Tailoring Thermoelectric Transport Properties of Ag-Alloyed PbTe

Oct 23, 2018 - Abstract · Supporting Info · Figures · References. Tools & Sharing. Add to Favorites ... Read the ACS privacy policy. CONTINUE. pubs lo...
0 downloads 0 Views 5MB Size
Subscriber access provided by UNIV OF LOUISIANA

Energy, Environmental, and Catalysis Applications

Tailoring Thermoelectric Transport Properties of AgAlloyed PbTe: Effects of Microstructure Evolution Ariel Sheskin, Torsten Schwarz, Yuan Yu, Siyuan Zhang, Lamya Abdellaoui, Baptiste Gault, Oana Cojocaru-Miredin, Christina Scheu, Dierk Raabe, Matthias Wuttig, and Yaron Amouyal ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b15204 • Publication Date (Web): 23 Oct 2018 Downloaded from http://pubs.acs.org on October 26, 2018

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

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

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

ACS Applied Materials & Interfaces

Tailoring Thermoelectric Transport Properties of Ag-Alloyed PbTe: Effects of Microstructure Evolution Ariel Sheskin,

a

Torsten Schwarz,

b

Baptiste Gault,

b

Oana Cojocaru-Mirédin,

Yuan Yu, b,c

c

Siyuan Zhang,

Christina Scheu,

b

b

Lamya Abdellaoui,

Dierk Raabe,

b

b

Matthias

Wuttig, c,d and Yaron Amouyal a,*

a

Department of Materials Science and Engineering, Technion-Israel Institute of Technology, Technion City, 32000 Haifa, Israel

b

Max-Planck Institut für Eisenforschung GmbH, 40237, Düsseldorf, Germany

c I.

d

Physikalisches Institut (IA), RWTH Aachen, 52074, Aachen, Germany

JARA-Institut Green IT, JARA-FIT, Forschungszentrum Jülich GmbH and RWTH Aachen University, 52056 Aachen, Germany

Abstract Capturing and converting waste heat into electrical power through thermoelectric generators based on the Seebeck effect is a promising alternative energy source. Among thermoelectric compounds, PbTe can be alloyed and form precipitates by aging at elevated temperatures, thus reducing thermal conductivity by phonon scattering. Here, PbTe is alloyed with Ag to form Ag-rich precipitates having a number density controlled by heat treatments. We employ complementary scanning transmission electron microscopy and

*

Corresponding author: [email protected] (Y. Amouyal) 1 ACS Paragon Plus Environment

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

Page 2 of 38

atom probe tomography to analyze the precipitate number density and the PbTe-matrix composition. We measure the temperature dependent transport coefficients, and correlate them with the microstructure. The thermal and electrical conductivities, as well as the Seebeck coefficients, are found to be highly-sensitive to the microstructure and its temporal evolution, e.g. the number density of Ag- rich precipitates increases by ca. three orders of magnitude and reaches (1.68 ± 0.92)×1024 m-3 upon aging at 380 °C for 6 h, which is pronounced by reduction of thermal conductivity to a value as low as 0.85 Wm-1K-1 at 300 °C. Our findings will help to guide predictive tools for further design of materials for energy harvesting.

Keywords: thermoelectric materials, lead-telluride, transmission electron microscopy, atom probe tomography, phonon scattering

1.

Introduction Lead telluride (PbTe) is one of the most interesting materials for thermoelectric (TE)

heat-to electrical energy conversion in mid-temperature ranges (300 - 500 °C), which are technologically important for thermal energy harvesting. 1 Among the advantages of PbTe, such as low band gap (0.3 eV at room temperature) with relatively high electrical conductivity, as well as intrinsically non-covalent weak interatomic bonding referred to as metavalent bonding, 2 giving rise to low thermal conductivity, is its capability of forming nano-precipitates upon alloying with another elements,

3-6

thereby reducing the lattice

component of the thermal conductivity, 𝜅𝑙. 7-9 𝜅𝑙 directly affects the TE figure of merit (ZT) 2 ACS Paragon Plus Environment

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

ACS Applied Materials & Interfaces

and, eventually, dictates the TE power conversion efficiency.

10-12

Formation of nano-

precipitates, however, affects not only 𝜅𝑙, but also other TE transport coefficients, such as the Seebeck coefficient, S, and the electrical conductivity, 𝜎, as well as the electronic component of the thermal conductivity, 𝜅𝑒.

13-15

Nanostructures are subjected to thermal

instability, and their temporal evolution under service conditions dictates all of the above mentioned TE transport coefficients. It is, therefore, of prime technological interest to study temporal evolution of nanostructures. It has been shown that Ag-alloying exhibits great potential in formation of Ag2Teprecipitates dispersed in the PbTe-matrix. 16-21 The number density of precipitates, Nv, can be controlled in a way typically exploited in physical metallurgy, by adjusting the time and temperatures of aging heat treatments. 22 Here, the guideline directing this methodology is that precipitates with high number density values serve as efficient phonon scattering centers, resulting in reduced 𝜅𝑙.

23-24

Additional considerations that are essential, though

commonly disregarded, are the effects of dilution of the matrix with Ag during nucleation, growth, and coarsening of the Ag-rich precipitates on S and 𝜎. 13-14, 22 It has been reported that nucleation of Ag2Te-precipitates in a 0.04 at. % Bi-doped (PbTe)0.97(Ag2Te)0.03 compound subjected to aging heat treatments at 380 °C yield Nv = 2.7·1020 m-3 upon 6 h aging, which decreases down to ~1019 m-3 after 48 h aging. 20 This trend was associated to nucleation of Ag2Te-precipitates for relatively short aging times, followed by their coarsening observed for longer times. Alongside this microstructure evolution, 𝜅𝑙 reaches a minimum value as low as 0.4 Wm-1K-1 at 500 °C upon 6 h aging at 380 °C due to intensive phonon scattering by a high number of precipitates, which increases up to 1.2 Wm-1K-1 upon aging for longer times. This finding was associated to coarsening 3 ACS Paragon Plus Environment

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

Page 4 of 38

of the Ag2Te-precipitates and reduction of matrix supersaturation. Major changes in the electronic transport properties were observed, as well. It is noteworthy that the above-reported number density values, as high as 2.7×1020 m-3, were evaluated based on scanning electron microscopy (SEM) analysis solely.

