Enhanced Thermoelectric Properties in Bulk Nanowire Heterostructure

Subscriber access provided by Johns Hopkins Libraries

Communication

Enhanced Thermoelectric Properties in Bulk Nanowire Heterostructurebased Nanocomposites through Minority Carrier Blocking Haoran Yang, Je-Hyeong Bahk, Tristan W Day, Amr Mohammed, G. Jeffrey Snyder, Ali Shakouri, and Yue Wu Nano Lett., Just Accepted Manuscript • DOI: 10.1021/nl504624r • Publication Date (Web): 09 Jan 2015 Downloaded from http://pubs.acs.org on January 10, 2015

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

Nano Letters 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 21

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

Nano Letters

Enhanced Thermoelectric Properties in Bulk Nanowire Heterostructure-based Nanocomposites through Minority Carrier Blocking

Haoran Yang1, Je-Hyeong Bahk2, Tristan Day3, Amr M. S. Mohammed2, 4, G. Jeffrey Snyder3, Ali Shakouri2, 4 and Yue Wu1, 2, 5 * 1. School of Chemical Engineering, Purdue University, West Lafayette, Indiana 47907, USA 2. Birck Nanotechnology Center, Purdue University, West Lafayette, Indiana 47907, USA 3. Materials Science, California Institute of Technology, Pasadena, California 91125, USA 4. School of Electrical and Computer Engineering, Purdue University, West Lafayette, Indiana 47907, USA 5. Department of Chemical and Biological Engineering, Iowa State University, Ames, Iowa 50011, USA

ACS Paragon Plus Environment

1

Nano Letters

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 21

ABSTRACT: To design superior thermoelectric materials, the minority carrier blocking effect, in which the unwanted bipolar transport is prevented by the interfacial energy barriers in the heterogeneous nanostructures, has been theoretically proposed recently. The theory predicts an enhanced power factor and a reduced bipolar thermal conductivity for materials with a relatively low doping level, which could lead to an improvement in the thermoelectric figure of merit (ZT). Here we show the first experimental demonstration of the minority carrier blocking in lead telluride-silver telluride (PbTe-Ag2Te) nanowire heterostructure-based nanocomposites. The nanocomposites are made by sintering PbTe-Ag2Te nanowire heterostructures produced in a highly-scalable solution-phase synthesis. Compared with Ag2Te nanowire-based nanocomposite produced in similar method, the PbTe-Ag2Te nanocomposite containing ~5 atomic % PbTe exhibits enhanced Seebeck coefficient, reduced thermal conductivity, and ~40% improved ZT, which can be well explained by the theoretical modeling based on the Boltzmann transport equations when energy barriers for both electrons and holes at the heterostructure interfaces are considered in the calculations. For this p-type PbTe-Ag2Te nanocomposite, the barriers for electrons, i.e. minority carriers, are primarily responsible for the ZT enhancement. By extending this approach to other nanostructured systems, it represents a key step towards low-cost solutionprocessable nanomaterials without heavy doping level for high-performance thermoelectric energy harvesting.

KEYWORDS: Thermoelectric, heterostructures, PbTe, Ag2Te, nanocomposites, minority carrier blocking.


