Impurity band engineering and hierarchical defect scattering enable high zT in Co-doped AgSbTe₂

Abstract

AgSbTe₂ is a promising p-type thermoelectric material for mid-temperature applications. However, its performance is limited by its low electrical conductivity and the presence of an n-type Ag₂Te secondary phase. In this study, a series of AgSb_1−xCo_xTe samples were synthesized by conventional melting followed by spark plasma sintering. Co doping suppresses the formation of the detrimental Ag₂Te phase and introduces an impurity band above the valence band maximum, as revealed by the calculated electronic structures. Moreover, Co incorporation generates multi-scale lattice defects that act as efficient phonon scattering centers, resulting in remarkably low thermal conductivity. Consequently, a peak thermoelectric figure of merit of 1.6 was achieved at 648 K. These results demonstrate a practical doping strategy that preserves favorable charge transport in AgSbTe₂ while strongly reducing its thermal conductivity, offering a clear pathway toward improving mid-temperature thermoelectric modules.

---

Broader context

Thermoelectric materials can directly convert waste heat into electrical energy, offering a scalable and maintenance-free pathway for decarbonizing medium-temperature industrial and automotive processes. Among p-type compounds operating in this temperature range, AgSbTe₂ stands out due to its inherently high Seebeck coefficient and favorable band structure. However, its practical application is constrained by low electrical conductivity and the formation of parasitic n-type Ag₂Te secondary phases, which degrade charge transport and overall device performance. To address this challenge, we synthesized cobalt-doped AgSbTe₂ using conventional melt-bonded spark plasma sintering technology. Co doping effectively suppresses the formation of the detrimental Ag₂Te secondary phase while inducing favorable electronic properties and microstructural modifications. This approach significantly reduces lattice thermal conductivity while maintaining electrical transport performance. Ultimately, a peak thermoelectric figure of merit of 1.6 was achieved at 648 K, providing a practical and efficient doping strategy for enhancing the performance of medium-temperature thermoelectric devices.

Introduction

The growing global demand for energy, combined with the depletion of fossil fuel reserves, calls for clean, efficient, and renewable energy sources.^1–3 Thermoelectric (TE) technology can directly convert heat into electric power and electricity back into heat, providing an environment-friendly route for utilizing wasted energy.^4–6 TE generators can be used in vehicle waste heat recovery, industrial energy utilization, space exploration, wearable devices and sensors due to their reliability, stability, and suitability in different environments.^7–9 Meanwhile, TE coolers, have been applied in small devices, electronics, and building cooling due to their compact size and precise control of temperature.^10–12 The widespread application of thermoelectric devices across various fields stems from their simple design, reliable operation, low noise levels, and minimal maintenance requirements.¹³ However, their relatively low conversion efficiency and high cost limit their practical application.¹⁴

The key measure of a material's thermoelectric performance is the dimensionless figure of merit (zT):

zT = S²σT/κ (1)

where S is the Seebeck coefficient, σ is the electrical conductivity, κ is the total thermal conductivity (lattice thermal conductivity, κ_l, plus electronic thermal conductivity, κ_e), and T is the absolute temperature.¹⁵ To achieve a high zT, a material needs a high power factor (PF = S²σ) together with a low κ.¹⁶ In practice, while increasing carrier concentration enhances σ, it often leads to a decrease in S. Methods such as introducing nanostructures or defects can scatter phonons and reduce κ_l but inhibit carrier mobility.¹⁷ Balancing these competing effects to produce materials with high S, high σ, and low κ remains the central challenge in thermoelectric material development.¹⁸

TE materials achieve their peak zT at different temperature ranges, making them suitable for various applications. For example, Bi₂Te₃ materials usually have a zT of 1 to 1.9 between 300 K and 400 K, which makes them commonly used in commercial thermoelectric devices.^19–22 Half-Heusler alloys,^23–26 skutterudites,^27,28 and some oxides^29,30 are high-temperature thermoelectric materials used in space exploration and industrial waste-heat recovery. In the mid-temperature range, PbTe,^31–33 GeTe,^34–39 SnSe,^40–43 and argyrodite compounds^44,45 can achieve peak zT values of over 2.5, making them ideal for waste heat recovery from vehicles and industrial furnaces.

Among mid-temperature TE materials, AgSbTe₂ is an intrinsic p-type material which has low κ and potentially achieves high zT in the temperature range from 500 to 700 K.^46–49 AgSbTe₂ exhibits an ultra-low κ_l, resulting from the synergistic interaction between its inherent cation disorder, strong anharmonicity and dynamic Ag⁺ diffusion. The former strongly scatters mid-to-high-frequency phonons, while the latter induces a phonon liquid-like state, thereby suppressing heat conduction across all frequency ranges.^50–56 AgSbTe₂ also has a high S coming from a narrow band gap and high intrinsic carrier mobility.^57,58 However, AgSbTe₂ is prone to phase decomposition, forming an n-type Ag₂Te secondary phase. Its relatively low σ also limits its performance.