20

Significant phonon scattering, however, can be initiated by precipitates with significantly larger Nv-values, ca. 1021 -1022 m-3. For instance, reduction of 𝜅𝑙 by approx. 20% was evaluated for PbTe containing precipitates of Nv = 1022 m-3, based on the Callaway model for lattice thermal conductivity. 15 It is, therefore, required to apply techniques with higher spatial resolution to resolve microstructural features at the nanoscale, and to account for their effects on the transport coefficients. Moreover, the effects of microstructure evolution on the electronic transport coefficients should be elucidated also in terms of the matrix concentration in the inter-precipitate regions and its temporal evolution. This requires capabilities of chemical analysis at the nanoscale along with high sensitivity. To meet these requirements, we apply herein scanning transmission electron microscopy (STEM) in concert with atom probe tomography (APT).

25-29

We address three main questions: (1)

what are the precipitates’ number density values and how do they evolve with aging time? (2) what are the average Ag concentrations in the inter-precipitate regions, and how do they evolve with aging time? and (3) how do the above quantities correlate with the TE transport coefficients? Addressing these questions will help us better understand the thermal stability of two-phase PbTe compounds, providing us with predictive alloy design tools.

2.

Experimental procedure

4 ACS Paragon Plus Environment

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

ACS Applied Materials & Interfaces

(PbTe)0.97(Ag2Te)0.03 samples were produced by mixing of pure elemental Pb powder (99.96%, Riedel-de Haën®), Te ingots (99.99%, STREM CHEMICALS®), and Ag shots (99.999%, Alfa Aesar®) in the corresponding molar ratios in quartz ampoules and were subjected to melting at 1000 °C for 6 h in a 120 torr Ar-7% H2 protecting atmosphere. The quartz ampoules were moderately cooled down and dwelled at 700 °C for 48 h for homogenization, followed by quenching in iced-water bath. The resulting ingot, comprising the super-saturated PbTe-based solid solution, was grinded into fine powder and hot-pressed at 650 °C within the single-phase region in a 12.5 mm diameter die at 45 MPa for 15 minutes under Ar-7% H2 atmosphere, followed by iced-water quenching. Part of the hot-pressed disks were, finally, aged at 380 °C for 6 h or 48 h in sealed quartz ampoules under 120 torr Ar-7% H2 atmosphere followed by iced-water quenching. 12.5 mm diameter pellets were prepared and their bulk densities were determined applying the Archimedes method using water immersion. Determination of electronic transport coefficients was carried out using an SBA-458 Nemesis® apparatus (Netzsch GmbH, Selb, Germany) applying a four-point probe with a locally-heated specimen holder. 13

Thermal diffusivity was directly determined by applying an LFA-457 MicroFlash® laser

flash analyzer (LFA; Netzsch GmbH, Selb, Germany). Specific heat capacity values were evaluated in the LFA using a pure Al2O3-reference sample with similar geometry of the sample, and the temperature dependent thermal conductivity was evaluated as a product of the thermal diffusivity and volumetric specific heat capacity.

30

A detailed description is

provided in the Supporting Information. SEM observations were carried out using a Zeiss Ultra Plus field emission gun (FEG) microscope. To obtain better contrast so that Ag-rich precipitates can be 5 ACS Paragon Plus Environment

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

Page 6 of 38

distinguished, secondary electrons (SE) signals were collected by Ga+ ion excitation using a FEI Helios NanoLab 600i dual-beam focused ion beam (FIB) microscope. Specimens for STEM and APT analyzes were prepared in the form of lamellae and sharp needles, respectively, applying a site-specific “lift-out” method as implemented in the FIB.

25

Conventional TEM investigations were performed using a Philips CM20 TEM working at an acceleration voltage of 200 kV. STEM observations were carried out in an FEI Cs-probe corrected Titan Themis using an acceleration voltage of 300 kV, equipped with a detector for energy dispersive x-ray spectroscopy (EDS) analysis. Multivariate statistical analysis was applied for a component-based quantification of the STEM-EDS dataset.

31

APT

measurements were conducted using a laser-assisted local electrode atom probe (LEAP™, Cameca Instr.) of either the 4000X Si or 5000 XS types. The APT analyses were carried out at a base temperature of 40 K in laser pulsing mode with a pulse duration of ~10 ps, 355 nm wavelength, 200 kHz pulse frequency, and a pulse energy of 30 pJ. A detection rate of 10 ions of 1000 pulses was maintained. The resulting APT datasets were processed using the IVAS 3.6.12 software.

3.

Results and discussion

3.1.

Microstructure and composition To capture the mesoscale nature of the precipitates distribution, we apply scale-

bridging techniques ranging from SEM/FIB to TEM and APT. Figure 1 shows SE micrographs collected from the as-quenched (AQ) specimens, as well as from those aged at 6 h and 48 h, under Ga+ ion excitation using FIB. The average precipitate number densities are evaluated by counting the number of precipitates (several dozens) per unit

6 ACS Paragon Plus Environment

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

ACS Applied Materials & Interfaces

area in several micrographs taken from different locations. Their average and standard deviation of the mean are then calculated, assuming that their spatial distribution is uniform throughout the entire volume. It is shown that even the AQ samples contain precipitates of relatively low Nv = (5.0 ± 1.5)×1017 m-3, which increases up to (1.07 ± 0.45)×1018 m-3 for the 6 h aged sample, followed by a decrease down to a value as small as (4.0 ± 1.5)×1017 m-3 for the 48 h aged sample. We highlight that this SE/FIB observation reveals precipitates that are typically 1-2 μm long, with a similar characteristic inter-precipitate separating distance. Populations of smaller precipitates having higher Nv-values, if exist, may be overlooked in this observation. These data appear in Table 1. Detailed explanation for this observation based on nucleation and coarsening steps, though with different numerical values, is given elsewhere. 20 [Fig. 1] A population of highly-dispersed, nanoscale precipitates having larger Nv-values is revealed by TEM analysis. Figure 2 (a) displays a dark field TEM micrograph taken from the 48 h aged sample along the [011] zone axis and the corresponding selected area electron diffraction (SAED) pattern, showing (b) monoclinic Ag2Te-precipitates with the

