Hierarchical structures lead to high thermoelectric performance in Cu_m+nPb₁₀₀Sb_mTe₁₀₀Se_2m (CLAST)

Abstract

Ternary compound CuSbSe₂-alloyed PbTe, Cu_mPb₁₀₀Sb_mTe₁₀₀Se_2m (CLAST), presents outstanding n-type thermoelectric transport behavior and features hierarchical Cu-based precipitates and interstitials that can balance phonon and carrier transport. Results show that a small amount of CuSbSe₂ (∼3%) alloying in CLAST can realize a room-temperature carrier concentration of ∼1.7 × 10¹⁸ cm⁻³ and then optimize the power factor, and simultaneously precipitate out embedded Cu-based nanostructures in the matrix to lower the lattice thermal conductivity. Additionally, extra Cu atoms adding in CLAST can form interstitials and further improve both the carrier concentration to ∼3.0 × 10¹⁸ cm⁻³ and carrier mobility to ∼1227.8 cm² V⁻¹ s⁻¹ at room temperature, which benefits a maximum power factor of ∼20.0 μW cm⁻¹ K⁻² in Cu_3.3Pb₁₀₀Sb₃Te₁₀₀Se₆. Moreover, the Cu interstitials together with massive Cu-based nanoprecipitates can strongly scatter a wide set of phonons, and largely lower the lattice thermal conductivity to ∼0.44 W m⁻¹ K⁻¹ in Cu_3.4Pb₁₀₀Sb₃Te₁₀₀Se₆ at 623 K. Finally, these Cu-based hierarchical structures in CLAST samples can synergistically optimize the phonon and carrier transport properties and contribute to a high ZT of ∼0.5 at 300 K and a peak ZT of ∼1.4 at 723 K. A remarkably high ZT_ave of ∼0.94 at 300–723 K is achieved in Cu_3.3Pb₁₀₀Sb₃Te₁₀₀Se₆ due to high ZT values in the low temperature range, outperforming other high-performance n-type PbTe-based thermoelectric materials.

---

Broader context

With more than 60% of input energy being lost as waste heat, thermoelectric materials which make it possible to directly and reversibly convert heat into electrical energy have received ever-increasing attention. Lead telluride (PbTe), a kind of traditional mid-temperature thermoelectric material, has been well-developed in the past few decades. For PbTe-based materials, due to the favorable complex electronic band structures, abundant studies indicate that p-type materials exhibit better thermoelectric performance than n-type counterparts, therefore the optimization of n-type materials to match p-type counterparts is becoming an urgent issue. Here, we report an n-type PbTe system, Cu_m+nPb₁₀₀Sb_mTe₁₀₀Se_2m (abbreviated as CLAST), achieving a room-temperature superior ZT of ∼0.5 and a high ZT of ∼1.4 at 723 K through designing hierarchical microstructures of nanoprecipitates and atomic interstitials. Alloying CuSbSe₂ into PbTe improves the power factor and concurrently reduces the thermal conductivity through producing nanoprecipitations. And then, extra Cu interstitials are added, which enhance carrier concentration, maintain carrier mobility, and further suppress thermal conductivity. This work provides a valuable path to collectively optimize carrier and phonon transport.

Introduction

Thermoelectric technology makes it possible to directly convert heat into electricity based on the Seebeck effect, which is a promising energy source.^1–5 The thermoelectric conversion efficiency strongly depends on the dimensionless figure of merit ZT of materials, ZT = S²σT/(κ_lat + κ_ele), where S is the Seebeck coefficient, σ is electrical conductivity, T is absolute temperature in kelvin, κ_lat is lattice thermal conductivity and κ_ele is electronic thermal conductivity. Achieving high S, high σ, and low κ_lat and κ_ele is the key to acquiring high-performance thermoelectric materials. Due to the coupled relationships between these thermoelectric parameters, obtaining a high ZT value by synchronously optimizing carrier and phonon transport properties is pretty difficult.^6–9 Strategies for scattering phonons with different wavelengths include point defects (including substitutional atoms,^10,11 interstitial atoms,^12,13 and vacancies^14,15), grain boundaries^16–18 and nanoscale precipitates embedded in the matrix.^19–21 Various kinds of methods aim to enhance the power factor (S²σ), including optimization of the carrier concentration and carrier mobility^22–24 and band structure manipulations (resonant states²⁵ and band convergence^26–28).

As a traditional mid-temperature range material for thermoelectric power generation, PbTe has been intensively investigated.^29–32 Recently, p-type PbTe has attained record-high ZT and ZT_ave with synergistically optimized strategies of manipulating valence band convergence and nanostructuring, such as PbTe–Na–SrTe,³³ PbTe–Na–PbS,³⁴ PbTe–S–K,³⁵etc. Although p-type PbTe has been well developed and presents high thermoelectric performance, the thermoelectric performance in n-type counterparts still maintains at a relatively low level, causing a hindrance to widely practical applications of PbTe-based materials. Due to the large offset (∼0.45 eV) between the conduction bands in n-type PbTe,^36,37 the strategies to enhance the thermoelectric performance are very different with the p-type PbTe. Most previous research focuses on improving the electrical transport properties by optimizing the carrier concentration or suppressing thermal conductivity by nanostructure design. For instance, In or Ga elements doped in n-type PbTe to form a defect level can well optimize the carrier concentration and finally obtain high power factor.^38–41 Sn alloying in n-type PbTe to sharpen the conduction band shape can well optimize the carrier effective mass, and maintain relatively high carrier mobility and power factor.^42,43 However, these studies mainly focus on optimizing carrier transport but still maintain high thermal conductivity, which limits the development of thermoelectric performance. Some researchers are devoted to reducing the lattice thermal conductivity by nanostructures, such as PbTe–ZnTe,⁴⁴ PbTe–CdTe,⁴⁵ PbTe–CaTe,⁴⁶ PbTe–PbSnS₂,⁴⁷etc. However, the existence of nanoprecipitates reinforces the scattering of carriers, in turn damaging the power factor. It can be seen that previous studies could not achieve high power factor and low lattice thermal conductivity at the same time. Therefore, it is essential to develop new approaches to realize high thermoelectric performance by balancing carrier and phonon transport properties.

