed Scopus (673) Google Scholar Such record-high ZT values are attributed to intrinsic ultralow κlat (∼0.20 W m−1K−1), which derives from strong anharmonic and anisotropic behavior in the structure. Arguably, the lamellar and easily cleavable SnSe single crystals are delicate to handle and unsuitable for mass production and broad applications of this technology. They are brittle, and growing them is a time-consuming, labor-intensive, and expensive process. Because of the highly anisotropic layered structure, SnSe single crystals are highly prone to mechanical cracks.22Zhao L.-D. Chang C. Tan G. Kanatzidis M.G. SnSe: a remarkable new thermoelectric material.Energy Environ. Sci. 2016; 9: 3044-3060Crossref Google Scholar On those accounts, it is imperative to realize polycrystalline SnSe-based materials with comparable ZT to the single crystals. However, serious debates have arisen on its ultralow κlat because many reports on polycrystalline SnSe described higher κlat values than the corresponding single crystals,23Wei P.C. Bhattacharya S. He J. Neeleshwar S. Podila R. Chen Y.Y. Rao A.M. The intrinsic thermal conductivity of SnSe.Nature. 2016; 539: E1-E2Crossref PubMed Scopus (119) Google Scholar leading to much lower ZTs. These results are strikingly contrary to the general understanding, given that polycrystalline samples tend to exhibit lower κlat values than single crystals by virtue of additional phonon scattering at grain boundaries. Two mechanisms were evoked for this anomalous observation.23Wei P.C. Bhattacharya S. He J. Neeleshwar S. Podila R. Chen Y.Y. Rao A.M. The intrinsic thermal conductivity of SnSe.Nature. 2016; 539: E1-E2Crossref PubMed Scopus (119) Google Scholar First, exposure of SnSe powders to air could readily oxidize their surface to form a layer of SnO and SnO2, being very high thermal conductivity materials and thus obscuring the intrinsically ultralow κlat.22Zhao L.-D. Chang C. Tan G. Kanatzidis M.G. SnSe: a remarkable new thermoelectric material.Energy Environ. Sci. 2016; 9: 3044-3060Crossref Google Scholar Second, vast off-stoichiometric defects were recently observed in SnSe single crystals with ultralow κlat by aberration-corrected transmission electron microscopy (TEM).24Wu D. Wu L. He D. Zhao L.-D. Li W. Wu M. Jin M. Xu J. Jiang J. Huang L. et al.Direct observation of vast off-stoichiometric defects in single crystalline SnSe.Nano Energy. 2017; 35: 321-330Crossref Scopus (89) Google Scholar We hypothesized that if SnO and SnO2 were the main reasons for the higher apparent thermal conductivity of the polycrystalline samples, then removing these oxides from the surface and protecting the resulting samples from re-oxidation should not only reveal the true intrinsic ultralow thermal conductivity in polycrystalline SnSe but also (because of the added scattering from the grain boundaries) result in even lower thermal conductivity than single crystals. Here, we report that a combined process of ball milling and chemical reduction efficiently removes the tin oxide layers from the surface of powders of SnSe-based materials, indeed revealing their intrinsically ultralow κlat and achieving an exceptionally high ZT comparable to that of the single crystals. Samples of composition SnSe-5%PbSe doped p-type with 1 mol % Na, namely Na0.01(Sn0.95Pb0.05)0.99Se, were synthesized in ingot form by melting constituent elements, pulverized into fine powders, and sieved to less than 45 μm. The pristine powders were ball milled (denoted as BM) for increasing surface area and consequently maximizing the effect of reduction, reduced under a 4% H2/Ar atmosphere at 613 K for 6 h to clean off oxide residues and possibly physisorbed O2 (RD), and ball milled and subsequently reduced (BR), respectively. A schematic illustration of this process is given in Figure 1. These four kinds of samples were consolidated into dense pellets by spark plasma sintering (SPS). The mass densities of the pristine, BM, RD, and BR SPS specimens were 96.5%, 96.8%, 93.5%, and 94.1%, respectively, of the theoretical value (Table S1). All procedures for sample preparation, including grinding, sieving, ball milling of powders, as well as opening of reaction tubes, were strictly performed in an argon atmosphere (99.99% purity). All samples adopt the SnSe structure with no observable impurities according to powder X-ray diffraction (PXRD) (Figure S1). The BM sample features a broad peak width because of the nanoscale (20–50 nm) and partially amorphized powders by a ball mill process (Figure S2). The high crystallinity is recovered in the BR sample by the H2 reduction process at 613 K. Pristine, BM, RD, and BR samples show a negligible weight loss up to 800 K under an Ar flow according to thermogravimetric analysis (TGA), indicating high thermal stability (Figure S3). It is important to note that all chalcogenide materials are prone to oxidation when exposed to air with varying degrees of sensitivity. Tin (Sn) chalcogenide materials have to be dealt with with special care because of the spontaneous formation of oxide thin films on the surface that prohibits further oxidation.25Lawless K.R. The oxidation of metals.Rep. Prog. Phys. 1974; 37: 231-316Crossref Scopus (301) Google Scholar On that account, powders of tin chalcogenide compounds would be contaminated with oxide thin films either inherently originating from the starting reagents or upon exposure to air. Furthermore, Sn can adopt both divalent and slightly more stable tetravalent formal oxidation states, which can coexist in Sn-containing chalcogenides, oxides, and halides, giving unavoidable vacancies in cationic lattices in their structures.26Chung I. Song J.H. Im J. Androulakis J. Malliakas C.D. Li H. Freeman A.J. Kenney J.T. Kanatzidis M.G. CsSnI3: semiconductor or metal? High electrical conductivity and strong near-infrared photoluminescence from a single material. High hole mobility and phase-transitions.J. Am. Chem. Soc. 2012; 134: 8579-8587Crossref PubMed Scopus (726) Google Scholar As a result, removal of SnO and SnO2 is essential to unveil the inherent TE properties of SnSe-based materials. Indeed, tetragonal SnO2 shows total thermal conductivity (κtot) of ∼98 and ∼55 W m−1 K−1 along the c and a axes, respectively, at room temperature.22Zhao L.-D. Chang C. Tan G. Kanatzidis M.G. SnSe: a remarkable new thermoelectric material.Energy Environ. Sci. 2016; 9: 3044-3060Crossref Google Scholar SnO has κtot of ∼2 W m−1 K−1 at room temperature, and it readily transforms to SnO2 with time or at high temperature.27Miller S.A. Gorai P. Aydemir U. Mason T.O. Stevanović V. Toberer E.S. Snyder G.J. SnO as a potential oxide thermoelectric candidate.J. Mater. Chem. C. 2017; 5: 8854-8861Crossref Google Scholar As a consequence, their presence even in low levels can significantly raise κtot and depress electrical conductivity because of carrier scattering at the interface, thus damaging the TE performance of SnSe-based materials. Furthermore, because as-grown SnO2 typically shows n-type conductivity, its presence can also be detrimental to the electrical transport properties of intrinsic p-type SnSe.28Singh A.K. Janotti A. Scheffler M. Van de Walle C.G. Sources of electrical conductivity in SnO2.Phys. Rev. Lett. 2008; 101 (055502)Crossref Scopus (345) Google Scholar To establish the presence of tin oxide layers on the pristine powder, we performed TEM. We first investigated the SPS-processed pristine pellet employing spherical aberration-corrected scanning TEM (Cs-STEM) equipped with an energy dispersive spectroscopy (EDS) detector. Its low-magnification bright-field (BF) STEM image, given in Figure 2A, shows nanoscale ∼20 nm precipitates appearing brighter than the surrounding matrix. Elemental mapping reveals that the precipitates are devoid of selenium but are richer in oxygen, while tin atoms are homogeneously distributed throughout the specimen. Surprisingly, nanoscale particles of tin oxides are present in the pristine samples that were strictly prepared in an Ar atmosphere. Subsequently, we examined the BM pristine powders to look for oxide contaminants. The size of the obtained particles generally ranges from ∼20 to 50 nm. Typical low-magnification STEM image for the BM sample shows randomly oriented nanoscale grains with amorphous regions (Figure S2A). Fast Fourier transform (FFT) images taken on the rectangles in Figure S2A also give diffuse features. The representative dark-field (DF) STEM image of the BM sample in Figure 2B shows aggregates of