2

Page 3 of 21

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

Nano Letters

A high thermoelectric figure of merit (ZT) requires simultaneous optimization of electrical conductivity (σ), thermal conductivity (κ), and Seebeck coefficient (S) in one material, which is a great challenge because these three properties are highly interrelated.1,2 To partially decouple σ and κ, a heterogeneous nanostructuring approach has been developed with secondary-phase nano-inclusions introduced into the bulk matrix. The nano-inclusions can be designed to scatter phonons more effectively than electrons, leading to a net enhancement of ZT.3-6 In addition, nano-inclusions can be intentionally tuned to affect electron transport in a constructive way: by selectively blocking low-energy carriers, the Seebeck coefficient and the power factor can be enhanced, which is previously known as the energy filtering effect.3,4,7-10 The fulfillment of energy filtering, however, requires a high carrier density and a precise control of the energy barrier height relative to the Fermi level, which is very difficult to achieve experimentally.9 Recently, Bahk et al. proposed an alternative approach of using heterostructures to block the detrimental minority carriers and ZT is expected to increase due to a reduction of the bipolar thermal conductivity and an increase in power factor even at relatively low doping levels.10 This effect is of great interest especially for thermoelectric nanomaterials produced from scalable solution phase synthesis, since high carrier densities (1019/cm3 to 1021/cm3) usually required for optimal thermoelectric properties3 are very hard to achieve in solution phase synthesized nanomaterials due to the “self-purification” effect.11,12 Ag2Te is a promising thermoelectric material due to its high electron mobility13 and very low intrinsic lattice thermal conductivity.14,15 Recently, our group reported carrier density modulation in solution phase synthesized Ag2Te nanowires.16 Although we showed that the electron density could be greatly increased through silver doping to achieve higher ZT than that of the undoped Ag2Te, the doping process is inefficient as evidenced by the demand for a large amount of extra


3

Nano Letters

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 21

Ag precursor. Here, we report the first experimental demonstration of the minority carrier blocking effect, which is observed in a PbTe-Ag2Te nanowire heterostructure-based nanocomposite system. The dumbbell-shaped PbTe-Ag2Te nanowire heterostructures feature well-defined morphology and size with two PbTe nanocubes grown at the two ends of each Ag2Te nanowire. Using these heterostructures as building blocks, we fabricated PbTe-Ag2Te nanocomposites with PbTe nano-inclusions uniformly distributed in the Ag2Te matrix through hot pressing. We show that the Seebeck coefficient and the power factor are enhanced, while the thermal conductivity is reduced in these nanocomposites, compared with the values of the pure Ag2Te nanowire nanocomposites produced by a similar process.16 As a result, ZT is greatly enhanced by more than 40% to reach ~ 0.66 at 390 K in the PbTe-Ag2Te nanocomposites, even though the system is not intentionally doped and the carrier density still remains low. We propose that the band offset between Ag2Te and PbTe provides energy barriers for electron transport, which introduces the minority carrier blocking effect responsible for the power factor enhancement. Under the minority carrier blocking scheme, the measured thermoelectric transport properties are in excellent agreement with the theoretical calculations based on the Boltzmann transport equations (BTE), taking into account the interfacial energy barriers for both carrier types.


4

Page 5 of 21

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

Nano Letters

Figure 1. Figure 1. Synthesis and characterization of the Te nanowires, PbTe-Te nanowire heterostructures and PbTe-Ag2Te nanowire heterostructures. (a) Schematic for the three-step synthesis of PbTe-Ag2Te nanowire heterostructures; (b) XRD patterns compared with the standards, (c-e) SEM images, (f-h) low-magnification TEM images with high-magnification TEM image insets of the Te nanowires (c and f), PbTe-Te nanowire heterostructures (d and g), and PbTe-Ag2Te nanowire heterostructures (e and h). Scale bars in the main images represent 500 nm and scale bars in the insets represent 50 nm. The synthesis of PbTe-Ag2Te nanowire heterostructures, highlighted in Fig. 1a, is accomplished through a three-step reaction with the details described in the Supporting