Various composition and microstructure engineering methods have significantly improved the performance of AgSbTe₂. Cation alloying can improve atomic ordering within the cation sublattice;^59–62 for example, Cd-doped AgSbTe₂ achieved a peak zT value of 2.6 at 573 K by reducing its atomic disorder.⁶³ Anion alloying on the Te sublattice, such as co-alloying of Se and S, has been shown to manipulate the band structure, achieving a peak zT of 2.3 at 673 K in AgSbTe_1.9Se_0.05S_0.05.^64,65 In addition to doping technology, thermal processing is also an important method to improve material performance.^66,67 For example, thermal cycling was reported to induce a metastable state and promote a uniformly distributed secondary phase.⁶⁸ Moreover, alloying AgSbTe₂ with PbTe, GeTe, or SnTe matrices can form nanostructured secondary phases to selectively scatter mid- to long-wavelength phonons, thereby synergistically reducing κ of the composite material.^69–73

Herein, we synthesized AgSb_1−xCo_xTe (0 ≤ x ≤ 0.04) samples using a two-step process combining conventional melting with subsequent spark plasma sintering (SPS). A maximum zT of 1.6 at 648 K was achieved in the AgSb_0.97Co_0.03Te₂ sample. Co doping significantly reduces its κ_l while maintaining the power factor. Detailed investigations of their microstructure and thermoelectric properties revealed that the incorporation of Co effectively suppresses the formation of Ag₂Te and facilitates an impurity band above the valence band, leading to increased electronic transport properties (Fig. 1a and b). In addition, the AgSb_0.97Co_0.03Te₂ sample has a low κ_l, which is associated with multiscale phonon scattering centers. Co doping leads to the formation of secondary phases, holes, dislocations, twin boundaries, and nanoscale superstructures in AgSb_0.97Co_0.03Te₂, which act as phonon scattering centers at the microscopic scale (Fig. 1c). This led to a low κ_l value and an improved zT (Fig. 1d).

Fig. 1 Overview of the conceptual framework of enhancing thermoelectric performance. (a) Crystal structure of Co-doped AgSbTe₂. (b) Schematic illustration of the electronic band structure modulation induced by Co doping. (c) Schematic of the micro- and nanostructure features in the doped sample, including grain boundaries, pores, secondary phases, and nanoscale superstructures that promote phonon scattering and reduce κ_l. (d) Comparison of κ_l and zT between pristine AgSbTe₂ and Co-doped AgSbTe₂ samples.

Experimental section

Synthesis

AgSb_1−xCo_xTe₂ (x = 0, 0.01, 0.02, 0.03, and 0.04) samples were prepared via a melting method, followed by spark plasma sintering (SPS). High-purity elemental Ag (99.99%), Sb (99.99%), Co (99.99%), and Te (99.99%) were weighed according to their nominal compositions, loaded into evacuated quartz ampoules, and melted in a muffle furnace. The furnace was heated from room temperature to 1273 K for 10 h, held at this temperature for 6 h to ensure complete homogenization, and then cooled to room temperature for 20 h. The ingots were manually ground into fine powders using an agate mortar and a pestle. The powders were subsequently consolidated into disks that were 12 mm in diameter by SPS at 673 K under an applied pressure of 35 MPa. During the SPS process, the sample was first heated to 623 K in 10 min and subsequently heated to 673 K in 5 min. The SPS cycle involved ramping to 623 K in 10 min, then to 673 K in 5 min, and finally holding at 673 K for 2 min. The sintered AgSb_0.97Co_0.03Te₂ pellet has a diameter of 11.87 mm, a height of 1.46 mm, and a density of 6.81 g cm⁻³, corresponding to a relative density exceeding 95%.

Characterization

The phases of the samples were determined using X-ray diffraction (XRD; D8 ADVANCE powder X-ray diffractometer, Bruker) and differential scanning calorimetry (DSC; Q100, TA Instruments). The micro- and nanostructures of the samples were investigated using a scanning electron microscope (JCM-7000 NeoScope, JEOL) and a field-emission transmission electron microscope (HF5000 TEM, Hitachi).

TE transport properties

σ and S were measured simultaneously using the standard four-probe method in a nitrogen atmosphere (SBA 467 HT, NETZSCH). κ was calculated using κ = λC_pd, where λ, the thermal diffusivity, was obtained using the laser flash diffusivity method (LFA 458 HyperFlash, NETZSCH), C_p, the specific heat, was determined from a reported paper,⁷⁴ and d, the density, was obtained by calculation.

DFT calculations

The structure optimization and energy band structure were calculated within the framework of density functional theory (DFT) using the generalized gradient approximation (GGA)⁷⁵ Perdew–Burke–Ernzerhof (PBE) exchange and correlation functional.⁷⁶ The pseudopotential projector augmented wave (PAW) method⁷⁷ was implemented in the Vienna Ab initio Simulation Package (VASP).⁷⁸ All calculations were based on the AF–IIb (F3̄m) phase, containing 64 atoms in a cubic cell with the lattice parameter a = 12.333 Å. For the Co-doped compositions, one Sb atom was substituted with Co, corresponding to the intended stoichiometries. The lattice parameters and atomic coordination numbers were fully relaxed under the following conditions: the residual force for ion convergence was 2 × 10⁻² eV Å⁻¹, and the total energy converged to 10⁻⁸ eV Å⁻¹. The cut-off energy of the plane-wave basis set was 600 eV. Brillouin zone (BZ) integrations were sampled on a Γ-center Monkhorst–Pack 4 × 4 × 4 mesh of k-points in the BZ of pristine AgSb_1−xCo_xTe₂. We calculated the electron spectrum at the Bloch wave vectors along the high-symmetry lines (Γ − X − M − Γ − R − X − M − R) in the Brillouin zone.

Results and discussion

Fig. 2a shows the X-ray diffraction (XRD) patterns of the AgSb_1−xCo_xTe₂ samples, which conform to the rock-salt AgSbTe₂ structure (PDF #15-0540; space group: Fm3̄m). For compositions with x < 0.03, minor Ag₂Te impurity peaks (PDF #34-0142) are observed (Fig. 2b). As an intrinsically n-type material, the Ag₂Te secondary phase acts as a compensating dopant in the p-type AgSbTe₂ matrix. It introduces excess electrons that recombine with and neutralize the dominant hole carriers, thereby directly reducing the hole concentration (n) and σ of the composite system.⁷⁹ Additionally, the phase change of Ag₂Te at 418 K significantly affects the σ of AgSbTe₂. Suppressing the formation of the Ag₂Te secondary phase is an important way to improve the performance of AgSbTe₂.⁸⁰ Therefore, suppressing the Ag₂Te formation is crucial for improving the performance of AgSbTe₂. Upon increasing the Co content, the Ag₂Te peaks progressively disappear entirely at x = 0.03 and 0.04, demonstrating that Co incorporation effectively inhibits the formation of Ag₂Te. However, at these higher doping levels (x = 0.03 and 0.04), additional reflections corresponding to CoTe₂ and Ag₅Te₃ emerge, signifying the onset of phase decomposition and indicating that the solubility limit of Co in the AgSbTe₂ matrix has been reached. The suppression of the Ag₂Te secondary phase leads to an increase in σ, while the formed CoTe₂ and Ag₅Te₃ secondary phases act as phonon scattering centers, lowering the κ.