smaller and darker nanoscale particles and larger and brighter crystallites. This indicates that the former would have a smaller molecular weight given the DF-STEM image. Indeed, the former are tin oxides with a negligible amount of Se atoms, and the latter comprises a significantly higher degree of Se than O atoms according to elemental scanning on the corresponding area and quantitative analysis by STEM-EDS (Figures 2B and S4). Interestingly, the oxygen concentration is much higher at the edges than in the central part of the BM powders (Figure S5). We also observed that oxide aggregates would be broken off the SnSe particles by the ball mill process. These exist as isolated nanoscale particles or as thin films on the surface of the SnSe particles. The features are readily seen in the BM samples (see additional STEM images in Figures S6 and S7). In contrast to the pristine and BM samples, the BR sample shows well-ordered nanoscale grains with recovered high crystallinity according to representative low-magnification STEM images and the corresponding FFT images taken at the different areas (Figure S2B and insets). Elemental analysis by STEM-EDS of the corresponding area shows only a trace of oxygen (Figures 2C and S8). However, despite the very significant decrease of the oxygen content, some powders still showed some residual level of oxygen even after the H2 reduction process. The surface state of BM and BR powders was further investigated using X-ray photoemission spectroscopy (XPS) (Figure S9). To directly observe surface overlayers of oxide contaminants, we did not perform any surface treatment such as ion sputtering and mechanical cleaning. For both samples, the Sn 3d3/2 and 3d5/2 peaks are split into two profiles, indicating the two different oxidation states in Sn. The deconvoluted peaks of 494.7 and 486.3 eV are clearly assigned to the Sn 3d3/2 and 3d5/2 peaks from SnO2, respectively. Note that the binding energy of Sn 3d3/2 and 3d5/2 for SnO and SnSe is too close to be deconvoluted to each other. On that account, those of 493.2 and 484.8 eV mainly correspond to the Sn 3d3/2 and 3d5/2 peaks from SnSe while the presence of SnO cannot be excluded.29Wang J.J. Lv A.F. Wang Y.Q. Cui B. Yan H.J. Hu J.S. Hu W.P. Guo Y.G. Wan L.J. Integrated prototype nanodevices via SnO2 nanoparticles decorated SnSe nanosheets.Sci. Rep. 2013; 3: 2613Crossref PubMed Scopus (44) Google Scholar The XPS data demonstrate that a ratio of SnSe to SnO2 is 1.15 for the BM sample and is raised to 1.36 for the BR sample. The latter also shows the greatly reduced oxygen concentration. These observations reveal that a substantial degree of tin oxides (SnO2 and SnO) exists in SnSe-based materials, even after accounting for the fact that facile adsorption of oxygen on sample surfaces can lead to an exaggerated concentration. It is important to note that the presence of Sn4+ species and the resulting structural change could not be observed by X-ray absorption near edge spectroscopy (XANES) and extended X-ray absorption fine structure (EXAFS) employing synchrotron X-ray for both samples. This suggests that only remnants of the tin oxides form as thin films on the surface of the powders rather than inside the bulk. Ultraviolet-visible near infra-red (UV-vis-NIR) diffuse reflectance measurements on pristine and BM samples show an absorption edge at 0.88 eV with residual absorption below the band gap because of surface impurities (Figure S10). Interestingly, the H2-treated BR sample lacks the residual absorption below the band gap and exhibits a much steeper and slightly blue-shifted absorption edge at 0.97 eV. This observation indicates that a reduction process removes tin oxide residues and defect states, thereby unveiling the intrinsic absorption spectrum of this material. Indeed, SnO bulk powders display a well-known direct band gap at ∼2.7 eV with extended band tailing down to ∼0.7 eV as a result of an indirect band gap.30Ogo Y. Hiramatsu H. Nomura K. Yanagi H. Kamiya T. Hirano M. Hosono H. p-Channel thin-film transistor using p-type oxide semiconductor, SnO.Appl. Phys. Lett. 2008; 93Crossref Scopus (569) Google Scholar Similarly, the BM and BR processed samples of undoped SnSe powders give an energy gap at 0.91 