5

Nano Letters

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 21

Information. To obtain the nanowire heterostructures, Tellurium (Te) nanowires are synthesized first; a small amount of Lead precursor is then injected, which triggers the growth of PbTe nanocubes at the two ends of each Te nanowire to form dumbbell-like PbTe-Te nanowire heterostructures; and finally the rest of the Te part is further converted into Ag2Te to obtain the desired structures. Notably, this is a very general method developed by our group to produce many kinds of Telluride-based nanowire heterostructures, such as Bi2Te3-Te, Bi2Te3-PbTe and Bi2Te3-Ag2Te nanowire heterostructures.17-19 The conversion from Te to Ag2Te follows the same procedure as described in our previous report on the synthesis of undoped Ag2Te nanowires.16 Fig. 1b presents the powder X-ray diffraction (XRD) patterns of the products from each step. After the first step, the XRD pattern of the product (Fig. 1b top) can be readily indexed to the pure hexagonal Te phase (JCPDS 36-1452). After the second step, another set of peaks corresponding to the altaite PbTe (JCPDS 38-1435) appear in addition to the Te peaks, indicating the formation of PbTe. After the final step, the Te peaks are replaced by the monoclinic Ag2Te peaks (JCPDS 34-0142) in the XRD pattern (Fig. 1b bottom), indicating complete transformation of Te into Ag2Te. The microstructures of the products are further investigated by electron microscopy. The lowmagnification scanning electron microscopy (SEM) images (Fig. 1c to 1e) show that the Te nanowires, dumbbell-like PbTe-Te, and PbTe-Ag2Te nanowire heterostructures from each step are all uniform, which proves the excellent control of the morphology of these nanostructures. The transmission electron microscopy (TEM) images (Fig. 1f to 1h) reveal more details of the nanostructures. As shown in Fig. 1f, the Te nanowires synthesized in the first step have an average length of 1563±34 nm and an average diameter of 19.9±0.5 nm. The products obtained after the second step (Fig. 1e) contain two PbTe nanocubes grown at the two ends of each Te


6

Page 7 of 21

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

Nano Letters

nanowire. The Te part of the nanowire heterostructure grows slightly longer and thicker compared to the pure Te nanowires, the average diameter reaches 23.6±0.9 nm and the average length reaches 1723±33 nm. Meanwhile, the PbTe nanocubes have an average side length of 34.4±1.8 nm. The increase of the length and diameter of the Te part in the nanowire heterostructure is likely due to the elevated reaction temperature in the second step where the unreacted Te precursors decompose and incorporate into the existing Te nanowires (see Methods). After the completion of the final step, as shown in Fig. 1h, the PbTe nanocubes remain almost unchanged in size and shape with an average side length of 34.6±1.5 nm. On the contrary, the originally straight Te parts transform into curved Ag2Te nanowires, and the average diameter become 27.7±1.0 nm. Since the nanowires become curved, the length distribution cannot be easily measured accurately and the average length is increased to about 2 µm. The diameter and length growth during this self-templated transformation from Te to Ag2Te is likely due to the volume expansion from the hexagonal Te lattice to the monoclinic Ag2Te lattice.20 According to ref. 20, the mechanical stress accumulation due to large volume change during the transformation could break the nanowires into shorter nanorods. In our study, however, the nanowires remained intact after transformation. We believe the stress is possibly relaxed by forming bends and kinks. Similar morphology has been observed in many other solution phase synthesized Ag2Te nanowires.21-23


7

Nano Letters

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 8 of 21

Figure 2. HRTEM characterizations. (a), (b), and (c) are the HRTEM images of the Te wire part, the PbTe head part, and the junction part of a typical PbTe-Te nanowire heterostructure. The insets are the corresponding FFT images. (d) and (e) are the HRTEM images of the PbTe head part and the Ag2Te wire part of a typical PbTe-Ag2Te nanowire heterostructure. (f) and (g) are the FFT images of the wire region (red solid rectangle) and the head region (blue dashed rectangle) of the PbTe-Te nanowire heterostructure in (c). (h) and (i) are the FFT images of (d) and (e). In order to further understand the formation of the dumbbell structures, high-resolution transmission electron microscopy (HRTEM) has been used to study the wire part, the head part, and the junction part of the PbTe-Te and PbTe-Ag2Te nanowire heterostructures, respectively.