[011](100)𝑃𝑏𝑇𝑒 ||[010](100)𝐴𝑔2𝑇𝑒 orientation relationship with respect to the PbTematrix. Figure 2 (c) is a STEM micrograph showing two precipitates. The corresponding EDS map (d) indicates a homogeneous distribution of Ag and Te in the precipitates and their immediate surroundings, in which 10-100 nm long precipitates possess a nearly Ag2Te-stoichiometry. [Fig. 2] 7 ACS Paragon Plus Environment

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

To accurately determine the Ag-rich precipitate number density and the matrix composition in the inter-precipitate regions of approx. 10 to 100 nm length scales, we apply APT analysis. APT is capable of 3D analysis of volumes comprising precipitate of number densities as high as 1025 m-3 as well as to quantify composition of tiny volumes on the nm3 scale. 25, 27 We utilize the iso-concentration method to discern the Ag-rich precipitates from their surrounding matrix.

25

To this end, we compare 3D APT reconstructed volumes

comprising precipitates of different spatial distributions to STEM micrographs showing similar features. Thus, we can optimize the reconstruction of the APT datasets by comparing the size of the precipitates obtained for a given iso-concentration value to the precipitates’ actual size as analyzed by TEM. The following precipitates’ population analysis is based on the following. One APT reconstructed volume was analyzed for the AQ sample, revealing six elongated large precipitates (up to 100 nm long) and 37 small (up to ~10 nm) spheroidal precipitates. Four reconstructed volumes were analyzed for the 6 h aged sample, each containing several dozens of small (up to ~10 nm) spheroidal precipitates. Part of these precipitates are interconnected by dislocations, which are clearly indicated by strong Ag-segregation to their cores; 21 they comprise typically 4-5 at. % Ag. Analysis of the 48 h sample is based on a total of three APT reconstructed volumes, each containing both populations of elongated and spheroidal precipitates that are larger than those observed for the 6 h samples, probably due to coarsening that is possibly stimulated by the interconnecting dislocations. The strongly anisotropic morphology of such precipitates is determined by matrix/precipitates interfacial energy as well as the degree of interfacial lattice mismatch; both factors can be tuned by changing the composition of the interface or its local environment, e.g. due to variation of the entire alloy composition or

8 ACS Paragon Plus Environment

Page 8 of 38

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

ACS Applied Materials & Interfaces

interfacial segregation. These effects were thoroughly investigated particularly for telluride-based TE compounds, which are considered to be crystallographically anisotropic, e.g. for Na-doped PbTe-PbS,

32

Sb2Te3-PbTe,

33

and PbTe-Cu2Te compounds.

34

We

believe that the aging heat treatments implemented in our study change the composition of the matrix/precipitate interface (see Section 3.3), thereby initiating the Ag-rich precipitate morphology change. Figure 3 displays typical APT reconstructed volumes acquired for the (a) AQ sample and those aged for (b) 6 h and (c) 48 h, where the red Ag iso-concentration surfaces mark the interface between the Ag-rich precipitates and the matrix. It is clearly shown that Nv is relatively small for the AQ and 48 h samples, however attains much larger values for 6 h aging. [Fig. 3] The composition of the matrix is evaluated for the exterior volumes far from the precipitates. The resulting precipitate number density as well as the matrix composition for the three states are listed in Table 1. [Table 1] The expected equilibrium composition of precipitates is Ag2Te, and this stoichiometry is observed for part of the precipitates, e.g. those shown in Fig. 2. We note, however, that some of the precipitates exhibit different stoichiometry. For example, large precipitates (10-100 nm dia.) observed for the AQ sample exhibit a Ag:Te ratio of nearly 2:1, whereas the smaller ones (1-10 nm dia.) exhibit a nearly 1:1 ratio. This is exemplified by the proximity histograms (proxigrams),

25

which are plotted across precipitate/matrix

interfaces of the AQ samples, Fig. 4. 9 ACS Paragon Plus Environment

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

[Fig. 4] 3.2.

Thermoelectric transport properties Measurements of electronic and thermal transport coefficients were performed

using SBA and LFA, respectively, between room temperature and 300 °C. The electrical conductivity, Seebeck coefficient, and thermal conductivity are shown in Fig. 5. [Fig. 5] The AQ sample exhibits the highest electrical conductivity and the 48 h one exhibits the lowest. Also, the electrical conductivity of the 48 h sample increases monotonously with temperature, which indicates a non-degenerate semiconductor behavior. Conversely, the AQ and 6 h samples do not exhibit such monotonous trend. Instead, their electrical conductivity values reaches a minimum at around 200 °C. The Seebeck coefficients exhibit rather complex trends: for temperatures of up to 150 °C, S is positive and increases with temperature. Here, the AQ sample shows the smallest and the 48 h aged sample the largest S values and the largest and smallest electrical conductivity, respectively. For elevated temperatures we observe a reversed behavior. In addition, for the 48 h sample, S decreases significantly and attains negative values for T > 250 °C, which indicates a p-n transition. Moreover, the thermal conductivity values decrease monotonously with temperature. Interestingly, we observe that 𝜅(48 h) > 𝜅(AQ) > 𝜅(6 h), which is discussed later. 3.3.