and 0.98 eV, respectively, with a steeper absorption edge in the latter. After having established the presence of tin oxides on the surfaces of polycrystalline pristine and BM samples and the considerable decrease of these oxides in the H2-reduced BR samples, we proceeded with the comparison of the thermal and charge transport properties of the polycrystalline SnSe-based samples. The measurements were done both parallel and perpendicular to the press direction of SPS. We will mainly focus on those along the parallel direction that show higher ZT, unless noted otherwise. The most striking changes caused by the H2-reducing treatment on the polycrystalline SnSe samples were observed in the thermal conductivity. Before this treatment, the thermal conductivity is in the range of ∼1 W m−1 K−1 at 300 K and decreases to ∼0.38 W m−1 K−1 at 773 K. After chemical reduction, the thermal conductivity drops to ∼0.7 W m−1 K−1 at 300 K and diminishes to ∼0.20 W m−1 K−1 at 773 K. κtot of the BR sample decreases continuously to 773 K with increasing temperature (Figure 3A). It is lower than that of pristine over the entire range of temperature. Note that κtot of pristine decreases sluggishly and deviates largely from that of the BR sample above ∼673 K. The BR sample reaches a minimum κtot of ∼0.20 W m−1 K−1 at 773 K. This value is about half of ∼0.38 W m−1 K−1 and is also lower than the κtot of undoped and Na-doped SnSe single crystals along the a axis, which is the out-of-plane direction giving the lowest κtot of SnSe materials (Figure 3B). Up to ∼750 K, its κtot is similar to and slightly higher than that for single crystals of SnSe along the b (in-plane) axis and Na-doped SnSe along the a axis, respectively. This observation can be ascribed to tin oxides that remain in the BR sample as observed in the STEM and XPS results. More harsh reduction processes or rigorous purification of reactant elements can further decrease κtot of polycrystalline SnSe materials. A ratio of lattice thermal conductivity (κlat) to κtot demonstrates that lattice vibrations dominate κtot (Figure S11). The BR sample shows a minimum κlat of ∼0.11 W m−1 K−1 at 773 K (Figure S12). This value is about one-third of pristine (∼0.31 W m−1 K−1) and one-half of SnSe single crystals (∼0.20 W m−1 K−1) (Figure 3C). These results clearly confirm that ultralow thermal conductivity is intrinsic for SnSe materials, and in the polycrystalline form they indeed exhibit lower κlat than corresponding single crystals. κtot perpendicular to the SPS direction is moderately higher. It is 1.20 W m−1 K−1 at 300 K and reaches a minimum of 0.33 W m−1 K−1 at 773 K, slightly higher than 0.20 W m−1 K−1 parallel to the SPS direction. The thermal diffusivity (D) of the BR samples is nearly independent of the thickness of specimens (0.650, 1.000, 1.215, 1.350, and 1.700 mm) (Figure S13). The standard deviation in D with respect to the thickness is 7.3% at 773 K. This observation ensures that ultralow thermal conductivity is an intrinsic property of SnSe materials. The ball milling process followed by chemical reduction described in this work can uncover the intrinsically ultralow thermal conductivity in polycrystalline SnSe-based materials with excellent reproducibility. We prepared twelve independent BR specimens, and their ultralow κtot was cross-confirmed from two institutions of Seoul National University (SNU) and Northwestern University (NU) and the company of Netzsch instruments (Figure 4). The relative density of all BR samples for the reproducibility test ranges from 93.9% to 95.6% (average density: 94.6%). The uncertainty of κtot for those specimens (8 from SNU, 3 from NU, and 1 from Netzsch) ranges from 7% to 20% in the temperature range from 300 to 773 K. Note that this uncertainty originates from both individual specimens and instruments. Their average κtot at 773 K is ∼0.21 W m−1 K−1, confirming the reproducibility of intrinsically ultralow thermal conductivity of polycrystalline SnSe-based materials. Note that the uncertainty for samples 1–8 measured at SNU is 11.1% at 773 K. Many research groups have obtained higher κlat in polycrystalline SnSe materials than in single crystals.23Wei P.C. Bhattacharya S. He J. Neeleshwar S. Podila R. Chen Y.Y. Rao A.M. The intrinsic thermal conductivity of SnSe.Nature. 