8

Page 9 of 21

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

Nano Letters

As shown in the HRTEM image (Fig. 2a) and its fast Fourier transform (FFT, Fig. 2a inset), the wire part of the PbTe-Te heterostructure is single crystalline hexagonal Te with growth direction along [001]. The head part (Fig. 2b) is indexed to single crystalline altaite PbTe phase. HRTEM image (Fig. 2c) and the corresponding FFT images (Fig. 2f, and 2g) at the junction part between Te wire and PbTe head reveal several important features. First, PbTe heads grow epitaxially on Te nanowires, and the (111) planes of PbTe heads align in parallel with the (001) planes of Te nanowires. Crystallographically, the (111) planes of the fcc PbTe and the (001) planes of the hexagonal Te share the same close packed atomic arrangement,24 and the in-plane lattice mismatch is only 2.4% while the cross plane lattice mismatch is 5.6%, which is schematically shown in Supporting Information Fig. S1. The small lattice mismatch allows for epitaxial growth of PbTe on Te. Second, the PbTe heads wrap and cover the tips of the Te wires. After converting the Te parts into Ag2Te, the HRTEM of the heads (Fig. 2d) and its corresponding FFT pattern (Fig. 2h) show that the heads remain unchanged as PbTe, however, the nanowire parts indeed transform into monoclinic Ag2Te (Fig. 2e and 2i). As shown in Supporting Information Fig. S2, the junctions between the PbTe heads and the Ag2Te wires become not clearly defined, and various features at the boundaries have been observed. This is probably due to the fast diffusion of Ag into the Te lattice.20,25 The well-controlled synthesis of PbTe-Ag2Te nanowire heterostructures allows us to use them as building blocks to fabricate bulk nanocomposites with Ag2Te as the primary phase (matrix) and PbTe as secondary-phase (nano-inclusions) through hot pressing. Before hot pressing, the electrically insulating surface ligands on the heterostructures are removed with hydrazine hydrate treatment.17 We keep the hot press temperature low (423 K) intentionally to avoid grain growth as well as alloying and elemental diffusion between PbTe and Ag2Te, which could become


9

Nano Letters

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 10 of 21

significant when the temperature is above 420 K according to the PbTe-Ag2Te phase diagram.15,26

Figure 3. Characterization of the hot pressed PbTe-Ag2Te nanocomposite sample. (a) XRD, (b) cross-section SEM image, and (c) HRTEM image, in which the dashed blue line defined region is PbTe while the solid purple lines defined regions are Ag2Te. Inset of (a) shows the photograph of a hot-pressed bulk PbTe-Ag2Te nanocomposite sample disk. The hot-pressed sample (Fig. 3a, inset) is re-examined by XRD and the diffraction pattern is presented in Fig. 3a. The diffraction pattern of the hot-pressed disk is identical to the diffraction


10

Page 11 of 21

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

Nano Letters

pattern of the as-synthesized PbTe-Ag2Te nanowire heterostructures, which confirms that hot pressing does not result in any noticeable phase change. Additionally, the XRD pattern from the hot-pressed disk shows a similar degree of peak broadening, which indicates that the grain growth during hot pressing is negligible. Indeed, the SEM image (Fig 3b) of the cross section of the hot pressed sample shows that the original nanowires are broken during hot pressing, which possibly occurs at the bend or kink sites, with a lot of short rod-like grains as long as 150 nm and with diameters around 30 to 40 nm. The SEM observation confirms that grain growth is insignificant and the morphological features of the raw materials are partially preserved in the hot pressed sample. To further confirm the phase separation, HRTEM studies are performed on the hot pressed sample. Fig. 3c shows that distinguishable PbTe grains and Ag2Te grains around 30~50 nm coexist, which verifies that the heterogeneity of the raw material is preserved after hot pressing and a uniform distribution of PbTe nano-inclusions in the Ag2Te matrix structure is obtained. In both the SEM and HRTEM images, nanoscale pores can be seen, which is consistent with the fact that the relative density of the hot-pressed pellet is about 88.5% of the theoretical density. In comparison, the pure Ag2Te nanowires-based nanocomposite sample we previously reported has a similar relative density of 88.7%. The composition of the binary composite is determined to be (PbTe)0.049(Ag2Te)0.951 by energy dispersive X-ray spectroscopy (EDX), and the cation/anion ratio is close to stoichiometric ((Ag/2+Pb)/Te=100/98.6).