How do the microstructure and composition dictate the transport properties? Complementary TEM and APT analysis enables us identify precipitates having Nv-

values that are larger than those detected by SE/FIB. From this point of view, the Nv-values

10 ACS Paragon Plus Environment

Page 10 of 38

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

ACS Applied Materials & Interfaces

derived from SE/FIB and complementary TEM and APT behave in the same manner qualitatively, i.e. they indicate the same nucleation and coarsening stages during aging. This is confirmed by the relatively low Nv-value observed by SE/FIB for the AQ state, which attains a maximum value for 6 h aging, followed by a decrease down to a value that is similar to the AQ state upon 48 h aging. Similar trends observed for the same materials class are reported elsewhere.

20

The same behavior is observed for the precipitate

population at the nanometer scale analyzed by TEM and APT, as shown in Table 1 and displayed in Fig. 6. Quantitatively, the Nv-values acquired from TEM and APT analysis are larger than those acquired from FIB by 4-6 orders of magnitude. The AQ state is characterized by a small (though nonzero) precipitate number density and high level of lattice strain, since it is closely described as a super-saturated solid solution. Upon aging, enhanced nucleation of Ag-rich precipitates takes place, which is accompanied by simultaneous rise in Nv and decrease in the level of lattice strain. We note that the composition of these precipitates evolve with aging times, with Ag:Te ratio nearly 1:1 for smaller ones (1-10 nm dia.), which approaches 2:1 for larger ones (>10-100 nm dia.), as shown in Fig. 4; this agrees well with Grossfeld et al, 20 who reported that large (1-2 μm long) precipitates possess the Ag2Te-stoichiometry. For adequately longer aging times the coarsening regime begins, in which Nv decreases along with a decrease of lattice strain. 35 Accordingly, the average concentration of solute Ag atoms in the matrix continues to decrease with aging time, until it reaches an equilibrium value for longer aging times. Interestingly, such a non-monotonous behavior of Nv with aging time was observed in other material systems, e.g. Ni-doped ZnO, AgSbTe2 based compounds, and Ag-alloyed PbTe,

11 ACS Paragon Plus Environment

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

13-14, 20 with implications for TE performance. Comparison between the Ag-rich precipitates

Nv-values and Ag-concentration in the matrix is displayed in Fig. 6. [Fig. 6] As shown in Fig. 6, Nv is highest for the 6 h sample, while the Ag concentration in the matrix is maximum for the AQ state and decreases upon aging, and is identical for 6 h and 48 h aging. Based on the above-described microstructure evolution, it is suggested that nucleation of Ag-rich precipitates takes place up to 6 h aging, and the onset of coarsening is between 6 h and 48 h. We elucidate the measured TE transport properties following this microstructure evolution sequence. Ag solute atoms in PbTe have been reported to behave as both n- or p-type dopants, depending on their concentration.

16-17, 36

When occupying an interstitial

site, the monovalent Ag atoms act as donors; alternatively, when they reside in Pbsubstitution sites, they act as acceptors, since Pb-atoms are divalent. This yields an amphoteric behavior of Ag in PbTe. 36 Interestingly, variation or reversal of the occupancy state is a dynamic process, which may take place during heat treatments, since the solubility of Ag in PbTe strongly depends on temperature; 37 this process is, therefore, both time- and temperature-dependent. It turns out that Ag-rich precipitates act as reservoirs for Agdopants, and their evolution modifies the p/n-type polarity of the PbTe-matrix upon heat treatments. This phenomenon, referred to as dynamic doping, was reported also for Cudoped PbSe, where Cu-dopants were found to occupy the PbSe-matrix more intensively with increasing temperatures.

38

Furthermore, it was reported that Cu-dopants in PbTe

initially occupy intrinsic Pb-vacancies, thereby improving charge carrier mobility. For

12 ACS Paragon Plus Environment

Page 12 of 38

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

ACS Applied Materials & Interfaces

larger concentrations, excess Cu atoms reside in interstitial sites, thereby reducing the lattice thermal conductivity; and for larger concentrations, they form Cu2Te-precipitates that further reduce thermal conductivity. 34 This scenario well-exemplifies the essence of the dynamic doping effect, having implications for both electronic and thermal transport. Based on the aforementioned concepts, we maintain that Ag atoms act as electron acceptors for temperatures up to 150 °C and for the current Ag concentration range, yielding positive S-coefficients. Since Ag atoms are consumed by Ag-rich precipitates during their nucleation and growth, their average concentration in the PbTe matrix decreases with aging time (see Fig. 6), resulting in a reduction of the electrical conductivity and increase of |S|, as indicated by Figs. 5 (a) and (b), respectively. The behavior of the sample aged for 48 h, indicating very low electrical conductivity with positive S-coefficient that changes its sign to negative upon heating, can be elucidated by transition from p- to ntype polarity, which occurs since the volume fraction of Ag2Te is the greatest after 48 h aging, so that the amount of Ag atoms dissolved in the PbTe-matrix is the smallest one. The thermal conductivity dependence on temperature and aging time, Fig. 5 (c), can be tightly correlated to the microstructure and composition, as depicted in Fig. 6. Our analysis relies on the assumption that thermal conductivity is dominated by the lattice component, i.e. by phonon scattering. To differentiate between the electronic and lattice components we evaluate the former applying the Wiedemann-Franz relationship: 𝜅𝑒 = 𝐿𝜎𝑇, 39

where L is the Lorenz number and the 𝜎-values are given in Fig. 5 (a). L is usually

evaluated for the degenerate limit and is considered to be constant; however, to address the more general case of a non-degenerate semiconducting behavior, we apply a semiempirical expression reported by Kim et al., 40 where L is a function of S. 20 Figure 7 shows 13 ACS Paragon Plus Environment