2016; 539: E1-E2Crossref PubMed Scopus (119) Google Scholar The discrepancy can now be explained in light of the characteristic difference in preparing single-crystal and polycrystalline specimens. We found in this work that tin oxides form thin films on powders of SnSe-based materials and probably even in single crystals upon sufficient air exposure. A consolidation process such as SPS spreads the oxide films on grain boundaries throughout the pellets (Figure 1). Consequentially, grain boundaries involving tin oxides could even boost phonon transport rather than impede it because SnO2 has thermal conductivity ∼140-fold highter than that of SnSe.22Zhao L.-D. Chang C. Tan G. Kanatzidis M.G. SnSe: a remarkable new thermoelectric material.Energy Environ. Sci. 2016; 9: 3044-3060Crossref Google Scholar, 31Turkes P. Pluntke C. Helbig R. Thermal conductivity of SnO2 single crystals.J. Phys. C. 1980; 13: 4941Crossref Scopus (67) Google Scholar Figure 5A shows the electrical conductivity (σ) of the samples. It is clear that the samples having undergone the H2 treatment have higher electric conductivity than the ones that have not. The BR sample reaches a maximum (σmax) of 93 and 107 S cm−1 at 773 K, taken parallel and perpendicular to press direction of SPS, respectively, which is about double that of the pristine. These values are much higher than those reported for other polycrystalline SnSe-based materials. In comparison, undoped and Na-doped SnSe single crystals show σmax of ∼85 and 148 S cm−1 along the b axis (in plane) and ∼14 and ∼40 S cm−1 along the a axis (out of plane), respectively. The remarkably enhanced σ of the BR sample reflects the removal of tin oxides at the grain boundaries, which can scatter charge carriers. The BR sample shows p-type character and a higher Seebeck coefficient (S) than the other samples (Figure 5B), and this is consistent with a lower hole carrier concentration (Figure S14). This is consistent with the H2 treatment, which is a reduction process. The Seebeck coefficient monotonously increases to reach ∼+375 μV K−1 at 673 K and then declines to +271 μV K−1 at 773 K. The values along the parallel and perpendicular directions are nearly the same (Figure S15). We confirmed the high reproducibility of σ and S values from the BR samples that were independently synthesized (Figure S16). The improved σ and S in the mid temperature range from ∼450 to 600 K result in relatively high power factor (σS2) for SnSe-based materials of ∼5 to 5.5 μW cm−1 K−1 forming a plateau in that temperature range. The BR sample reaches the enhanced power factor of 6.85 μW cm−1 K−1 at 773 K compared to 3.12 μW cm−1 K−1 for the pristine. The reduced sample without a ball mill process (RD) also gives an increased power factor of 4.85 μW cm−1 K−2 (Figure 5C). Power factor is slightly higher along the perpendicular direction to give ∼8.0 μW cm−1 K−1 at 773 K and a maximum of 8.14 μW cm−1 K−1 at 800 K. The temperature-dependent Hall effect measurement shows that μH at 300 K for the pristine an" @default.
• W2912055813 created "2019-02-21" @default.
• W2912055813 creator A5003515264 @default.
• W2912055813 creator A5029371959 @default.
• W2912055813 creator A5035758112 @default.
• W2912055813 creator A5077382159 @default.
• W2912055813 creator A5089183036 @default.
• W2912055813 date "2019-03-01" @default.
• W2912055813 modified "2023-09-30" @default.
• W2912055813 title "Surface Oxide Removal for Polycrystalline SnSe Reveals Near-Single-Crystal Thermoelectric Performance" @default.
• W2912055813 cites W1965905995 @default.
• W2912055813 cites W1973277081 @default.
• W2912055813 cites W1977057564 @default.
• W2912055813 cites W1981033590 @default.
• W2912055813 cites W1986250062 @default.
• W2912055813 cites W1987090462 @default.
• W2912055813 cites W1995997336 @default.
• W2912055813 cites W2007798826 @default.
• W2912055813 cites W2030650018 @default.
• W2912055813 cites W2076218685 @default.
• W2912055813 cites W2103285346 @default.
• W2912055813 cites W2112425685 @default.
• W2912055813 cites W2137813091 @default.
• W2912055813 cites W2175477086 @default.
• W2912055813 cites W2260501105 @default.
• W2912055813 cites W2317983131 @default.