This work reported a high-performance n-type Cu_m+nPb₁₀₀Sb_mTe₁₀₀Se_2m (abbreviated as CLAST in the following text) thermoelectric material with hierarchical structures including Cu-based nanoscale precipitates and atomic interstitials, and the proposed structures are schematically shown in Fig. 1. High-performance CLAST is obtained by firstly alloying ternary compound CuSbSe₂ to form nanoprecipitates and then adding extra Cu to form atomic Cu interstitials. Interestingly, these hierarchical structures in the CLAST matrix can not only well strengthen the power factor but also depress the lattice thermal conductivity. With CuSbSe₂ alloying and extra Cu added, the room-temperature carrier concentration and carrier mobility can be optimized to ∼3.0 × 10¹⁸ cm⁻³ and ∼1227.8 cm² V⁻¹ s⁻¹, which contributes to a maximum power factor of ∼20.0 μW cm⁻¹ K⁻² in Cu_3.3Pb₁₀₀Sb₃Te₁₀₀Se₆. Furthermore, the minimum lattice thermal conductivity can reach as low as ∼0.44 W m⁻¹ K⁻¹ in Cu_3.4Pb₁₀₀Sb₃Te₁₀₀Se₆ at 623 K. As a result, a room-temperature ZT of ∼0.5 and a peak ZT of ∼1.4 at 723 K are achieved in Cu_3.3Pb₁₀₀Sb₃Te₁₀₀Se₆ through two-step introducing hierarchical structures of nanoprecipitates and interstitials, leading to a transporting balance between carrier and phonon.

Fig. 1 Schematic of Cu-based forms in CLAST. Nanoscale Cu-based precipitates form in CLAST with ternary compound CuSbSe₂ alloying, and then atomic Cu interstitials appear with extra Cu atoms added. These hierarchical structures can effectively optimize the carrier concentration (n), maintain high carrier mobility (μ) and simultaneously intensively reduce the lattice thermal conductivity (κ_lat).

Results and discussion

To obtain high-performance n-type PbTe-based thermoelectric materials, this work synergistically optimizes the carrier and phonon transport properties by artificially introducing Cu-based hierarchical structures into Cu_mPb₁₀₀Sb_mTe₁₀₀Se_2m (CLAST). More experiment details are in the ESI.† Firstly, ternary compound CuSbSe₂ alloying in PbTe forms Cu-based nanostructures to dramatically lower the lattice thermal conductivity. Then, extra Cu is utilized and forms interstitial atoms, leading to both a good synergistic optimization of carrier and phonon transport and a further diminution in lattice thermal conductivity. To unveil the origin of high performance in CLAST thermoelectrics, the microstructure observations are conducted to systematically investigate the existing forms of Cu atoms.

Crystal structure identification in CLAST

As shown in Fig. 2(a) and (b), PbTe has a cubic rock-salt structure with the Fm3m space group, and CuSbSe₂ exhibits a 2D double-layered structure with the Pnma space group. Besides, the lattice parameters between these two phases are very different, CuSbSe₂: a ∼ 6.40 Å, b ∼ 3.95 Å, c ∼ 15.3 Å, and PbTe: a = b = c ∼ 6.44 Å. From the XRD patterns, a small amount of CuSbSe₂ alloyed with PbTe is dominated by the NaCl structure, named as Cu_mPb₁₀₀Sb_mTe₁₀₀Se_2m (CLAST). Meanwhile, no impurity phase can be found in the powder XRD measurement of the Cu_mPb₁₀₀Sb_mTe₁₀₀Se_2m (m = 0, 2, 3, 4) samples due to the detection limit of the equipment. Noticeably, the following TEM results disclose that some Cu-based nanostructures indeed precipitate out in the CLAST sample, which will be discussed later. In Fig. 2(d), the falling lattice parameter with the increase of CuSbSe₂ content indicates that small-sized Cu⁺ (∼0.46 Å) and Sb³⁺ (∼0.92 Å) can partly incorporate in Pb²⁺ (∼1.2 Å) sites, and small-sized Se²⁻ (∼1.98 Å) incorporates in Te²⁻ (∼2.1 Å) sites.

Fig. 2 Crystal structures of PbTe and CuSbSe₂, and phase identification of Cu_mPb₁₀₀Sb_mTe₁₀₀Se_2m (CLAST) (m = 0, 2, 3, 4): (a) PbTe crystal structure; (b) CuSbSe₂ crystal structure; (c) powder XRD pattern; (d) lattice parameter.