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

the temperature dependence of both the lattice and electronic components of the thermal conductivity. [Fig. 7] It is shown that the electronic component is lower by approx. four orders of magnitude than the lattice component, which justifies employing the well-known Callaway model for lattice thermal conductivity to account for the results shown in Fig. 5 (c). 9, 15, 21, 41-42 For instance, the temperature dependent values are typical for the Umklapp behavior, in which phonons are scattered inelastically by other phonons. 39 Besides, there are two other interesting features in Fig. 5 (c). First, the thermal conductivity decreases upon 6 h aging, e.g. from 0.93 Wm-1K-1 (AQ) to 0.85 Wm-1K-1 (6 h) at 300 °C, and then increases up to values that are even larger than those in the AQ state, e.g. 1.03 Wm-1K-1 at 300 °C upon 48 h aging. This behavior is associated to an increased phonon scattering by the highly dispersed Ag-rich precipitates after 6 h aging having a Nv value as high as (1.7 ± 0.92)×1024 m-3 (see Table 1). In turn, 48 h aging leads to coarsening of the precipitates, so that their number density drops down to (1.2 ± 0.75)×1022 m-3, rendering them less effective in phonon scattering, thereby 𝜅 increases. Simultaneously, the PbTe matrix becomes less super-saturated with aging time, which leads to a decreased level of strains. Lattice strains, by themselves, play a significant role in phonon scattering. This is why the specimen aged for 48 h exhibits thermal conductivity higher than the original AQ state. Second, only Nv-values as large as those listed in Table 1 (~1021-1024 m-3) are capable of significant phonon scattering, whereas those detected by FIB, of the order of magnitude ~1017-1018 m-3, can hardly account for the high sensitivity of 𝜅 to the

14 ACS Paragon Plus Environment

Page 14 of 38

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

ACS Applied Materials & Interfaces

aging time, as shown in Fig. 5 (c).

15

Such information could not be gained without the

complementary TEM and APT methodology. It is noteworthy that second-phase precipitates of number densities as high as 1024 m-3, as reported in our study, are found to be very efficient in phonon scattering; however, there are alternative strategies yielding comparable or greater effects, such as grain refinement. 43 A striking example is significant decrease of lattice thermal conductivity of PbS due to phonon scattering upon reduction of grain size down to ca. 30 nm. 44 4. Summary and conclusions We produced (PbTe)0.97(Ag2Te)0.03 based compounds for TE energy harvesting at the mid-temperature range, and performed heat treatments at 380 °C for different durations to control nucleation and growth of Ag-rich precipitates. The microstructures are characterized by applying SEM/FIB and complementary TEM and APT technique to obtain accurate information of the precipitates’ number density and the composition of the matrix. We measured the thermal and electrical conductivities as well as the Seebeck coefficients between room temperature and 300 °C, and establish the correlation between nanostructure and TE transport properties in this alloy. We can draw the following major conclusions: 1. Complementary TEM and APT is a promising methodology that enables us to characterize the precipitate number density and the PbTe matrix composition precisely. It is capable of analyzing Nv- values as high as 1025 m-3, which are sufficiently high to account for significant changes in thermal conductivity. 2. Analysis of the nanostructure and matrix composition for the different aging times enables us to identify both stages of nucleation and coarsening of 15 ACS Paragon Plus Environment

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

precipitates; the latter tightly correlates with thermal and electronic transport properties. 3. Explicitly, a maximum value of Nv = (1.7 ± 0.9)×1024 m-3 is obtained for 6 h aging, which is accompanied by a pronounced reduction of thermal conductivity with respect to the AQ- state. Further aging up to 48 h results in an increase of the thermal conductivity due to both precipitate coarsening and relaxation of the PbTe matrix strains. 4. The above trend is accompanied by a continuous reduction of electrical conductivity and an increase of low-temperature S-coefficients upon aging due to consumption of Ag atoms from the PbTe matrix by the growing of Ag-rich precipitates. These results may serve as a predictive tool for further materials design for TE energy harvesting.

Acknowledgments Y.A. wishes to acknowledge generous support from the German-Israeli Foundation for Research and Development (GIF), Grant No. I-2333-1150.10/2012, and the Asher Space Research Institute (ASRI) at the Technion. Partial support from the Nancy and Stephen Grand Technion Energy Program (GTEP) and the Russell-Berrie Nanotechnology Institute (RBNI) is acknowledged, as well. O. C-M and M. W. wish to acknowledge RWTH Aachen University’ support through “Seed Funds” project. The authors are grateful to Uwe Tezins, Andreas Sturm, and Volker Kree for their support to the APT, SEM, FIB, and TEM 16 ACS Paragon Plus Environment

Page 16 of 38

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

ACS Applied Materials & Interfaces

facilities at Max‐Planck‐Institut für Eisenforschung GmbH. T.S. thanks for the support by the German Research Foundation (DFG) (Contract GA 2450/1‐1).

Supporting Information. Raw data showing temperature dependent thermal diffusivity measured for the AQ sample, as well as for those aged at 380 °C for 6 h and 48 h (Figure S1); Raw data showing temperature dependent heat capacity values measured for the same specimens (Figure S2).

References (1)

LaLonde, A. D.; Pei, Y.; Wang, H.; Snyder, G. J. Lead Telluride Alloy Thermoelectrics. Materials Today 2011, 14, 526-532.

(2)

Zhu, M.; Cojocaru-Mirédin, O.; Mio, A. M.; Keutgen, J.; Küpers, M.; Yu, Y.; Cho, J.-Y.; Dronskowski, R.; Wuttig, M. Unique Bond Breaking in Crystalline Phase Change Materials and the Quest for Metavalent Bonding. Adv. Mater. 2018, 30, 1706735.

(3)

Snyder, G. J.; Toberer, E. S. Complex Thermoelectric Materials. Nat. Mater. 2008, 7, 105-114.

(4)

Tan, G.; Shi, F.; Hao, S.; Zhao, L.-D.; Chi, H.; Zhang, X.; Uher, C.; Wolverton, C.; Dravid, V. P.; Kanatzidis, M. G. Non-Equilibrium Processing Leads to Record High Thermoelectric Figure of Merit in PbTe-SrTe. Nat Commun 2016, 7, 12167.