Fig. 2 XRD and DSC characterization of AgSb_1−xCo_xTe₂ samples. XRD patterns (a) and enlarged peaks (b) of AgSb_1−xCo_xTe₂ (x = 0, 0.01, 0.02, 0.03, 0.04). (c) Lattice parameter as a function of Co (x) concentration of AgSb_1−xCo_xTe₂ samples. (d) DSC curves of AgSb_1−xCo_xTe₂ samples.

Fig. 2c plots the lattice parameters extracted from the XRD refinements of AgSb_1−xCo_xTe₂. The pristine AgSbTe₂ exhibits a lattice parameter of approximately 6.079 Å, matching literature reports. As Co content increases, the unit cell slightly increases until x reaches 0.03. As the ionic radius of Co²⁺ is larger than that of Sb³⁺ but smaller than that of Ag⁺, the increasing lattice parameter indicates the Co substitution of Sb. However, at x = 0.03 and x = 0.04, the lattice parameter slightly contracts, coinciding with the emergence of CoTe₂ precipitates and indicating the reach of the Co solubility limit in AgSbTe_2.

Fig. 2d shows the DSC traces for AgSb_1−xCo_xTe₂. The pristine AgSbTe₂ has a pronounced endothermic peak at 422 K, corresponding to the phase transition from high-symmetry cubic α-Ag₂Te to low-symmetry monoclinic β-Ag₂Te. The phase transition in Ag₂Te induces severe lattice distortion, which not only generates microcracks in the AgSbTe₂ compound but also creates a high density of scattering centers, thus dramatically increasing the scattering rate for both electrons and holes and consequently diminishing its overall σ. As the Co concentration increases, the 422 K peak progressively diminishes and is almost entirely absent for x = 0.03 and 0.04. The disappearance of this thermal signature in DSC confirms the XRD conclusion and provides complementary evidence that Co doping suppresses the formation and phase transition of Ag₂Te. The disappearance of the characteristic Ag₂Te peaks in both XRD patterns and DSC traces definitively proves that Co doping inhibits the formation of this detrimental n-type secondary phase across the critical composition range (x ≥ 0.03).

To examine the band structure evolution with the Co content in AgSb_1−xCo_xTe₂ and disclose the fundamental reasons for the enhanced performance, we performed DFT calculations. Fig. 3a depicts the calculated band structure of pristine AgSbTe₂, which exhibits a pronounced pseudo band gap characterized by notably flat valence bands, in agreement with earlier reports.⁸¹ In the highlighted VB1, VB2, and VB3, VB2 and VB3 are particularly flat, while VB1 is sharp. Such flatness near the maximum value of the valence band is the basis for the large S value observed in pure AgSbTe₂. Fig. 3b shows the electronic band structure of Co-doped AgSbTe₂. It shows a clear convergence of the valence and conduction bands at the M point, which is because of the over-doping of Co. The model has a higher Co content of 0.0625 than the solubility of about 0.03 obtained from experiments. More importantly, at the R point, Co incorporation drives an upward shift in the valence band maximum (VBM) toward the conduction band minimum (CBM). DFT calculations reveal that this shift originates from the formation of Co-induced impurity states near the valence band edge. The implications of these states are significant: the resulting band-gap narrowing increases the hole carrier concentration, which directly accounts for the increase in σ observed in the experiments. Moreover, VB3 has almost completely merged with VB2, which is expected to increase the effective mass and thus enhance the S. However, experimental data for Co-doped AgSbTe₂ show a reduction in S. This may be related to the significant increase in hole concentration induced by gap narrowing, which strongly suppresses S. Additionally, Co doping can form strong scattering centers, leading to reduced carrier mobility. As shown in Fig. S3, the calculated weighted mobility (µ_w), which remains significantly reduced upon doping, further indicates that the overall carrier transport quality is not markedly improved. Together, these factors explain why the anticipated benefit in S from band convergence is offset, resulting in the observed net decrease in S.

Fig. 3 DFT calculations of the band structure and density of states. DFT-calculated band structures of (a) Ag₁₆Sb₁₆Te₃₂ and (b) Ag₁₆Sb₁₅Co₁Te₃₂. DOS of (c) Ag₁₆Sb₁₆Te₃₂ and (d) Ag₁₆Sb₁₅Co₁Te₃₂.

Fig. 3c and d present the total and projected densities of states (DOSs) for pristine and Co-doped AgSbTe₂, respectively. Co incorporation introduces additional holes, strengthening the p-type conduction of AgSbTe₂ and leading to a downward shift in the Fermi level. Additionally, the broadened DOS peaks in the Co-doped samples indicate reduced electron localization, resulting from the hybridization between the Co 3d and the host p/s orbitals.