11

Nano Letters

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 12 of 21

Figure 4. Experimental data and theoretical fitting of the thermoelectric properties. (a) Electrical conductivity, (b) Seebeck coefficient, (c) power factor, (d) thermal conductivity, and (e) ZT of the PbTe-Ag2Te nanocomposite compared with those of the Ag2Te nanocomposite. (f) The schematic band alignment between PbTe and Ag2Te. Fig. 4a to Fig 4e present the electrical and thermal transport properties of the PbTe-Ag2Te nanocomposite sample at temperatures ranging from 300 K to 390 K, which is compared with our previously reported undoped Ag2Te nanowire-based nanocomposite sample hot pressed


12

Page 13 of 21

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

Nano Letters

under the identical conditions.16 Also shown are the fitting curves calculated based on the Boltzmann transport equations with barrier heights for both carrier types.9,10 Fig. 4f schematically shows the presumed band alignment between Ag2Te and PbTe in the nanocomposite. Because Ag2Te has a small band gap of ~0.05 eV14 whereas PbTe has a wider band gap of 0.31 eV27 at room temperature, after the Fermi level alignment, the discontinuity in the conduction band and the valence band will form energy barriers for both electrons and holes. Since the atomic percentage of PbTe is small, we can model the system such that most of the diffusive transport is assumed to be occurring in the Ag2Te matrix, and PbTe nano-inclusions simply act as heterostructure barriers to fit the experimental data. From the calculation, we find that the energy barrier heights that best fit the experimental data are ~110 meV and ~45 meV for electrons and holes, respectively. Note that these effective energy barrier heights are not simply the band offsets, because the charge transfer between the two materials can induce band bending at the interface, so that the barrier heights are adjusted. Also, tunneling through the narrowed potential width near the interface, as well as the non-negligible barrier influence on high-energy carriers can potentially vary the effective barrier height.11 As shown in Fig. 4a, the electrical conductivity of the PbTe-Ag2Te nanocomposite is lower than that of the Ag2Te sample over the entire temperature range. Such a reduction in the electrical conductivity is a result of the electron and hole blocking by the heterostructure barriers. The electron and hole mobilities of the Ag2Te sample at room temperature are estimated to be 837 and 169 cm2V-1s-1, respectively. Due to the scattering from the PbTe nano-inclusions, the electron and hole mobilities at room temperature are reduced to 289 and 103 cm2V-1s-1, respectively. The carrier concentrations that participate in the transport are also reduced by the


13

Nano Letters

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 14 of 21

barrier blocking, which, combined with the mobility reduction, lowered the electrical conductivity as anticipated in the theory.9 On the other hand, the Seebeck coefficient of the PbTe-Ag2Te sample is increased significantly compared to that of the Ag2Te sample over the entire temperature range. At 380 K, the peak Seebeck coefficient reaches 271 µV/K for the PbTe-Ag2Te nanocomposite, while for the Ag2Te sample it is only 192 µV/K, corresponding to a more than 40 % enhancement in the PbTe-Ag2Te nanocomposite. The Seebeck coefficient in the bipolar transport regime is given by28

,

(1)

where n and p are electron and hole densities respectively, the subscripts e and h refer to the partial properties of electrons and holes respectively, and b is the mobility ratio, i.e. b = µe/µh. Since electrons and holes have partial Seebeck coefficients with opposite signs, the absolute value of the total Seebeck becomes smaller than those of the partial Seebeck coefficients according to Eq. (1). Previously, we reported that the “undoped” Ag2Te nanowire based sample has a significant contribution from minority carriers (electrons) in the Seebeck coefficient even though the electron density (1×1017 cm-3) is much lower than the hole density (8×1017 cm-3), due to their higher mobility.15 When these minority electrons are effectively blocked from participating in transport by the heterostructure barriers, their contribution to the Seebeck coefficient is reduced, and thus the magnitude of the total Seebeck coefficient is increased. As shown in Fig. 4b, the theoretical calculations show that the enhancement of the Seebeck coefficient in the PbTe-Ag2Te sample is primarily due to the blocking of the electrons that have a negative contribution to the positive total Seebeck coefficient. Meanwhile, the barrier for holes is found to be detrimental, because when holes are blocked by the barriers in the valence band, the