(5)

Korkosz, 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.; Dravid, V. P.; 17 ACS Paragon Plus Environment

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

Page 18 of 38

Kanatzidis, M. G. High ZT in p-Type (PbTe)1–2x(PbSe)x(PbS)x Thermoelectric Materials. J. Am. Chem. Soc. 2014, 136, 3225-3237. (6)

Rawat, P. K.; Paul, B.; Banerji, P. Exploration of Zn Resonance Levels and Thermoelectric Properties in I-Doped PbTe with ZnTe Nanostructures. ACS Applied Materials & Interfaces 2014, 6, 3995-4004.

(7)

He, J. Q.; Girard, S. N.; Zheng, J. C.; Zhao, L. D.; Kanatzidis, M. G.; Dravid, V. P. Strong Phonon Scattering by Layer Structured PbSnS2 in PbTe Based Thermoelectric Materials. Adv. Mater. 2012, 24, 4440-4444.

(8)

He, J. Q.; Sootsman, J. R.; Girard, S. N.; Zheng, J. C.; Wen, J. G.; Zhu, Y. M.; Kanatzidis, M. G.; Dravid, V. P. On the Origin of Increased Phonon Scattering in Nanostructured PbTe Based Thermoelectric Materials. J. Am. Chem. Soc. 2010, 132, 8669-8675.

(9)

Lo, S.-H.; He, J.; Biswas, K.; Kanatzidis, M. G.; Dravid, V. P. Phonon Scattering and

Thermal

Conductivity

in

p-Type

Nanostructured

PbTe-BaTe

Bulk

Thermoelectric Materials. Adv. Funct. Mater. 2012, 22, 5175-5184. (10) Tritt, T. M. Thermoelectric Phenomena, Materials, and Applications. Annual Review of Materials Research 2011, 41, 433-448. (11) He, J.; Tritt, T. M. Advances in Thermoelectric Materials Research: Looking Back and Moving Forward. Science 2017, 357. (12) Zebarjadi, M.; Esfarjani, K.; Dresselhaus, M. S.; Ren, Z. F.; Chen, G. Perspectives on Thermoelectrics: From Fundamentals to Device Applications. Energy & Environmental Science 2012, 5, 5147-5162.

18 ACS Paragon Plus Environment

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

ACS Applied Materials & Interfaces

(13) Koresh, I.; Amouyal, Y. Effects of Microstructure Evolution on Transport Properties of Thermoelectric Nickel-Doped Zinc Oxide. Journal of the European Ceramic Society 2017, 37, 3541-3550. (14) Cojocaru-Mirédin, O.; Abdellaoui, L.; Nagli, M.; Zhang, S.; Yu, Y.; Scheu, C.; Raabe, D.; Wuttig, M.; Amouyal, Y. Role of Nanostructuring and Microstructuring in Silver Antimony Telluride Compounds for Thermoelectric Applications. ACS Applied Materials & Interfaces 2017, 9, 14779-14790. (15) Amouyal, Y. Reducing Lattice Thermal Conductivity of the Thermoelectric Compound AgSbTe2 (P4/mmm) by Lanthanum Substitution: Computational and Experimental Approaches. Journal of Electronic Materials 2014, 43, 3772-3779. (16) Pei, Y.; May, A. F.; Snyder, G. J. Self-Tuning the Carrier Concentration of PbTe/Ag2Te Composites with Excess Ag for High Thermoelectric Performance. Adv. Energy Mater. 2011, 1, 291-296. (17) Pei, Y.; Lensch-Falk, J.; Toberer, E. S.; Medlin, D. L.; Snyder, G. J. High Thermoelectric Performance in PbTe Due to Large Nanoscale Ag2Te Precipitates and La Doping. Adv. Funct. Mater. 2011, 21, 241-249. (18) Pei, Y.; Heinz, N. A.; LaLonde, A.; Snyder, G. J. Combination of Large Nanostructures and Complex Band Structure for High Performance Thermoelectric Lead Telluride. Energy & Environmental Science 2011, 4, 3640-3645. (19) Lensch-Falk, J. L.; Sugar, J. D.; Hekmaty, M. A.; Medlin, D. L. Morphological Evolution of Ag2Te Precipitates in Thermoelectric PbTe. Journal of Alloys and Compounds 2010, 504, 37-44.

19 ACS Paragon Plus Environment

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

(20) Grossfeld, T.; Sheskin, A.; Gelbstein, Y.; Amouyal, Y. Microstructure Evolution of Ag-Alloyed PbTe-Based Compounds and Implications for Thermoelectric Performance. Crystals 2017, 7, 281. (21) Yu, Y.; Zhang, S.; Mio, A. M.; Gault, B.; Sheskin, A.; Scheu, C.; Raabe, D.; Zu, F.; Wuttig, M.; Amouyal, Y.; Cojocaru-Mirédin, O. Ag-Segregation to Dislocations in PbTe-Based Thermoelectric Materials. ACS Applied Materials & Interfaces 2018, 10, 3609-3615. (22) Amouyal, Y.; Gelbstein, Y.; Fuks, D. Physical Metallurgy Inspired Nano-Features for Enhancement of Thermoelectric Conversion Efficiency. Advanced Theory and Simulations 2018, 1, 1800072. (23) Kim, W.; Majumdar, A. Phonon Scattering Cross Section of Polydispersed Spherical Nanoparticles. J. Appl. Phys. 2006, 99, 084306. (24) Cahill, D. G.; Braun, P. V.; Chen, G.; Clarke, D. R.; Fan, S.; Goodson, K. E.; Keblinski, P.; King, W. P.; Mahan, G. D.; Majumdar, A.; Maris, H. J.; Phillpot, S. R.; Pop, E.; Shi, L. Nanoscale Thermal Transport. I. 2003–2012. Applied Physics Reviews 2014, 1, 011305. (25) Gault, B.; Moody, M. P.; Cairney, J. M.; Ringer, S. P., Atom Probe Microscopy. Springer New York: 2012. (26) Seidman, D. N.; Stiller, K. An Atom-Probe Tomography Primer. Materials Research Society Bulletin 2009, 34, 717-721. (27) Amouyal, Y.; Schmitz, G. Atom Probe Tomography—a Cornerstone in Materials Characterization. MRS Bulletin 2016, 41, 13-18.