Fig. 4 shows the SEM images of as-sintered AgSbTe₂ and AgSb_0.97Co_0.03Te₂. The pure AgSbTe₂ sample had a dense surface (Fig. 4a), whereas the Co-doped sample had numerous pores on its surface (Fig. 4d). These large pores with irregular shapes indicate that they originated from secondary phases pulled out during the polishing process. The large pores with irregular shapes in the Co-doped sample are attributed to secondary phases that were pulled out during the polishing process. Co-doping suppresses the formation of Ag₂Te and causes the precipitation of Co–Sb–Te phases. These Co–Sb–Te phases likely have a considerable lattice mismatch and form weakly bonded interfaces with the AgSbTe₂ matrix, which makes them easy to detach, thus creating pores. In contrast, the Ag₂Te secondary phases in the pure AgSbTe₂ sample exhibit only a small lattice mismatch with the matrix. This strong interfacial cohesion makes them difficult to be pulled out during polishing, which explains why no significant pores are observed in the undoped material. The small pores have sizes ranging from 1 to 3 µm, which can act as effective phonon scattering centers, thereby contributing to the reduction of κ_l.

Fig. 4 SEM analysis of pristine and AgSb_0.97Co_0.03Te₂ samples. (a) Backscattered electron (BSE) image, (b) its enlarged image, and (c) EDS mapping of pristine AgSbTe₂. (d) BSE image, (e) its enlarged image, and (f) EDS mapping of pristine AgSb_0.97Co_0.03Te₂. The holes and EDS analysis regions are pointed out in yellow.

Fig. 4b and c are the enlarged EDS elemental maps of the pure sample, which indicate the presence of a Ag₂Te secondary phase. The element distribution of AgSb_0.97Co_0.03Te₂ is shown in Fig. 4e and f. It can be seen in Fig. 4e that region 1 is poor in Ag and rich in Co, while region 2 is rich in Ag and poor in Sb. Based on the quantitative EDS data for regions 1–3 in Table S1, region 1 is mainly a Co–Sb–Te phase. Its composition is close to that of CoSb_0.5Te_1.5, and its peaks match those of CoTe₂ in the XRD pattern. To better understand this phase, we performed Rietveld refinement on the Co-doped AgSbTe₂ sample. We used two models, one for pure CoTe₂ and another for the non-stoichiometric CoSb_0.5Te_1.5, as suggested by EDS. The refinement confirms that the CoSb_0.5Te_1.5 model fits the data better. With this model, the lattice parameters change as expected, and the b-axis expands clearly from 3.817 A to 3.883 A. This expansion occurs because larger Sb³⁺ ions partly replace Te²⁻ ions, providing clear structural proof of the Sb-rich composition found by EDS. Overall, the Co–Sb–Te secondary phase likely has a composition of CoSb_0.5Te_1.5, which is consistent with the Co–Sb–Te phase reported in the study of Ag–Sb–Te systems.⁸² The composition of region 2 deviates from Ag₂Te and instead aligns more closely with Ag₅Te₃. This observation is consistent with the suppression of Ag₂Te phases, as indicated by the XRD and DSC results. The microprecipitates can pin grain boundaries and increase the grain boundary density, thereby enhancing phonon scattering and reducing κ_l.

The sub-micro- and nanostructures of the AgSb_0.97Co_0.03Te₂ sample are characterized using TEM. The TEM images and related EDS maps in Fig. 5a and b indicate the presence of Ag-rich and Sb-rich secondary phases at the sub-micro-scale. The TEM image in Fig. 5c displays a polycrystalline morphology with grain sizes ranging from 2 µm down to sub-micrometers, indicating a broad size distribution of the crystals. In Fig. 5d, the high-resolution TEM image and the selected area electron diffraction (SAED) pattern show clear lattice fringes along the [111] direction, confirming the highly crystalline and long-range structure order in the matrix. To investigate sub-micro- and nanoscale structures in detail, several regions marked in Fig. 5c were examined under high magnification. The enlarged images of region 1 (Fig. 5e) indicate the twin boundaries and pinning dislocations along the [110] direction, which can act as efficient scattering centers for mid-to low-frequency phonons. The enlarged image of region 2 (Fig. 5f) shows clear boundaries between grains. The interfaces cause abrupt changes in elastic properties and lattice periodicity, thereby scattering long-wavelength (low-frequency) phonons.⁸³ The enlarged images of region 3 (Fig. 5g–i) show nanoscale superstructures in grains, complex interfaces between adjacent grains, and dislocations within individual grains. Nanoscale superstructures effectively scatter high-frequency phonons. The combination of different defects scatters phonons across a broad frequency range, contributing to the low κ_l of the AgSb_0.97Co_0.03Te₂ sample.

Fig. 5 TEM images of AgSb_0.97Co_0.03Te₂. (a) TEM and (b) related EDS images of the AgSb_0.97Co_0.03Te₂ sample at a sub-micro-scale. (c) TEM image of the AgSb_0.97Co_0.03Te₂ sample; the yellow squares are the enlarged regions of the following images. (d) Atomic-resolution HAADF-STEM image obtained from the AgSb_0.97Co_0.03Te₂ sample and its corresponding SAED along the [111] direction. (e) HRTEM micrograph of AgSb_0.97Co_0.03Te₂ grains in region 1, where the yellow lines show the pinning dislocations. The inset shows the corresponding SAED pattern along the [111] direction. (f) HRTEM image of region 2, the yellow line displays the lattice fringes of the two grains. The HRTEM images of region 3 show the nanoscale superstructure (g), grain boundaries (h), and dislocation (i) in yellow.

The TE performance of the sintered AgSb_1−xCo_xTe₂ pellets over the full thermal cycle (heating and cooling) is summarized in Fig. 6. Fig. 6a presents the temperature dependence of σ for each composition. For pristine AgSbTe₂, σ is 225 S cm⁻¹ at 298 K, which decreases slightly until the Ag₂Te phase transition at 418 K, then remains roughly constant at about 180 S cm⁻¹ up to the second transition at 633 K, before recovering to 225 S cm⁻¹ at 648 K. Co incorporation elevates σ across the whole temperature range. The AgSb_0.97Co_0.03Te₂ sample exhibits σ = 325 S cm⁻¹ at 298 K, which declines to about 275 S cm⁻¹ at 418 K and to about 240 S cm⁻¹ at 673 K during heating but reaches a peak of 350 S cm⁻¹ at 623 K on cooling. This is related to the suppression of the Ag₂Te secondary phase and the introduction of an impurity band.