• W2912055813 cites W2407654267 @default.
• W2912055813 cites W2412292761 @default.
• W2912055813 cites W2507875578 @default.
• W2912055813 cites W2511197735 @default.
• W2912055813 cites W2534277249 @default.
• W2912055813 cites W2548132273 @default.
• W2912055813 cites W2581847710 @default.
• W2912055813 cites W2591932046 @default.
• W2912055813 cites W2605096937 @default.
• W2912055813 cites W2735316262 @default.
• W2912055813 cites W2741192582 @default.
• W2912055813 cites W2743854742 @default.
• W2912055813 cites W2755006415 @default.
• W2912055813 cites W2767698201 @default.
• W2912055813 cites W2803116553 @default.
• W2912055813 cites W2805531970 @default.
• W2912055813 doi "https://doi.org/10.1016/j.joule.2019.01.001" @default.
• W2912055813 hasPublicationYear "2019" @default.
• W2912055813 type Work @default.
• W2912055813 sameAs 2912055813 @default.
• W2912055813 citedByCount "151" @default.
• W2912055813 countsByYear W29120558132019 @default.
• W2912055813 countsByYear W29120558132020 @default.
• W2912055813 countsByYear W29120558132021 @default.
• W2912055813 countsByYear W29120558132022 @default.
• W2912055813 countsByYear W29120558132023 @default.
• W2912055813 crossrefType "journal-article" @default.
• W2912055813 hasAuthorship W2912055813A5003515264 @default.
• W2912055813 hasAuthorship W2912055813A5029371959 @default.
• W2912055813 hasAuthorship W2912055813A5035758112 @default.
• W2912055813 hasAuthorship W2912055813A5077382159 @default.
• W2912055813 hasAuthorship W2912055813A5089183036 @default.
• W2912055813 hasBestOaLocation W29120558131 @default.
• W2912055813 hasConcept C121332964 @default.
• W2912055813 hasConcept C127413603 @default.
• W2912055813 hasConcept C137637335 @default.
• W2912055813 hasConcept C185592680 @default.
• W2912055813 hasConcept C191897082 @default.
• W2912055813 hasConcept C192562407 @default.
• W2912055813 hasConcept C2524010 @default.
• W2912055813 hasConcept C2776799497 @default.
• W2912055813 hasConcept C2779851234 @default.
• W2912055813 hasConcept C33923547 @default.
• W2912055813 hasConcept C49040817 @default.
• W2912055813 hasConcept C61696701 @default.
• W2912055813 hasConcept C63024428 @default.
• W2912055813 hasConcept C73922627 @default.
• W2912055813 hasConcept C8010536 @default.
• W2912055813 hasConcept C97355855 @default.
• W2912055813 hasConceptScore W2912055813C121332964 @default.
• W2912055813 hasConceptScore W2912055813C127413603 @default.
• W2912055813 hasConceptScore W2912055813C137637335 @default.
• W2912055813 hasConceptScore W2912055813C185592680 @default.
• W2912055813 hasConceptScore W2912055813C191897082 @default.
• W2912055813 hasConceptScore W2912055813C192562407 @default.
• W2912055813 hasConceptScore W2912055813C2524010 @default.
• W2912055813 hasConceptScore W2912055813C2776799497 @default.
• W2912055813 hasConceptScore W2912055813C2779851234 @default.
• W2912055813 hasConceptScore W2912055813C33923547 @default.
• W2912055813 hasConceptScore W2912055813C49040817 @default.
• W2912055813 hasConceptScore W2912055813C61696701 @default.
• W2912055813 hasConceptScore W2912055813C63024428 @default.
• W2912055813 hasConceptScore W2912055813C73922627 @default.
• W2912055813 hasConceptScore W2912055813C8010536 @default.
• W2912055813 hasConceptScore W2912055813C97355855 @default.
• W2912055813 hasFunder F4320306084 @default.
• W2912055813 hasFunder F4320322120 @default.
• W2912055813 hasFunder F4320326441 @default.
• W2912055813 hasFunder F4320332359 @default.
• W2912055813 hasFunder F4320337480 @default.
• W2912055813 hasIssue "3" @default.
• W2912055813 hasLocation W29120558131 @default.

• next