Thermal and electrical transport properties in CLAST

In Fig. 3(a), compared with undoped PbTe, the total thermal conductivity of CLAST is obviously low. The lattice thermal conductivity is equal to the total thermal conductivity minus electronic thermal conductivity, and the electronic thermal conductivity is computed by κ_ele = LσT, where L is the Lorenz number, σ is electrical conductivity and T is the temperature in kelvin. With the incremental CuSbSe₂ content, the total thermal conductivity is gradually lowered because of the decreased lattice thermal conductivity with nano-precipitate scattering (detailed discussion is given below), as shown in Fig. 3(b). The minimum lattice thermal conductivity is suppressed from ∼1.28 W m⁻¹ K⁻¹ in PbTe to ∼0.65 W m⁻¹ K⁻¹ in Cu₄Pb₁₀₀Sb₄Te₁₀₀Se₈. In Fig. 3(c), the optimal electrical conductivity is promoted to ∼320.7 S cm⁻¹ in Cu₃Pb₁₀₀Sb₃Te₁₀₀Se₆ at 573 K because of the improved carrier concentration. Specifically, the temperature-dependent electrical conductivity shows a trend of first rising and then falling in CLAST samples, which may originate from the combination of dynamic doping and phase transition of the Cu-containing phase in CLAST.^48,49 In Fig. 3(d), undoped PbTe experiences a p–n transformation, and CLAST exhibits n-type semiconductor characteristic with the negative Seebeck coefficient. The absolute value of room-temperature Seebeck coefficient is reduced from ∼316.7 μV K⁻¹ to ∼239.3 μV K⁻¹. Integrating the improved electrical conductivity and Seebeck coefficient, the maximum power factor reaches ∼16.3 μW cm⁻¹ K⁻¹ in Cu₃Pb₁₀₀Sb₃Te₁₀₀Se₆ at 573 K, as shown in Fig. 3(e). Finally, the ZT values reflect an incremental trend with temperature, and a maximum ZT of ∼1.0 is achieved in Cu₃Pb₁₀₀Sb₃Te₁₀₀Se₆ at 723 K, as shown in Fig. 3(f). Other thermoelectric data with respect to CLAST are listed in Fig. S1 and S2 (ESI†), such as Lorenz number L, electronic thermal conductivity κ_ele, heat capacity C_p, thermal diffusivity D, and sample density ρ.

Fig. 3 Thermoelectric transport properties of Cu_mPb₁₀₀Sb_mTe₁₀₀Se_2m (m = 0, 2, 3, 4): (a) total thermal conductivity; (b) lattice thermal conductivity; (c) electrical conductivity; (d) Seebeck coefficient; (e) power factor; (f) ZT value.

Microstructures in CLAST

To disclose the low thermal conductivity in CLAST, we carried out transmission electron microscopy (TEM) analysis to explore the microstructures of CLAST. Fig. 4(a) displays the bright-field TEM image obtained from Cu₃Pb₁₀₀Sb₃Te₁₀₀Se₆, where nanoscale precipitates with an average size of ∼100 nm are clearly observed. Fig. 4(b) and (c) show the selected area electron diffraction (SAED) patterns taken along [100] from the CLAST matrix as well as the areas with nanoprecipitates. The splitting spots in Fig. 4(c) indicate that the precipitates have a similar cubic structure to the rock-salt PbTe matrix. It can also be found that the lattice mismatch between the PbTe matrix and the precipitates along [100] is about 7% by comparing the SAED patterns.

Fig. 4 Nanoprecipitates in Cu₃Pb₁₀₀Sb₃Te₁₀₀Se₆: (a) low magnification bright-field TEM image, with red arrows marking some of the precipitates; electron diffraction along [100] obtained from (b) a small local region that is almost free from nanoprecipitates and (c) a relatively larger region that contains considerable nanoprecipitates. The splitting spots in (c) show that the nanoprecipitates have a similar cubic structure to the PbTe matrix, although a large lattice mismatch by ∼7% is found; (d1) The HRTEM image of the nano-precipitate. Inset showing the Fast-Fourier-Transform (FFT) image that reveals the presence of dislocations, together with the corresponding diffraction spots chosen marked in red; (d2 and d3) GPA maps of (d1) from horizontal (E_xx) and vertical (E_yy) directions; (e1–e6) the ADF image and corresponding EDS element maps of Te, Pb, Cu, Sb and Se, respectively.

Fig. 4(d1) displays the high resolution TEM (HRTEM) image of the nanoprecipitates, and the dislocations appear as periodic structures as shown in the Fourier transform image of the inset of Fig. 4(d1), consistent with the lattice mismatch of the splitting diffraction spots in Fig. 4(c). Thus, the lattice distortion could be compensated by forming a semi-coherent interface between the matrix and the precipitates. Geometric phase analysis (GPA) is used to determine lattice distortion with respect to the reference matrix structure, as depicted in Fig. 4(d2 and d3). Fig. 4(e1–e6) are the annular dark-field (ADF) images and the corresponding energy dispersive spectroscopy (EDS) elemental maps. We observed a homogeneous distribution of each element in either the matrix or the precipitates. In particular, it is obvious that the precipitates show a composition that is Pb-poor and Cu-rich.

Crystal structure identification in extra Cu-added CLAST

In Fig. 5(a), all the samples can be indexed as a NaCl structure with no distinct secondary phases detected. From the enlarged XRD patterns at ∼27.6 degree, no obvious peak shifts can be observed compared with a CLAST sample, indicating small changes of lattice parameter in extra Cu-added CLAST samples. In Fig. 5(b), the calculated lattice parameters in Cu-added CLAST samples show slight vibrations around ∼6.44 Å.

Fig. 5 Phase identification of Cu_3+nPb₁₀₀Sb₃Te₁₀₀Se₆ (n = 0–0.5): (a) powder XRD pattern; (b) lattice parameter.