Fig. 6 Thermoelectric properties of the AgSb_0.97Co_0.03Te₂ samples during heating and cooling cycles. (a) σ, (b) S, (c) PF, (d) κ, (e) κ_l, and (f) zT as a function of temperature.

Fig. 6b shows the S versus temperature plot. Pristine AgSbTe₂ displays an S of approximately 200 µV K⁻¹ at 298 K, increasing to approximately 250 µV K⁻¹ at 623 K before dropping to approximately 225 µV K⁻¹ at 648 K. In contrast, Co doping reduces S sharply to below 125 µV K⁻¹ at room temperature, with only a slight decrease at elevated temperatures (approximately 220 µV K⁻¹ at 673 K), confirming the merging VB. Specifically, the x = 0.03 sample shows an S of approximately 80 µV K⁻¹ at 298 K, increasing to approximately 220 µV K⁻¹ at 648 K.

Fig. 6c shows the resulting PF values. At x = 0.02, the material exhibits maximum PFs of 7 µW cm⁻¹ K⁻² at 289 K and 15 µW cm⁻¹ K⁻² at 623 K. Other Co-doped samples exhibited PF close to the pristine material level, below 4 µW cm⁻¹ K⁻² at room temperature and increasing to approximately 12 µW cm⁻¹ K⁻² at 623 K. The doped samples exhibited behavior similar to the pristine material, with a PF below 4 µW cm⁻¹ K⁻² at room temperature, increasing to about 12 µW cm⁻¹ K⁻² at 648 K.

Fig. 6d plots the total κ of AgSb_1−xCo_xTe₂ as a function of temperature. The pristine AgSbTe₂ sample exhibits a nearly constant κ of about 0.7 W m⁻¹ K⁻¹ between 298 K and 648 K, with a slight uptick to about 0.8 W m⁻¹ K⁻¹ at 623 K during the cooling cycle. Doping levels of x = 0.01 and x = 0.02 produce negligible changes in κ, whereas the x = 0.03 composition displays a significant reduction: κ is approximately 0.50 W m⁻¹ K⁻¹ at 298 K, rising modestly to approximately 0.60 W m⁻¹ K⁻¹ at 573 K before decreasing to ≈0.45 W m⁻¹ K⁻¹ at 648 K. Fig. 6e displays κ_l components. AgSb_0.97Co_0.03Te₂ has a low κ_l, which is near 0.30 W m⁻¹ K⁻¹ between 289 K and 573 K, decreasing further to approximately 0.20 W m⁻¹ K⁻¹ above 600 K, and dropping to about 0.10 W m⁻¹ K⁻¹ at 648 K on cooling. By comparison, pristine AgSbTe₂ displays κ_l = 0.59 W m⁻¹ K⁻¹ at 289 K and 0.42 W m⁻¹ K⁻¹ at 648 K. This pronounced suppression of κ_l in the AgSb_0.97Co_0.03Te₂ sample can be understood as a combination of intrinsic and extrinsic effects. The AgSbTe₂ matrix possesses intrinsically low κ_l due to its characteristic cationic disorder and anharmonic lattice dynamics.^84,85 Co doping further reduces κ_l by creating a spectrum of scattering centers that target phonons of different mean free paths. Point defects arise from atomic substitution and enhanced disorder at the Co doping sites. Nanoscale secondary phases, including CoTe₂ and Ag₅Te₃, effectively scatter mid-frequency phonons. Additionally, the microstructure revealed by SEM and TEM shows increased porosity and defects at different scale, which are particularly effective for scattering long-wavelength phonons. The collective action of these scattering mechanisms across multiscales is responsible for driving κ_l to the observed low values.

Finally, Fig. 6f presents zT as a function of temperature. When x = 0.03, the total κ value decreases significantly while maintaining the electrical properties, enabling the sample to achieve a maximum zT value of 1.6 at 648 K. In contrast, the zT value of pristine AgSbTe₂ at the same temperature is only 1.2.

Conclusion

AgSb_1−xCo_xTe₂ (0 ≤ x ≤ 0.04) samples were synthesized by conventional melting followed by spark plasma sintering, and their microstructures and transport properties were systematically evaluated. The incorporation of Co effectively suppresses the formation of the Ag₂Te precipitate and manipulates the band structure, maintaining the electrical properties. The introduction of multiscale defects and secondary phases significantly enhances phonon scattering, leading to a substantially reduced κ_l of 0.45 W m⁻¹ K⁻¹ at 648 K. The AgSb_0.97Co_0.03Te₂ composition achieves the lowest κ and a resulting peak zT of 1.6 at 648 K. These results highlight the critical role of κ optimization via controlled phase formation in advancing the performance of AgSbTe₂-based thermoelectric materials.

Conflicts of interest

The authors declare no conflict of interest.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5ta09240j.

Acknowledgements

The authors acknowledge the financial support from the Australian Research Council, including the ARC LIEF grant (No. LE230100179), and the iLAuNCH Trailblazer Program.