20 ACS Paragon Plus Environment

Page 20 of 38

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

ACS Applied Materials & Interfaces

(28) Devaraj, A.; Perea, D. E.; Liu, J.; Gordon, L. M.; Prosa, T. J.; Parikh, P.; Diercks, D. R.; Meher, S.; Kolli, R. P.; Meng, Y. S.; Thevuthasan, S. Three-Dimensional Nanoscale Characterisation of Materials by Atom Probe Tomography. International Materials Reviews 2016, 63, 68-101. (29) Kelly, T. F.; Larson, D. J. Atom Probe Tomography 2012. Annual Review of Materials Research 2012, 42, 1-31. (30) Graff, A.; Amouyal, Y. Effects of Lattice Defects and Niobium Doping on Thermoelectric Properties of Calcium Manganate Compounds for Energy Harvesting Applications. Journal of Electronic Materials 2016, 45, 1508-1516. (31) Zhang, S.; Scheu, C. Evaluation of EELS Spectrum Imaging Data by Spectral Components and Factors from Multivariate Analysis. Microscopy 2018, 67, i133i141. (32) He, J.; Blum, I. D.; Wang, H.-Q.; Girard, S. N.; Doak, J.; Zhao, L.-D.; Zheng, J.-C.; Casillas, G.; Wolverton, C.; Jose-Yacaman, M.; Seidman, D. N.; Kanatzidis, M. G.; Dravid, V. P. Morphology Control of Nanostructures: Na-Doped PbTe-PbS System. Nano Lett. 2012, 12, 5979-5984. (33) Liu, Y.; Chen, L.; Li, J. Precipitate Morphologies of Pseudobinary – Thermoelectric Compounds. Acta Mater. 2013, 65, 308-315. (34) Xiao, Y.; Wu, H.; Li, W.; Yin, M.; Pei, Y.; Zhang, Y.; Fu, L.; Chen, Y.; Pennycook, S. J.; Huang, L.; He, J.; Zhao, L.-D. Remarkable Roles of Cu to Synergistically Optimize Phonon and Carrier Transport in n-Type PbTe-Cu2Te. J. Am. Chem. Soc. 2017, 139, 18732-18738.

21 ACS Paragon Plus Environment

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

(35) Gottstein, G., Physical Foundations of Materials Science. 1st ed.; Springer-Verlag Berlin Heidelberg New York: 2004. (36) Strauss, A. J. Effect of Pb- and Te-Saturation on Carrier Concentrations in ImpurityDoped PbTe. Journal of Electronic Materials 1973, 2, 553-569. (37) Bergum, K.; Ikeda, T.; Jeffrey Snyder, G. Solubility and Microstructure in the Pseudo-Binary PbTe-Ag2Te System. Journal of Solid State Chemistry 2011, 184, 2543-2552. (38) You, L.; Liu, Y.; Li, X.; Nan, P.; Ge, B.; Jiang, Y.; Luo, P.; Pan, S.; Pei, Y.; Zhang, W.; Snyder, G. J.; Yang, J.; Zhang, J.; Luo, J. Boosting the Thermoelectric Performance of PbSe through Dynamic Doping and Hierarchical Phonon Scattering. Energy & Environmental Science 2018, 11, 1848-1858. (39) Kittel, C., Introduction to Solid State Physics. 6th ed.; John Wiley & Sons Inc.: 1986. (40) Kim, H.-S.; Gibbs, Z. M.; Tang, Y.; Wang, H.; Snyder, G. J. Characterization of Lorenz Number with Seebeck Coefficient Measurement. APL Materials 2015, 3, 041506. (41) Callaway, J. Model for Lattice Thermal Conductivity at Low Temperatures. Physical Review 1959, 113, 1046. (42) He, J.; Zhao, L.-D.; Zheng, J.-C.; Doak, J.; Wu, H.; Wang, H.-Q.; Lee, Y.; Wolverton, C.; Kanatzidis, M.; Dravid, V. Role of Sodium Doping in Lead Chalcogenide Thermoelectrics. J. Am. Chem. Soc. 2013, 135, 4624-7. (43) Liu, W.; Hu, J.; Zhang, S.; Deng, M.; Han, C.-G.; Liu, Y. New Trends, Strategies and Opportunities in Thermoelectric Materials: A Perspective. Materials Today Physics 2017, 1, 50-60.

22 ACS Paragon Plus Environment

Page 22 of 38

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

ACS Applied Materials & Interfaces

(44) Wu, H.; Carrete, J.; Zhang, Z.; Qu, Y.; Shen, X.; Wang, Z.; Zhao, L.-D.; He, J. Strong Enhancement of Phonon Scattering through Nanoscale Grains in Lead Sulfide Thermoelectrics. NPG Asia Materials 2014, 6, e108.

23 ACS Paragon Plus Environment

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

Figures and Tables

Figure 1. Micrographs representing secondary electrons (SE) signal collected by Ga+ ion excitation in the dual-beam focused ion beam (FIB), taken from the (a) as-quenched (AQ) sample as well as those aged for (b) 6 h and (c) 48 h at 380 °C. Large Ag2Te-precipitates appear as elongated bright spots on the background of the PbTe-based matrix. 24 ACS Paragon Plus Environment

Page 24 of 38

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

ACS Applied Materials & Interfaces

Figure 2. (a) A dark field transmission electron microscopy (TEM) micrograph taken from the 48 h aged sample along the [011] zone axis and the (b) corresponding selected area electron diffraction (SAED) pattern, showing monoclinic Ag2Te-precipitates (red indices) oriented with respect to the PbTe-matrix (black indices) as [011](100)𝑃𝑏𝑇𝑒 || [010](100)𝐴𝑔2𝑇𝑒 (c) A scanning transmission electron microscopy (STEM) micrograph showing several precipitates and (d) corresponding energy dispersive x-ray spectroscopy (EDS) map of a region containing two precipitates, marked by red rectangle in (c), indicating the nearly 2:1 stoichiometry of Ag:Te. 25 ACS Paragon Plus Environment

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

Figure 3. Three-dimensional (3D) atom probe tomography (APT) reconstructed volumes acquired from the (a) as-quenched (AQ) alloy and those aged for (b) 6 h and (c) 48 h. For clarity, only the Te ions are marked (green dots). Ag-rich precipitates are indicated by isoconcentration surfaces, marked in red, whose Ag-concentrations are indicated by the values marked in red.