14

Page 15 of 21

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

Nano Letters

relative contribution of electrons becomes more significant, which results in the reduction of the total Seebeck coefficient. Consequently, if there were hole barriers only, the magnitude of the total Seebeck coefficient would have been even smaller than that of the Ag2Te sample due to the increased electron contribution. This is clearly shown in Fig. 4b, in which the simulated cases of only hole barriers and only electron barriers are separately presented. When only electron barriers are assumed to exist in the material, the Seebeck coefficient is overestimated at low temperatures. At high temperatures above 360 K, however, a small effect of hot-carrier energy filtering comes into play for the holes when the hole barriers are also included because the Fermi level approaches to the optimal position as temperature increases. This results in a slight increase in the Seebeck coefficient compared to that in the case of only electron barriers. Only when both the electron and the hole barriers are taken into account in the BTE calculations, the measured Seebeck coefficients can be fit very well over the entire temperature range. Fig. 4c presents the measured and calculated power factors. In the whole measured temperature range, the PbTe-Ag2Te sample exhibits larger power factors than those of the Ag2Te sample, and the relative power factor increase is around 17% to 38%. As shown as a separate curve in Fig. 4c, if only hole barriers were to exist, the power factor would have decreased significantly due to the reduced electrical conductivity and Seebeck coefficient, which indicates that the blocking of majority carriers (holes in this case) has a detrimental influence on the thermoelectric performance in the bipolar transport region. On the contrary, if only electron barriers were to exist, the power factor could have been enhanced even more, by greater than 50 % at around 380 K, and greater than 200 % at around room temperature, mainly due to the lessened electrical conductivity reduction. The calculation suggests that an even higher power factor could


15

Nano Letters

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 16 of 21

be expected if the valence band alignment between the two phases were improved to reduce the effective hole barrier height. The experimental realization is subject to future research. Fig. 4d shows that the measured thermal conductivity of the PbTe-Ag2Te sample is 0.29 W m-1 K-1 at 300 K and decreases to 0.19 W m-1 K-1 as temperature increases to 390 K, which is lower than that of the Ag2Te sample by ~ 35 %. The electronic thermal conductivity (κe) is calculated from the Wiedemann-Franz relation29 using the Lorenz number obtained from our electron transport modeling, and it is seen that the electronic contribution to the total thermal conductivity is very small, less than 0.02 W m-1 K-1 in the PbTe-Ag2Te nanocomposite. This is much lower than that of the Ag2Te sample (0.04 W m-1 K-1) due to the carrier blocking by the barriers. The reduction in total thermal conductivity largely comes from the reduction of the lattice plus bipolar part (κl + κbi), which is obtained by subtracting κe from the total thermal conductivity. Accurate quantification of the bipolar thermal conductivity is difficult since κbi is very sensitive to the band structure but the valence band of Ag2Te has not been well studied so far. Although it is difficult to quantitatively separate κbi from κl, we expect that a larger portion of κ reduction is associated with a reduction in kbi since the electrical measurements clearly indicate a reduced bipolar effect in this sample compared to the Ag2Te sample. On the other hand, although it’s well known that nano-inclusions may introduce phonon scattering to suppress κl,30-32 a significant reduction of κl in the PbTe-Ag2Te nanocomposites would be unlikely because bulk Ag2Te already exhibits a remarkably low κl of 0.1~0.2 W m-1 K-1.14,15 Due to the enhanced power factor and the extremely low thermal conductivity, the figure of merit ZT reaches 0.66 at 390 K for the PbTe-Ag2Te nanocomposite sample, as shown in Fig. 4e, which is, to our best knowledge, the highest ZT among ever reported Ag2Te based thermoelectric materials at the same temperature range. The enhanced ZT demonstrates that the minority carrier