Electrical transport properties in extra Cu-added CLAST

In Fig. 6(a), the extra Cu added in CLAST can largely enhance the electrical transport properties, and the room-temperature electrical conductivity is improved from ∼177.6 S cm⁻¹ in Cu₃Pb₁₀₀Sb₃Te₁₀₀Se₆ to ∼1077.3 S cm⁻¹ in Cu_3.4Pb₁₀₀Sb₃Te₁₀₀Se₆ through adding a small amount of Cu, n = 0.4. The negative Seebeck coefficient indicates that extra Cu-added CLAST samples are dominated with charge carriers of electrons, and the absolute values of the Seebeck coefficients decrease with increasing Cu content, as shown in Fig. 6(b). The Pisarenko relationship in Fig. S3 (ESI†) unveils that the carrier effective mass in extra Cu-added CLAST samples maintains around ∼0.3m₀. The variations of the electrical conductivity and Seebeck coefficient can be commendably accounted for by the enhanced carrier concentration, as shown in Fig. 6(e). In Fig. 6(e), the Hall measurement shows that carrier concentration manifests an obvious upward trend with an increasing amount of Cu, indicating that Cu atoms form massive interstitials to work as a donor in CLAST rather than the hole-doping by Cu in Pb sites, similar as Ag works in Ag_nPb₁₀₀In_nTe_100+2n (LIST) samples.³⁹ Additionally, the carrier mobility of Cu_3+nPb₁₀₀Sb₃Te₁₀₀Se₆ is generally higher than that of other n-type PbTe systems (PbTe–GeTe,⁵⁰ PbTe–Se,⁵¹ PbTe–MgTe,⁵² PbTe–MnTe,⁵³ PbTe–CdTe,⁵⁴ PbTe–Ag₂Te,⁵⁵ Ag_nPb₁₀₀In_nTe_100+2n³⁹), as shown in Fig. 6(f). It is well known that carrier concentration and carrier mobility are two competing parameters, however, it is very interesting that Cu addition improves both carrier concentration and carrier mobility, which may be related to the interstitials causing weak scattering on carriers⁵⁶ and compensating Pb vacancies.⁵⁷ Combining improved electrical conductivity and reduced Seebeck coefficient over the wide temperature, the power factor of Cu-added CLAST is almost two times larger than CLAST samples across the whole temperature range, and the maximum power factor reaches ∼20.4 μW cm⁻¹ K⁻² for Cu_3.4Pb₁₀₀Sb₃Te₁₀₀Se₆ in Fig. 6(c). The power factor at low temperature is superior to other n-type PbTe systems, such as La-doped PbTe–Ag₂Te,⁵⁸ I-doped PbTe–MgTe,⁵² PbTe–InSb,⁵⁹etc. As we know, good electrical transport properties of materials are determined by the average power factor value PF_ave, calculated as:⁶⁰where T_c is the cold-end temperature (300 K), and T_h is the hot-end temperature (723 K). As shown in Fig. 6(d), the PF_ave of Cu_3+nPb₁₀₀Sb₃Te₁₀₀Se₆ (n = 0.1–0.5) reaches a much higher value, compared with Cu₃Pb₁₀₀Sb₃Te₁₀₀Se₆. Therefore, by virtue of improved carrier concentration and carrier mobility, CLAST with added Cu can exhibit an outstanding electrical transport performance.

Fig. 6 Electrical transport properties of Cu_3+nPb₁₀₀Sb₃Te₁₀₀Se₆ (n = 0–0.5): (a) electrical conductivity; (b) Seebeck coefficient; (c) power factor; (d) average power factor; (e) carrier concentration and carrier mobility at room temperature; and (f) carrier mobility as a function of carrier concentration at room temperature (PbTe–GeTe,⁵⁰ PbTe–Se,⁵¹ PbTe–MgTe,⁵² PbTe–MnTe,⁵³ PbTe–CdTe,⁵⁴ PbTe–Ag₂Te,⁵⁵ Ag_nPb₁₀₀In_nTe_100+2n³⁹).

Microstructures in extra Cu-added CLAST

To further examine the location of Cu atoms in the extra Cu-added CLAST sample, we utilized aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) analysis. Two types of atom structures are observed in the Cu_3+nPb₁₀₀Sb₃Te₁₀₀Se₆ sample along the [100] zone axis, one shows an atom structure projection that is similar to that of PbTe (cf.Fig. 7(a)), and another displays the existence of Cu interstitials (cf.Fig. 7(b)). The detailed atom arrangement is exemplified in the magnified HAADF image (cf.Fig. 7(c)). Noticeably, the Cu interstitials randomly disperse in the CLAST lattice because of their strong vibrations,⁵⁷ which can contribute to intensified phonon scattering and then low lattice thermal conductivity. The additional peaks corresponding to Cu interstitials can also be clearly observed in Fig. 7(d).

Fig. 7 Atom structure observation of Cu_3.3Pb₁₀₀Sb₃Te₁₀₀Se₆. HAADF images obtained from regions (a) without and (b) with Cu interstitial atoms along the [100] zone axis; (c) enlarged image showing Cu interstitial atoms from the marked regions in (b); and (d) intensity scan profiles obtained from the ADF images, as indicated by hollow arrows shown in (a) and (b), respectively. Inset of (c) shows the schematic atom structure of the CLAST with Cu interstitials, which is labeled by white circles. The arrowed peaks in (d) mark the additional peaks corresponding to Cu interstitials.