26 ACS Paragon Plus Environment

Page 26 of 38

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

ACS Applied Materials & Interfaces

Figure 4. Typical proximity histograms (proxigrams) acquired across the interfaces between (a) large and (b) small Ag-rich precipitates and the PbTe-matrix of the asquenched (AQ) sample, indicating a difference in the Ag/Te stoichiometry of the precipitates.

27 ACS Paragon Plus Environment

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

Figure 5. The temperature dependent (a) electrical conductivity, (b) Seebeck coefficients, and (c) thermal conductivity, measured for the as-quenched (AQ) sample (red circles), as well as for those aged at 380 °C for 6 h (blue squares) and 48 h (olive diamonds). 28 ACS Paragon Plus Environment

Page 28 of 38

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

ACS Applied Materials & Interfaces

Figure 6. Ag-rich precipitates number density, Nv, (red bars) evaluated from complementary transmission electron microscopy (TEM) and atom probe tomography (APT) for the different aging times. The blue bars indicate the average concentration of Ag evaluated from APT analysis for the inter-precipitate regions, for the respective aging times.

29 ACS Paragon Plus Environment

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

Figure 7. The temperature dependent lattice (open symbols) and electronic (filled symbols) components of thermal conductivity, evaluated for the as-quenched (AQ) sample (red circles), as well as for those aged at 380 °C for 6 h (blue squares) and 48 h (olive diamonds) based on the measured thermal conductivities and the Wiedemann-Franz relationship.

30 ACS Paragon Plus Environment

Page 30 of 38

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

ACS Applied Materials & Interfaces

Table 1. The Ag2Te-precipitate number density, Nv, evaluated from secondary electron (SE) signal initiated by Ga+ ion beams in the dual-beam focused ion beam (FIB) and complementary transmission electron microscopy (TEM) and atom probe tomography (APT) for the different aging times (1st and 2nd columns, respectively). The 4th - 6th columns indicate the average matrix composition evaluated from APT analysis for the interprecipitate regions.

Aging time [h]

Nv, SE/FIB [m-3]

Nv, TEM/APT [m-3]

Pb [at. %]

Te [at. %]

Ag [at. %]

0 (AQ)

(5.0 ± 1.5)·1017

(3.61 ± 1.51)·1021

44.88 ± 0.23

54.33 ± 0.23

0.78 ± 0.04

6

(1.07 ± 0.45)·1018

(1.68 ± 0.92)·1024

40.94 ± 0.26

58.47 ± 0.26

0.62 ± 0.04

48

(4.0 ± 1.5)·1017

(1.21 ± 0.75)·1022

39.67 ± 0.25

59.71 ± 0.24

0.62 ± 0.04

31 ACS Paragon Plus Environment

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

Caption : Figure 1. Micrographs representing secondary electrons (SE) signal collected by Ga+ ion excitation in the dual-beam focused ion beam (FIB), taken from the (a) as-quenched (AQ) sample as well as those aged for (b) 6 h and (c) 48 h at 380 °C. Large Ag2Te-precipitates appear as elongated bright spots on the background of the PbTe-based matrix. 99x209mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 32 of 38

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

ACS Applied Materials & Interfaces

Figure 2. (a) A dark field transmission electron microscopy (TEM) micrograph taken from the 48 h aged sample along the [01 1] zone axis and the (b) corresponding selected area electron diffraction (SAED) pattern, showing monoclinic Ag2Te-precipitates (red indices) oriented with respect to the PbTe-matrix (black indices) as [011 ](100)_PbTe ||[010](100)_(〖Ag〗_2 Te) (c) A scanning transmission electron microscopy (STEM) micrograph showing several precipitates and (d) corresponding energy dispersive x-ray spectroscopy (EDS) map of a region containing two precipitates, marked by red rectangle in (c), indicating the nearly 2:1 stoichiometry of Ag:Te. 199x199mm (300 x 300 DPI)

ACS Paragon Plus Environment

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

Figure 3. Three-dimensional (3D) atom probe tomography (APT) reconstructed volumes acquired from the (a) as-quenched (AQ) alloy and those aged for (b) 6 h and (c) 48 h. For clarity, only the Te ions are marked (green dots). Ag-rich precipitates are indicated by iso-concentration surfaces, marked in red, whose Agconcentrations are indicated by the values marked in red. 149x149mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 34 of 38

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

ACS Applied Materials & Interfaces

Figure 4. Typical proximity histograms (proxigrams) acquired across the interfaces between (a) large and (b) small Ag-rich precipitates and the PbTe-matrix of the as-quenched (AQ) sample, indicating a difference in the Ag/Te stoichiometry of the precipitates. 131x199mm (300 x 300 DPI)

ACS Paragon Plus Environment

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

Figure 6. Ag-rich precipitates number density, Nv, (red bars) evaluated from complementary transmission electron microscopy (TEM) and atom probe tomography (APT) for the different aging times. The blue bars indicate the average concentration of Ag evaluated from APT analysis for the inter-precipitate regions, for the respective aging times. 208x159mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 36 of 38

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

ACS Applied Materials & Interfaces

Figure 7. The temperature dependent lattice (open symbols) and electronic (filled symbols) components of thermal conductivity, evaluated for the as-quenched (AQ) sample (red circles), as well as for those aged at 380 °C for 6 h (blue squares) and 48 h (olive diamonds) based on the measured thermal conductivities and the Wiedemann-Franz relationship. 208x159mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

ACS Paragon Plus Environment

Page 38 of 38