16

Page 17 of 21

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

Nano Letters

blocking with heterostructure barriers is an effective approach for improving the thermoelectric performance of solution-processed nanomaterials. Compared to the conventional hightemperature solid-state synthesis of thermoelectric bulk materials, our novel approach represents an alternative design towards high-performance thermoelectric bulk materials because: (1) it does not require an energy-intensive process to achieve high performance through heavy doping and it bypasses the doping limitation in most typical solution-phase nanostructure growth methods; (2) the colloidal nanowire suspension could enable solution processing of nanomaterials for direct printing of thermoelectric devices; (3) the low-cost solution-phase synthesis approach and the hot press process have been demonstrated to be general and highlyscalable for other Telluride-based material systems based on our previous research. Finally, we would like to discuss the difference between our nano-composite sample and the bulk PbTe-Ag2Te sample with similar composition reported in ref. 14. The bulk composites are n-type materials that contain micron-size PbTe region and the thermoelectric properties are dominated by the Ag2Te matrix, which might suggest an important nanostructuring size effect on the thermoelectric properties of the binary composites, which needs further investigation. In summary, we have developed a novel low-cost scalable solution-phase synthesis of dumbbell-like PbTe-Ag2Te nanowire heterostructures. These nanowire heterostructures have been utilized as building blocks to fabricate bulk PbTe-Ag2Te nanocomposites through hot pressing, after which the PbTe nano-inclusions are uniformly distributed in the Ag2Te matrix. The thermoelectric transport properties of the binary nanocomposite have been studied and compared with our previously reported undoped Ag2Te nanowire-based nanocomposite fabricated under identical conditions. The experimental data have been explained by taking into account the band offset between PbTe and Ag2Te and adding energy barriers for both electrons


17

Nano Letters

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 21

and holes in the BTE calculations. We have shown evidence of power factor enhancement due to minority carrier blocking, which leads to a 40% to 55% enhancement of ZT in the heterostructure nanocomposite. The energy barrier for the majority carriers has been found to be detrimental to the thermoelectric properties, and a further ZT enhancement could be possible if the hole barrier height could be reduced in the future. Furthermore, the minority carrier blocking approach through heterostructuring may be applied to other systems to improve ZT in which heavy doping is difficult to achieve. ASSOCIATED CONTENT Supporting Information. Detailed experimental processes, theoretical modeling procedures, schematic presentation of the lattice mismatch between Te (001) planes and PbTe (111) planes, and HRTEM images showing the junctions of several PbTe-Ag2Te heterostructures. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author *[email protected] Author Contributions Yue Wu led the project. Haoran Yang was the main contributor for the PbTe-Ag2Te heterostructure synthesis, nanocomposite fabrications and structural characterizations. JeHyeong Bahk and Ali Shakouri performed the theoretical modeling. Tristan Day and G. Jeffrey Snyder performed the electrical conductivity measurements. Haoran Yang and Amr M. S. Mohammed performed Seebeck coefficient measurements. Funding Sources


18

Page 19 of 21

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

Nano Letters

This work is supported by US Air Force Office of Scientific Research (Award Number FA955012-1-0061).

REFERENCES (1)

Vineis, C. J.; Shakouri, A.; Majumdar, A.; Kanatzidis, M. G. Adv. Mater. 2010, 22, (36),

3970-3980. (2)

Shakouri, A. Ann. Rev. Mater. Res. 2011, 41, (1), 399-431.

(3)

Zhang, Y.; Stucky, G. D. Chem. Mat. 2013, 26, (1) 837-848.

(4)

Hsu, K. F.; Loo, S.; Guo, F.; Chen, W.; Dyck, J. S.; Uher, C.; Hogan, T.; Polychroniadis,

E. K.; Kanatzidis, M. G. Science 2004, 303, (5659), 818-821. (5)

Wang, H.; Li, J.-F.; Nan, C.-W.; Zhou, M.; Liu, W.; Zhang, B.-P.; Kita, T. Appl. Phys.

Lett. 2006, 88, (9), 092104. (6)

He, J.; Kanatzidis, M. G.; Dravid, V. P. Mater. Today 2013, 16, (5), 166-176.

(7)

Zebarjadi, M.; Esfarjani, K.; Shakouri, A.; Bian, Z.; Bahk, J.-H.; Zeng, G.; Bowers, J.;

Lu, H.; Zide, J.; Gossard, A. J. Electron. Mater. 2009, 38, (7), 954-959. (8)

Faleev, S. V.; Léonard, F. Phys. Rev. B 2008, 77, (21), 214304.

(9)

Bahk, J.-H.; Bian, Z.; Shakouri, A. Phys. Rev. B 2013, 87, (7), 075204.