Thermal transport properties in extra Cu-added CLAST

As depicted in Fig. 8(a), the total thermal conductivity is almost unchanged when n < 0.3 and begins to increase when n > 0.3, which comes from the promoted electronic thermal conductivity. In Fig. 8(b), the increased electronic thermal conductivity is in agreement with the electrical conductivity in Fig. 8(a) because of the increasing carrier concentration. In contrast, the lattice thermal conductivity shows a further obvious reduction with Cu added in Fig. 8(c), and the minimum lattice thermal conductivity decreases sharply to ∼0.44 W m⁻¹ K⁻¹ at 623 K in Cu_3.4Pb₁₀₀Sb₃Te₁₀₀Se₆, which is extremely close to the minimum theoretical value of PbTe (∼0.36 W m⁻¹ K⁻¹). The suppressed lattice thermal conductivity is indicative of local lattice distortion with Cu atoms, which enhances phonon scattering. Meanwhile, the lattice thermal conductivity exhibits an upward trend above 623 K, which originates from the phase transition of Cu-containing phases and the slightly bipolar diffusion in CLAST samples.^61,62 Then, we compared the lattice thermal conductivity of Cu_3.4Pb₁₀₀Sb₃Te₁₀₀Se₆ to that of other high-ZT n-type PbTe, such as PbTe–CdTe,⁵⁴ PbTe–Cu₂Te,⁵⁷ PbTe–I,⁶³ PbTe–Ga,⁴¹ and PbTe–MgTe.⁵² It is obviously noticed that the lattice thermal conductivity of Cu_3.3Pb₁₀₀Sb₃Te₁₀₀Se₆ is relatively low compared with others, especially at low temperature. This comes from the superimposed effects of nanoprecipitates and interstitials. Other thermoelectric parameters, including Lorenz number L, heat capacity C_p, thermal diffusivity D, and sample density ρ, are shown in Fig. S4 (ESI†).

Fig. 8 Thermal transport properties of Cu_3+nPb₁₀₀Sb₃Te₁₀₀Se₆ (n = 0–0.5): (a) total thermal conductivity; (b) electronic thermal conductivity; (c) lattice thermal conductivity; (d) comparison of lattice thermal conductivity of n-type PbTe (PbTe–CdTe,⁵⁴ PbTe–Cu₂Te,⁵⁷ PbTe–I,⁶³ PbTe–Ga,⁴¹ PbTe–MgTe⁵²).

ZT values

To evaluate the comprehensive effects of hierarchical structures, precipitates and interstitials, on thermoelectric performance in CLAST samples, the minimum lattice thermal conductivity (κ_lat) and ratio of carrier mobility to lattice thermal conductivity (μ/κ_lat) are shown in Fig. 9(a). It is obvious that the lattice thermal conductivity is continuously suppressed through introducing precipitates and interstitials in CLAST. Besides intensifying the phonon scattering, these nanoprecipitates and interstitials can also preserve carrier transport, inducing a large enhancement of μ/κ_lat in extra Cu-added CLAST. Consequently, these special hierarchical structures, precipitate and interstitials, can contribute to a maximum ZT of ∼0.5 at 300 K and of ∼1.4 at 723 K in Cu_3.3Pb₁₀₀Sb₃Te₁₀₀Se₆, as shown in Fig. 9(b). When making a comparison for the room-temperature carrier mobility and lattice thermal conductivity in CLAST with others,^{39,41,44,45,50,52–55,57,64,65} it is found that the CLAST sample shows a superior performance through balancing the phonon and carrier transports, as shown in Fig. 9(c). These special hierarchical structures in CLAST make its thermoelectric performance competitive to other high-performance n-type PbTe-based materials, especially at a low temperature range, shown in Fig. 9(d). The average ZT values (ZT_ave) are also calculated as:^66–68where T_c is the cold-end temperature (300 K), and T_h is the hot-end temperature (723 K). In Fig. S5 (ESI†), a remarkably high ZT_ave of ∼0.94 is achieved in Cu_3.3Pb₁₀₀Sb₃Te₁₀₀Se₆ from 300 K to 723 K due to outstanding ZT values in the low temperature range.

Fig. 9 Comparisons of thermoelectric parameters of CLAST and others: (a) ratio of carrier mobility to lattice thermal conductivity (μ/κ_lat) at room temperature; (b) ZT values of Cu_3+nPb₁₀₀Sb₆Te₁₀₀Se₆ (n = 0–0.5); (c) diagram of μ and 1/κ_lat (PbTe–MnTe,⁵³ PbTe–PbSe,⁶⁴ PbTe–Cu₂Te,⁵⁷ PbTe–ZnTe,⁴⁴ PbTe–MgTe,⁵² PbTe–CdTe,⁵⁴ PbTe–Ag₂Te,⁵⁵ PbTe–Ga,⁴¹ Pb_0.98Te–Sb,⁶⁵ PbTe–GeTe,⁵⁰ Ag_nPb₁₀₀I_nTe_100+2n³⁹); (d) comparisons of ZT values in n-type PbTe (PbTe–CdTe,⁵⁴ PbTe–Cu₂Te,⁵⁷ PbTe–I,⁶³ PbTe–Ga,⁴¹ PbTe–MgTe⁵²).