References

1. X. L. Shi, J. Zou and Z. G. Chen, Chem. Rev., 2020, 120, 7399–7515.
2. Q. Zhang, K. Deng, L. Wilkens, H. Reith and K. Nielsch, Nat. Electron., 2022, 5, 333–347.
3. T. Lu, L. Chen, X.-L. Shi, W. Liu, M. Li, Y. Wang, X. Zeng, P. Song, J. Bell, G. J. Snyder, Z.-G. Chen and M. Hong, Small Struct., 2025, 6, 2400694.
4. Z. Wu, S. Zhang, Z. Liu, E. Mu and Z. Hu, Nano Energy, 2022, 91, 106692.
5. H. Han, L. Zhao, X. Wu, B. Zuo, S. Bian, T. Li, X. Liu, Y. Jiang, C. Chen, J. Bi, J. Xu and L. Yu, J. Mater. Chem. A, 2024, 12, 24041–24083.
6. L. Chen, T. Lu, X.-L. Shi, W.-D. Liu, M. Li, S. Huo, P. Song, J. Bell, Z.-G. Chen and M. Hong, Appl. Phys. Rev., 2025, 12, 041322.
7. F. Tohidi, S. Ghazanfari Holagh and A. Chitsaz, Appl. Therm. Eng., 2022, 201, 117793.
8. H. Jouhara, A. Żabnieńska-Góra, N. Khordehgah, Q. Doraghi, L. Ahmad, L. Norman, B. Axcell, L. Wrobel and S. Dai, Int. J. Thermofluids, 2021, 9, 100063.
9. H. Ma, T. Lu, X.-L. Shi, M. Li, S. Huo, P. Song, Z.-G. Chen and M. Hong, Prog. Mater. Sci., 2026, 158, 101619.
10. J. Mao, G. Chen and Z. Ren, Nat. Mater., 2021, 20, 454–461.
11. W.-Y. Chen, X.-L. Shi, J. Zou and Z.-G. Chen, Small Methods, 2022, 6, 2101235.
12. J. Choi, C. Dun, C. Forsythe, M. P. Gordon and J. J. Urban, J. Mater. Chem. A, 2021, 9, 15696–15703.
13. Z. Soleimani, S. Zoras, B. Ceranic, S. Shahzad and Y. Cui, Sustain. Energy Technol. Assessments, 2020, 37, 100604.
14. Y. Xiao and L.-D. Zhao, Science, 2020, 367, 1196–1197.
15. W. Chen, X.-L. Shi, M. Li, T. Liu, Y. Mao, Q. Liu, M. Dargusch, J. Zou, G. Q. Lu and Z.-G. Chen, Science, 2024, 386, 1265–1271.
16. Y. Pei, X. Shi, A. LaLonde, H. Wang, L. Chen and G. J. Snyder, Nature, 2011, 473, 66–69.
17. M. Zebarjadi, K. Esfarjani, M. Dresselhaus, Z. Ren and G. Chen, Energy Environ. Sci., 2012, 5, 5147–5162.
18. Z. Ma, J. Wei, P. Song, M. Zhang, L. Yang, J. Ma, W. Liu, F. Yang and X. Wang, Mater. Sci. Semicond. Process., 2021, 121, 105303.
19. J. Pei, B. Cai, H.-L. Zhuang and J.-F. Li, Natl. Sci. Rev., 2020, 7, 1856–1858.
20. Q. Shi, J. Li, X. Zhao, Y. Chen, F. Zhang, Y. Zhong and R. Ang, ACS Appl. Mater. Interfaces, 2022, 14, 49425–49445.
21. Y. Lu, Y. Zhou, W. Wang, M. Hu, X. Huang, D. Mao, S. Huang, L. Xie, P. Lin, B. Jiang, B. Zhu, J. Feng, J. Shi, Q. Lou, Y. Huang, J. Yang, J. Li, G. Li and J. He, Nat. Nanotechnol., 2023, 18, 1281–1288.
22. D.-W. Ao, W.-D. Liu, Y.-X. Chen, M. Wei, B. Jabar, F. Li, X.-L. Shi, Z.-H. Zheng, G.-X. Liang, X.-H. Zhang, P. Fan and Z.-G. Chen, Adv. Sci., 2022, 9, 2103547.
23. H. Zhu, W. Li, A. Nozariasbmarz, N. Liu, Y. Zhang, S. Priya and B. Poudel, Nat. Commun., 2023, 14, 3300.
24. Y. Zhang, G. Peng, S. Li, H. Wu, K. Chen, J. Wang, Z. Zhao, T. Lyu, Y. Yu, C. Zhang, Y. Zhang, C. Ma, S. Guo, X. Ding, J. Sun, F. Liu and L. Hu, Nat. Commun., 2024, 15, 5978.
25. X. Ai, Y. Wu, H. Lyu, L. Giebeler, W. Xue, A. Sotnikov, Y. Wang, Q. Zhang, D. Makarov, Y. Yu, G. J. Snyder, K. Nielsch and R. He, Nat. Commun., 2025, 16, 6497.
26. S. Guo, S. Anand, M. K. Brod, Y. Zhang and G. J. Snyder, J. Mater. Chem. A, 2022, 10, 3051–3057.
27. D. Li, X.-L. Shi, J. Zhu, T. Cao, X. Ma, M. Li, Z. Han, Z. Feng, Y. Chen, J. Wang, W.-D. Liu, H. Zhong, S. Li and Z.-G. Chen, Nat. Commun., 2024, 15, 4242.
28. S. El Oualid, I. Kogut, M. Benyahia, E. Geczi, U. Kruck, F. Kosior, P. Masschelein, C. Candolfi, A. Dauscher, J. D. Koenig, A. Jacquot, T. Caillat, E. Alleno and B. Lenoir, Adv. Energy Mater., 2021, 11, 2100580.
29. Y. Liu, J. Zhi, W. Li, Q. Yang, L. Zhang and Y. Zhang, Molecules, 2023, 28, 5894.
30. Y. Zheng, M. Zou, W. Zhang, D. Yi, J. Lan, C.-W. Nan and Y.-H. Lin, J. Adv. Ceram., 2021, 10, 377–384.