(10) Bahk, J.-H.; Shakouri, A. Appl. Phys. Lett. 2014, 105, (5), 052106. (11) Erwin, S. C.; Zu, L.; Haftel, M. I.; Efros, A. L.; Kennedy, T. A.; Norris, D. J. Nature 2005, 436, (7047), 91-94.


19

Nano Letters

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 20 of 21

(12) Dalpian, G. M.; Chelikowsky, J. R. Phys. Rev. Lett. 2006, 96, (22), 226802. (13) Taylor, P. F.; Wood, C. J. Appl. Phys. 1961, 32, (1), 1-3. (14) Pei, Y.; Heinz, N. A.; Snyder, G. J. J. Mater. Chem. 2011, 21, (45), 18256-18260. (15) Jung, D.-y.; Kurosaki, K.; Ohishi, Y.; Muta, H.; Yamanaka, S. Mater. Trans. 2012, 53, (7), 1216-1219. (16) Yang, H.; Bahk, J.-H.; Day, T.; Mohammed, A. M. S.; Min, B.; Snyder, G. J.; Shakouri, A.; Wu, Y. Nano Lett. 2014, 14, (9), 5398–5404. (17) Zhang, G.; Fang, H.; Yang, H.; Jauregui, L. A.; Chen, Y. P.; Wu, Y. Nano Lett. 2012, 12, (7), 3627-3633. (18) Fang, H.; Feng, T.; Yang, H.; Ruan, X.; Wu, Y. Nano Lett. 2013, 13, (5), 2058-2063. (19) Fang, H.; Yang, H.; Wu, Y. Chem. Mat. 2014, 26, (10), 3322-3327. (20) Moon, G. D.; Ko, S.; Xia, Y.; Jeong, U. ACS Nano 2010, 4, (4), 2307-2319. (21) Xiao, F.; Chen, G.; Wang, Q.; Wang, L.; Pei, J.; Zhou, N. J. Solid State Chem. 2010, 183, (10), 2382-2388. (22) Pei, J.; Chen, G.; Jia, D. C.; Yu, Y. G.; Sun, J. X.; Xu, H. M.; Qiu, Z. Z. New J. Chem. 2014, 38, (1), 59-62. (23) Manzoor, S.; Liu, Y.; Yu, Z.; Fu, X.; Ban, G. J. Nanomater. 2014, 2014. (24) Callister, W. D., Materials Science and Engineering: An Introduction, 8th Edition. John Wiley & Sons, Inc.: Hoboken, 2009.


20

Page 21 of 21

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

Nano Letters

(25) Zuo, P.; Zhang, S.; Jin, B.; Tian, Y.; Yang, J. J. Phys. Chem. C 2008, 112, (38), 1482514829. (26) Lensch-Falk, J. L.; Sugar, J. D.; Hekmaty, M. A.; Medlin, D. L. J. Alloys Compd. 2010, 504, (1), 37-44. (27) Madelung, O., Semiconductors: Data Handbook. Springer: Berlin, 2004. (28) Rowe, D. M., Modern Thermoelectrics. Reston Publishing Co.: Reston, 1983. (29) Minnich, A. J.; Dresselhaus, M. S.; Ren, Z. F.; Chen, G. Energ. Environ. Sci. 2009, 2, (5), 466-479. (30) Zhao, L. D.; Wu, H. J.; Hao, S. Q.; Wu, C. I.; Zhou, X. Y.; Biswas, K.; He, J. Q.; Hogan, T. P.; Uher, C.; Wolverton, C.; Dravid, V. P.; Kanatzidis, M. G. Energ. Environ. Sci. 2013, 6, (11), 3346-3355. (31) Zhao, X. Y.; Shi, X.; Chen, L. D.; Zhang, W. Q.; Bai, S. Q.; Pei, Y. Z.; Li, X. Y.; Goto, T. Applied Physics Letters 2006, 89, (9), 092121. (32) Zhang, T.; Zhang, Q.; Jiang, J.; Xiong, Z.; Chen, J.; Zhang, Y.; Li, W.; Xu, G. Appl. Phys. Lett. 2011, 98, (2), 022104.


21

Enhanced Thermoelectric Properties in Bulk Nanowire Heterostructure

Recommend Documents