Conclusions

In summary, a maximum ZT ∼ 0.5 at 300 K and ∼1.4 at 723 K, and average ZT ∼ 0.94 at 300–723 K are achieved in an n-type CLAST sample. Its high thermoelectric transport properties are mainly attributed to the hierarchical structures of nanoprecipitates and interstitials, which can well balance the phonon and carrier transport. The nanoprecipitates firstly are formed in the CLAST sample with ternary compound CuSbSe₂ alloying, which can intensify phonon scattering to largely depress the lattice thermal conductivity. With extra Cu added in CLAST, massive interstitials can be observed, and these Cu interstitials can well optimize the carrier concentration and carrier mobility, and simultaneously further lower the lattice thermal conductivity. These special hierarchical structures in CLAST finally make its thermoelectric performance obviously higher than other n-type PbTe-based thermoelectric materials. This work puts forward an effective strategy to synergistically optimize phonon and carrier transports through designing hierarchical structures, and also proves that the interstitials are of great importance to enhance the thermoelectric performance through balancing electrical and thermal transport.

Author contribution

Siqi Wang, Yu Xiao and Li-Dong Zhao designed the research; Siqi Wang compounded the samples and made thermoelectric performance tests; Yongjin Chen, Xiang Gao, Shang Peng and Tao Hong performed TEM characterization; Siqi Wang, Dongyang Wang, Zhi Yang and Yuejun Sun analyzed data and drew diagrams. Siqi Wang, Yu Xiao and Xiang Gao wrote the manuscript and all authors edited the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Key Research and Development Program of China (2018YFA0702100 and 2018YFB0703600), the National Natural Science Foundation of China (51772012 and 51671015), Beijing Natural Science Foundation (JQ18004), Shenzhen Peacock Plan team (KQTD2016022619565991), the National Postdoctoral Program for Innovative Talents (BX20190028 and BX20200028), 111 Project (B17002) and Postdoctoral Science Foundation of China (2019M660399). L. D. Z. is thankful for the support from the National Science Fund for Distinguished Young Scholars (51925101) and the high performance computing (HPC) resources at Beihang University.