31. B. Jia, D. Wu, L. Xie, W. Wang, T. Yu, S. Li, Y. Wang, Y. Xu, B. Jiang, Z. Chen, Y. Weng and J. He, Science, 2024, 384, 81–86.
32. R. Wu, Y. Yu, S. Jia, C. Zhou, O. Cojocaru-Mirédin and M. Wuttig, Nat. Commun., 2023, 14, 719.
33. H.-T. Liu, Q. Sun, Y. Zhong, Q. Deng, L. Gan, F.-L. Lv, X.-L. Shi, Z.-G. Chen and R. Ang, Nano Energy, 2022, 91, 106706.
34. B. Jiang, W. Wang, S. Liu, Y. Wang, C. Wang, Y. Chen, L. Xie, M. Huang and J. He, Science, 2022, 377, 208–213.
35. M. Hong, M. Li, Y. Wang, X.-L. Shi and Z.-G. Chen, Adv. Mater., 2023, 35, 2208272.
36. Y. Jiang, J. Dong, H.-L. Zhuang, J. Yu, B. Su, H. Li, J. Pei, F.-H. Sun, M. Zhou, H. Hu, J.-W. Li, Z. Han, B.-P. Zhang, T. Mori and J.-F. Li, Nat. Commun., 2022, 13, 6087.
37. C. Liu, Z. Zhang, Y. Peng, F. Li, L. Miao, E. Nishibori, R. Chetty, X. Bai, R. Si, J. Gao, X. Wang, Y. Zhu, N. Wang, H. Wei and T. Mori, Sci. Adv., 2025, 9, eadh0713.
38. M. Rakshit, D. Jana and D. Banerjee, J. Mater. Chem. A, 2022, 10, 6872–6926.
39. T. Zhang, S. Deng, X. Zhao, X. Ruan, N. Qi, Z. Chen, X. Su and X. Tang, J. Mater. Chem. A, 2022, 10, 3698–3709.
40. D. Liu, D. Wang, T. Hong, Z. Wang, Y. Wang, Y. Qin, L. Su, T. Yang, X. Gao, Z. Ge, B. Qin and L.-D. Zhao, Science, 2023, 380, 841–846.
41. C. Zhou, Y. K. Lee, Y. Yu, S. Byun, Z.-Z. Luo, H. Lee, B. Ge, Y.-L. Lee, X. Chen, J. Y. Lee, O. Cojocaru-Mirédin, H. Chang, J. Im, S.-P. Cho, M. Wuttig, V. P. Dravid, M. G. Kanatzidis and I. Chung, Nat. Mater., 2021, 20, 1378–1384.
42. B. Qin, D. Wang, T. Hong, Y. Wang, D. Liu, Z. Wang, X. Gao, Z.-H. Ge and L.-D. Zhao, Nat. Commun., 2023, 14, 1366.
43. S. Li, Y. Hou, D. Li, B. Zou, Q. Zhang, Y. Cao and G. Tang, J. Mater. Chem. A, 2022, 10, 12429–12437.
44. W. Li, Z. Yu, C. Liu, Y. Peng, B. Feng, J. Gao, G. Wu, X. Bai, J. Chen, X. Wang and L. Miao, J. Adv. Ceram., 2023, 12, 1511–1520.
45. H. Xu, H. Wan, R. Xu, Z. Hu, X. Liang, Z. Li and J. Song, J. Mater. Chem. A, 2023, 11, 4310–4318.
46. L. Li, B. Hu, Q. Liu, X. L. Shi and Z. G. Chen, Adv. Mater., 2024, 36, e2409275.
47. B. Du, Y. Yan and X. Tang, J. Electron. Mater., 2015, 44, 2118–2123.
48. K. Zhang, S. Liu, X. Wang, S. Wu, Q. Xiong, X. Wang, J. Chen, G. Wang, B. Zhang, H. Fu, G. Han, G. Wang, X. Lu, Y. Zhou and X. Zhou, Adv. Funct. Mater., 2024, 34, 2400679.
49. X. Hu, S. Zheng, Q. Xiong, S. Wu, Y. Huang, B. Zhang, W. Wang, X. Wang, N. Li, Z. Zhou, Y. Zhou, X. Lu and X. Zhou, Mater. Today Phys., 2023, 38, 101255.
50. D. T. Morelli, V. Jovovic and J. P. Heremans, Phys. Rev. Lett., 2008, 101, 035901.
51. E. D. Specht, J. Ma, O. Delaire, J. D. Budai, A. F. May and E. A. Karapetrova, J. Electron. Mater., 2015, 44, 1536–1539.
52. E. Quarez, K.-F. Hsu, R. Pcionek, N. Frangis, E. K. Polychroniadis and M. G. Kanatzidis, J. Am. Chem. Soc., 2005, 127, 9177–9190.
53. Y. Zhang, C. Xing, D. Wang, A. Genç, S. H. Lee, C. Chang, Z. Li, L. Zheng, K. H. Lim, H. Zhu, R. B. Smriti, Y. Liu, S. Cheng, M. Hong, X. Fan, Z. Mao, L.-D. Zhao, A. Cabot, T. Zhu and B. Poudel, Nat. Commun., 2025, 16, 22.
54. J. Ding, J. L. Niedziela, D. Bansal, J. Wang, X. He, A. F. May, G. Ehlers, D. L. Abernathy, A. Said, A. Alatas, Y. Ren, G. Arya and O. Delaire, Proc. Natl. Acad. Sci., 2020, 117, 3930–3937.
55. S. Bhattacharya, R. Basu, R. Bhatt, S. Pitale, A. Singh, D. K. Aswal, S. K. Gupta, M. Navaneethan and Y. Hayakawa, J. Mater. Chem. A, 2013, 1, 11289–11294.
56. F. Gascoin and A. Maignan, Chem. Mater., 2011, 23, 2510–2513.
57. K. Hoang, S. D. Mahanti, J. R. Salvador and M. G. Kanatzidis, Phys. Rev. Lett., 2007, 99, 156403.