References

1. Y. Xiao and L.-D. Zhao, Science, 2020, 367, 1196–1197.
2. L.-D. Zhao, V. P. Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2014, 7, 251–268.
3. X. L. Shi, J. Zou and Z. G. Chen, Chem. Rev., 2020, 120, 7399–7515.
4. Y. M. Zhou and L. D. Zhao, Adv. Mater., 2017, 29, 1702676.
5. D. Y. Wang, Z. W. Huang, Y. Zhang, L. J. Hao, G. T. Wang, S. H. Deng, H. L. Wang, J. Chen, L. H. He, B. Xiao, Y. D. Xu, S. J. Pennycook, H. J. Wu and L. D. Zhao, Sci. China Mater., 2020, 63, 1759–1768.
6. J. F. Li, Y. Pan, C. F. Wu, F. H. Sun and T. R. Wei, Sci. China: Technol. Sci., 2017, 60, 1347–1364.
7. M. G. Kanatzidis, Chem. Mater., 2010, 22, 648–659.
8. B.-C. Qin, Y. Xiao, Y.-M. Zhou and L.-D. Zhao, Rare Met., 2017, 37, 343–350.
9. Y. Xiao and L.-D. Zhao, npj Quantum Mater., 2018, 3, 55.
10. A. T. Duong, V. Q. Nguyen, G. Duvjir, V. T. Duong, S. Kwon, J. Y. Song, J. K. Lee, J. E. Lee, S. Park, T. Min, J. Lee, J. Kim and S. Cho, Nat. Commun., 2016, 7, 13713.
11. W. K. He, D. Y. Wang, H. J. Wu, Y. Xiao, Y. Zhang, D. S. He, Y. Feng, Y. J. Hao, J. F. Dong, R. Chetty, L. J. Hao, D. F. Chen, J. F. Qin, Q. Yang, X. Li, J. M. Song, Y. C. Zhu, W. Xu, C. L. Niu, X. Li, G. T. Wang, C. Liu, M. Ohta, S. J. Pennycook, J. Q. He, J. F. Li and L.-D. Zhao, Science, 2019, 365, 1418–1424.
12. H. S. Dow, M. Na, S. J. Kim and J. W. Lee, J. Mater. Chem. C, 2019, 7, 3787–3794.
13. L. You, J. Y. Zhang, S. S. Pan, Y. Jiang, K. Wang, J. Yang, Y. Z. Pei, Q. Zhu, M. T. Agne, G. J. Snyder, Z. F. Ren, W. Q. Zhang and J. Luo, Energy Environ. Sci., 2019, 12, 3089–3098.
14. L. P. Hu, T. J. Zhu, X. H. Liu and X. B. Zhao, Adv. Funct. Mater., 2014, 24, 5211–5218.
15. J. Shuai, Y. Sun, X. J. Tan and T. Mori, Small, 2020, 16, 1906921.
16. G. Z. Xie, Z. Li, T. T. Luo, H. Bai, J. C. Sun, Y. Xiao, L.-D. Zhao, J. S. Wu, G. J. Tan and X. F. Tang, Nano Energy, 2020, 69, 104395.
17. T. J. Slade, J. A. Grovogui, J. J. Kuo, S. Anand, T. P. Bailey, M. Wood, C. Uher, G. J. Snyder, V. P. Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2020, 13, 1509–1518.
18. W. W. Qu, X. X. Zhang, B. F. Yuan and L. D. Zhao, Rare Met., 2018, 37, 79–94.
19. D. Li, H. W. Ming, J. M. Li, B. Jabar, W. Xu, J. Zhang and X. Y. Qin, ACS Appl. Mater. Interfaces, 2020, 12, 3886–3892.
20. B. Xiang, J. Liu, J. Yan, M. Xia, Q. Zhang, L. Chen, J. Li, X. Y. Tan, Q. Yan and Y. Wu, J. Mater. Chem. A, 2019, 7, 18458–18467.
21. C. Chang and L. D. Zhao, Mater. Today Phys., 2018, 4, 50–57.
22. J. Li, J. H. Sui, Y. L. Pei, C. Barreteau, D. Berardan, N. Dragoe, W. Cai, J. Q. He and L.-D. Zhao, Energy Environ. Sci., 2012, 5, 8543–8547.
23. S. N. Guin and K. Biswas, J. Mater. Chem. C, 2015, 3, 10415–10421.
24. D. Wu, Y. L. Pei, Z. Wang, H. J. Wu, L. Huang, L. D. Zhao and J. Q. He, Adv. Funct. Mater., 2014, 24, 7763–7771.
25. G. J. Tan, F. Y. Shi, S. Q. Hao, H. Chi, L.-D. Zhao, C. Uher, C. Wolverton, V. P. Dravid and M. G. Kanatzidis, J. Am. Chem. Soc., 2015, 137, 5100–5112.
26. C. J. Zhou, Y. Yu, Y. K. Lee, O. Cojocaru-Miredin, B. Yoo, S. P. Cho, J. Im, M. Wuttig, T. Hyeon and I. Chung, J. Am. Chem. Soc., 2018, 140, 15535–15545.
27. S. Sarkar, X. Zhang, S. Hao, X. Hua, T. P. Bailey, C. Uher, C. Wolverton, V. P. Dravid and M. G. Kanatzidis, ACS Energy Lett., 2018, 3, 2593–2601.
28. Y. L. Pei, C. Chang, Z. Wang, M. J. Yin, M. H. Wu, G. J. Tan, H. J. Wu, Y. X. Chen, L. Zheng, S. K. Gong, T. J. Zhu, X. B. Zhao, L. Huang, J. Q. He, M. G. Kanatzidis and L. D. Zhao, J. Am. Chem. Soc., 2016, 138, 16364–16371.
29. A. D. LaLonde, Y. Z. Pei, H. Wang and G. J. Snyder, Mater. Today, 2011, 14, 526–532.
30. L. W. Fu, J. Cui, Y. Yu, Y. Huang, Y. F. Wang, Y. Chen and J. Q. He, J. Mater. Chem. A, 2019, 7, 6304–6311.
31. X. Zhang, D. Y. Wang, H. J. Wu, M. J. Yin, Y. L. Pei, S. K. Gong, L. Huang, S. J. Pennycook, J. Q. He and L.-D. Zhao, Energy Environ. Sci., 2017, 10, 2420–2431.
32. G. J. Tan, C. C. Stoumpos, S. Wang, T. P. Bailey, L.-D. Zhao, C. Uher and M. G. Kanatzidis, Adv. Energy Mater., 2017, 7, 1700099.
33. G. J. Tan, F. Y. Shi, S. Q. Hao, L.-D. Zhao, H. Chi, X. M. Zhang, C. Uher, C. Wolverton, V. P. Dravid and M. G. Kanatzidis, Nat. Commun., 2016, 7, 12167.
34. D. Wu, L.-D. Zhao, X. Tong, W. Li, L. J. Wu, Q. Tan, Y. L. Pei, L. Huang, J. F. Li, Y. M. Zhu, M. G. Kanatzidis and J. Q. He, Energy Environ. Sci., 2015, 8, 2056–2068.
35. H. J. Wu, L.-D. Zhao, F. S. Zheng, D. Wu, Y. L. Pei, X. Tong, M. G. Kanatzidis and J. Q. He, Nat. Commun., 2014, 5, 4515.
36. S. Q. Wang, Z. Yang, Y. J. Sun, Y. Xiao and L.-D. Zhao, J. Alloys Compd., 2020, 815, 152463.
37. K. Q. Zhang, H. C. Wang, W. B. Su, T. Wang, X. Wang, T. T. Chen, T. C. Huo, F. Dang, M. Y. Dong, C. L. Wang, B. B. Dong and Z. H. Guo, J. Phys. D: Appl. Phys., 2020, 53, 245501.
38. Q. Zhang, Q. C. Song, X. Y. Wang, J. Y. Sun, Q. Zhu, K. Dahal, X. Lin, F. Cao, J. W. Zhou, S. Chen, G. Chen, J. Mao and Z. F. Ren, Energy Environ. Sci., 2018, 11, 933–940.
39. Y. Xiao, H. J. Wu, D. Y. Wang, C. L. Niu, Y. L. Pei, Y. Zhang, I. Spanopoulos, I. T. Witting, X. Li, S. J. Pennycook, G. J. Snyder, M. G. Kanatzidis and L.-D. Zhao, Adv. Energy Mater., 2019, 9, 1900414.
40. X. L. Su, S. Q. Hao, T. P. Bailey, S. Wang, I. Hadar, G. J. Tan, T.-B. Song, Q. J. Zhang, C. Uher, C. Wolverton, X. F. Tang and M. G. Kanatzidis, Adv. Energy Mater., 2018, 8, 1800659.
41. Z. S. Wang, G. Y. Wang, R. F. Wang, X. Y. Zhou, Z. Y. Chen, C. Yin, M. J. Tang, Q. Hu, J. Tang and R. Ang, ACS Appl. Mater. Interfaces, 2018, 10, 22401–22407.
42. Y. Xiao, D. Wang, B. Qin, J. Wang, G. Wang and L.-D. Zhao, J. Am. Chem. Soc., 2018, 140, 13097–13102.
43. Y. Xiao, D. Wang, Y. Zhang, C. Chen, S. Zhang, K. Wang, G. Wang, S. J. Pennycook, G. J. Snyder, H. Wu and L.-D. Zhao, J. Am. Chem. Soc., 2020, 142, 4051–4060.
44. P. K. Rawat, B. Paul and P. Banerji, ACS Appl. Mater. Interfaces, 2014, 6, 3995–4004.
45. K. Ahn, M. K. Han, J. Q. He, J. Androulakis, S. Ballikaya, C. Uher, V. P. Dravid and M. G. Kanatzidis, J. Am. Chem. Soc., 2010, 132, 5227–5235.
46. K. Biswas, J. Q. He, G. Y. Wang, S. H. Lo, C. Uher, V. P. Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2011, 4, 4675–4684.
47. S. N. Girard, T. C. Chasapis, J. Q. He, X. Y. Zhou, E. Hatzikraniotis, C. Uher, K. M. Paraskevopoulos, V. P. Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2012, 5, 8716–8725.
48. S. Ballikaya, H. Chi, J. R. Salvador and C. Uher, J. Mater. Chem. A, 2013, 1, 12478–12484.
49. S. Mukherjee, R. Parasuraman, A. M. Umarji, G. Rogl, P. Rogl and K. Chattopadhyay, J. Alloys Compd., 2020, 817, 152729.
50. Z.-Z. Luo, X. M. Zhang, X. Hua, G. J. Tan, T. P. Bailey, J. W. Xu, C. Uher, C. Wolverton, V. P. Dravid, Q. Y. Yan and M. G. Kanatzidis, Adv. Funct. Mater., 2018, 28, 1801617.
51. K. M. Zhang, Q. H. Zhang, L. J. Wang, W. Jiang and L. D. Chen, J. Alloys Compd., 2017, 725, 563–572.
52. P. Jood, M. Ohta, M. Kunii, X. Hu, H. Nishiate, A. Yamamoto and M. G. Kanatzidis, J. Mater. Chem. C, 2015, 3, 10401–10408.
53. Y. Xiao, H. J. Wu, J. Cui, D. Y. Wang, L. W. Fu, Y. Zhang, Y. Chen, J. Q. He, S. J. Pennycook and L.-D. Zhao, Energy Environ. Sci., 2018, 11, 2486–2495.
54. G. J. Tan, X. M. Zhang, S. Q. Hao, H. Chi, T. P. Bailey, X. L. Su, C. Uher, V. P. Dravid, C. Wolverton and M. G. Kanatzidis, ACS Appl. Mater. Interfaces, 2019, 11, 9197–9204.
55. Y. Z. Pei, A. F. May and G. J. Snyder, Adv. Energy Mater., 2011, 1, 291–296.
56. Y. X. Qin, Y. Xiao and L.-D. Zhao, APL Mater., 2020, 8, 010901.
57. Y. Xiao, H. J. Wu, W. Li, M. J. Yin, Y. L. Pei, Y. Zhang, L. W. Fu, Y. X. Chen, S. J. Pennycook, L. Huang, J. He and L.-D. Zhao, J. Am. Chem. Soc., 2017, 139, 18732–18738.
58. Y. Z. Pei, J. Lensch-Falk, E. S. Toberer, D. L. Medlin and G. J. Snyder, Adv. Funct. Mater., 2011, 21, 241–249.
59. J. Zhang, D. Wu, D. S. He, D. Feng, M. J. Yin, X. Y. Qin and J. Q. He, Adv. Mater., 2017, 29, 1703148.
60. H. S. Kim, W. S. Liu, G. Chen, C. W. Chua and Z. F. Ren, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 8205–8210.
61. L.-D. Zhao, H. J. Wu, S. Q. Hao, C. I. Wu, X. Y. Zhou, K. Biswas, J. Q. He, T. P. Hogan, C. Uher, C. Wolverton, V. P. Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2013, 6, 3346–3355.
62. Y. L. Pei and Y. Liu, J. Alloys Compd., 2012, 514, 40–44.
63. A. D. LaLonde, Y. Z. Pei and G. J. Snyder, Energy Environ. Sci., 2011, 4, 2090–2096.
64. H. Sun, B. W. Cai, P. Zhao, F. R. Yu, L. Zhang, D. L. Yu, Y. J. Tian and B. Xu, J. Alloys Compd., 2019, 791, 786–791.
65. H. T. Liu, Z. Y. Chen, C. Yin, B. Q. Zhou, B. Liu and R. Ang, Appl. Phys. A: Mater. Sci. Process., 2019, 125, 225.
66. B. C. Qin, Y. Zhang, D. Y. Wang, Q. Zhao, B. C. Gu, H. J. Wu, H. J. Zhang, B. J. Ye, S. J. Pennycook and L.-D. Zhao, J. Am. Chem. Soc., 2020, 142, 5901–5909.
67. X. Qian, H. J. Wu, D. Y. Wang, Y. Zhang, J. F. Wang, G. T. Wang, L. Zheng, S. J. Pennycook and L.-D. Zhao, Energy Environ. Sci., 2019, 12, 1969–1978.
68. B. Qin, D. Wang, W. He, Y. Zhang, H. Wu, S. J. Pennycook and L. D. Zhao, J. Am. Chem. Soc., 2019, 141, 1141–1149.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d0ee03459b

‡ These authors contributed equally to this work.

---