58. P. Ganesan, C. S. Gantepogu, S. Duraisamy, S. Meledath Valiyaveettil, W.-H. Tsai, C.-R. Hsing, K.-H. Lin, K.-H. Chen, Y.-Y. Chen and M.-K. Wu, Mater. Today Phys., 2024, 42, 101358.
59. A. Bhui, S. Das, R. Arora, U. Bhat, P. Dutta, T. Ghosh, R. Pathak, R. Datta, U. V. Waghmare and K. Biswas, J. Am. Chem. Soc., 2023, 145, 25392–25400.
60. V. Taneja, S. Das, K. Dolui, T. Ghosh, A. Bhui, U. Bhat, D. K. Kedia, K. Pal, R. Datta and K. Biswas, Adv. Mater., 2024, 36, 2307058.
61. R. Pathak, L. Xie, S. Das, T. Ghosh, A. Bhui, K. Dolui, D. Sanyal, J. He and K. Biswas, Energy Environ. Sci., 2023, 16, 3110–3118.
62. B.-C. Chen, K.-K. Wang and H.-J. Wu, Small, 2024, 20, 2401723.
63. S. Roychowdhury, T. Ghosh, R. Arora, M. Samanta, L. Xie, N. K. Singh, A. Soni, J. He, U. V. Waghmare and K. Biswas, Science, 2021, 371, 722–727.
64. Y. Zhang, Z. Li, S. Singh, A. Nozariasbmarz, W. Li, A. Genc, Y. Xia, L. Zheng, S. H. Lee, S. K. Karan, G. K. Goyal, N. Liu, S. M. Mohan, Z. Mao, A. Cabot, C. Wolverton, B. Poudel and S. Priya, Adv. Mater., 2023, 35, e2208994.
65. M. Hong, Z. G. Chen, L. Yang, Z. M. Liao, Y. C. Zou, Y. H. Chen, S. Matsumura and J. Zou, Adv. Energy Mater., 2017, 8, 1702333.
66. Z. Gong, K. Saglik, J. Wu, A. Suwardi and J. Cao, Nanoscale, 2023, 15, 18283–18290.
67. J. Cao, J. Dong, K. Saglik, D. Zhang, S. F. D. Solco, I. J. W. J. You, H. Liu, Q. Zhu, J. Xu, J. Wu, F. Wei, Q. Yan and A. Suwardi, Nano Energy, 2023, 107, 108118.
68. J. H. Kim, J. H. Yun, S. Cha, S. Byeon, J. Park, H. Jin, S. Kim, S. J. Kim, J. Park, J. Jang, S. Park and J. S. Rhyee, Adv. Funct. Mater., 2024, 34, 202404886.
69. Z. Yang, S. Wang, Y. Sun, Y. Xiao and L.-D. Zhao, J. Alloys Compd., 2020, 828.
70. R. Zybała, Synth. Met., 2020, 270, 116606.
71. Z. Chen, B. Gao, J. Tang, X. Guo, W. Li and R. Ang, Appl. Phys. Lett., 2019, 115, 073903.
72. M. Yan, X. Tan, Z. Huang, G. Liu, P. Jiang and X. Bao, J. Mater. Chem. A, 2018, 6, 8215–8220.
73. H. Wang, H. Hu, N. Man, C. Xiong, Y. Xiao, X. Tan, G. Liu and J. Jiang, Mater. Today Phys., 2021, 16, 100298.
74. J. K. Lee, M.-W. Oh, B. Ryu, J. E. Lee, B.-S. Kim, B.-K. Min, S.-J. Joo, H.-W. Lee and S.-D. Park, Sci. Rep., 2017, 7, 4496.
75. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
76. J. P. Perdew, A. Ruzsinszky, G. I. Csonka, O. A. Vydrov, G. E. Scuseria, L. A. Constantin, X. Zhou and K. Burke, Phys. Rev. Lett., 2008, 100, 136406.
77. P. E. Blöchl, Phys. Rev. B:Condens. Matter Mater. Phys., 1994, 50, 17953–17979.
78. G. Kresse and J. Furthmüller, Phys. Rev. B:Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
79. Z. Hu, M. Yuan, W. Li, S. Wang, J. Li, J. Jiang, J. Shuai and Y. Hou, Acta Mater., 2025, 290, 120985.
80. R. Du, G. Zhang, M. Hao, X. Xuan, P. Peng, P. Fan, H. Si, G. Yang and C. Wang, ACS Appl. Mater. Interfaces, 2023, 15, 9508–9516.
81. L.-H. Ye, K. Hoang, A. J. Freeman, S. D. Mahanti, J. He, T. M. Tritt and M. G. Kanatzidis, Phys. Rev. B:Condens. Matter Mater. Phys., 2008, 77, 245203.
82. B. Kim, G. Chang, S. Seon, G. H. Lee, J. Park, C. Ju and S.-i. Kim, J. Korean Ceram. Soc., 2025 DOI:10.1007/s43207-025-00555-5.
83. D. G. Cahill, W. K. Ford, K. E. Goodson, G. D. Mahan, A. Majumdar, H. J. Maris, R. Merlin and S. R. Phillpot, J. Appl. Phys., 2003, 93, 793–818.
84. Q. Ren, M. K. Gupta, M. Jin, J. Ding, J. Wu, Z. Chen, S. Lin, O. Fabelo, J. A. Rodríguez-Velamazán, M. Kofu, K. Nakajima, M. Wolf, F. Zhu, J. Wang, Z. Cheng, G. Wang, X. Tong, Y. Pei, O. Delaire and J. Ma, Nat. Mater., 2023, 22, 999–1006.
85. M. K. Gupta, J. Ding, D. Bansal, D. L. Abernathy, G. Ehlers, N. C. Osti, W. G. Zeier and O. Delaire, Adv. Energy Mater., 2022, 12, 2200596.

---