저작자표시-비영리-변경금지 2.0 대한민국 이용자는 ... - S-Space

(1)

저작자표시-비영리-변경금지 2.0 대한민국 이용자는 아래의 조건을 따르는 경우에 한하여 자유롭게

l 이 저작물을 복제, 배포, 전송, 전시, 공연 및 방송할 수 있습니다. 다음과 같은 조건을 따라야 합니다:

l 귀하는, 이 저작물의 재이용이나 배포의 경우, 이 저작물에 적용된 이용허락조건 을 명확하게 나타내어야 합니다.

저작권법에 따른 이용자의 권리는 위의 내용에 의하여 영향을 받지 않습니다. 이것은 이용허락규약(Legal Code)을 이해하기 쉽게 요약한 것입니다.

Disclaimer

저작자표시. 귀하는 원저작자를 표시하여야 합니다.

비영리. 귀하는 이 저작물을 영리 목적으로 이용할 수 없습니다.

변경금지. 귀하는 이 저작물을 개작, 변형 또는 가공할 수 없습니다.

(2)

공학석사학위논문

Synthesis of chalcogenide compounds for energy harvesting applications

2017년 2월

서울대학교 대학원 화학생물공학부

신 호 철

(3)

공학석사학위논문

2017년 2월

서울대학교 대학원 화학생물공학부

신 호 철

(4)

지도 교수 정 인

이 논문을 공학석사 학위논문으로 제출함

2016년 12월

서울대학교 대학원

화학생물공학부

신 호 철

신호철의 공학석사 학위논문을 인준함

2017 년 2 월

위 원 장 (인) 부위원장 (인) 위 원 (인)

(5)

i

Synthesis of chalcogenide compounds for energy harvesting applications

Abstract

Hocheol Shin

The effect of ZnTe incorporation on PbTe to increase thermoelectric performance by valence band convergence was examined. Also structural and spectroscopic characterizations were noted from the theoretical calculations, and thermoelectric transport properties of Pb0.985Na0.015Te-x%

ZnTe (x = 0, 1, 2, 4). The fact that solid solubility limit of ZnTe in PbTe is less than 1 mol% is studied. A tendency of Hall coefficients with ZnTe contents implies that valence band is a little converged. Therefore, there was no significant

(6)

ii

enhancement in the Seebeck coefficient. But, the lattice thermal conductivity was reduced from introduction of ZnTe in p-type Pb0.985Na0.015Te. In conclusion, the spark plasma sintered Pb0.985Na0.015Te-2% ZnTe reaches the best ZT of 1.73 at 700K, which arises from a relatively high power factor of ~27.0 μW cm^-1 K^-2 and a low lattice thermal conductivity of ~0.69 W m^-1 K^-1 at ~700 K.

Keyword :

Thermoelectric material, Lead telluride , figure of merit ZT, Zinc teluride

Student Number : 2015-21070

(7)

iii

PART 1. INTRODUCTION

The lack of fossil energy and available energy is a big problem for mankind in the near future. In order to solve this problem, the recycling of waste heat has been emerging as a new concept of energy development. Thermoelectric (TE) Materials are semiconductor systems that enable direct conversion between thermal energy and electrical energy. It is with future-oriented characteristics as clean energy, therefore it can utilize waste heat as a renewable energy for industry. Therefore, thermoelectric materials (TE) have broad fields of applications.^1-3 The key parameter that defines the efficiency of TE materials is the dimensionless figure of merit ZT=(σS²T)/κ. To obtain a high ZT, both Seebeck coefficient (S) and electrical conductivity (σ) should be large in the meantime poor thermal conductivity (κ). To be more specific about thermal conductivity (κ), it consists of

(10)

2

κe and κl which are the electron and lattice contributions, respectively. The three factors (σ, S, κ) in the bulk material are each in a mutually related. For example, as the number of charge carriers increases the electrical conductivity but at the same time decreases the Seebeck coefficient and increases the electrical thermal conductivity. Control of this complementary relationship is a main factor that develops of materials with high performance indices. Depending on the temperature range the performance of thermoelectric materials differs. For instance, Bi-Te system is superior at about 300~500K and Sn-Se system is better at 900K or higher, and PbTe-based compounds are top TE materials operating in the mid-temperature range of 500 – 900 K.^4-14

Recently, their ZT values have been considerably advanced by many innovative strategies such as nanostructuring,^15, ¹⁶ hierarchical architecturing,¹⁷ and electronic band structure engineering^18,19. Their availability of

(11)

3

facile doping or alloying mainly attributes improvements. For instance, 4 mole% SrTe doping on PbTe coupled with spark plasma sintering (SPS) results in a remarkably high ZT value of ~2.2 at 915 K because of multi scale hierarchical architecture for maximally reducing lattice thermal conductivity.¹⁷ Higher degree of alloying with SrTe permits further enhancement of ZT that enables the noticeable improvement of Seebeck coefficient owing to valence band convergence.²⁰ The group 12 elements of Zn, Cd, and Hg are also reported as effective dopants to PbTe.^21-24 For CdTe- and HgTe-doped PbTe, the specimens prepared by SPS shows higher ZT than those by melt-processed ingot. For ZnTe-doped PbTe, TE properties of melt-processed samples that show n-type behavior were only studied.^25-28 Therefore, the p-type ZnTe-doped PbTe fabricated by SPS deserves to be studied for high TE performance by valence band convergence as p-type SrTe-doped PbTe.

(12)

4

In this work, the TE properties of the Pb0.985Na0.015Te- x% ZnTe (x = 0, 1, 2, 4) system processed by SPS is reported. Furthermore, the detailed investigations of structural, spectroscopic, electrical, and thermal transport properties as well as the results of ab initio density functional theory calculations also are studied. Unlike the PbTe-SrTe system, the solubility limit of ZnTe in PbTe is less than 1%

and consequently the effect of valence band convergence is negligible, but the modest reduction in lattice thermal conductivity occurs upon ZnTe incorporation. Consequently, the Pb0.985Na0.015Te-2% ZnTe shows the enhanced ZT value of ~1.73 at ~700 K compared to that of Pb0.985Na0.015Te without the incorporation of ZnTe (ZT ~ 1.53).

(13)

5

PART 2. EXPERIMENTAL SECTION

2.1 Synthesis of bulk ingot and powder processing.

Stoichiometric amounts of elemental Pb, Na, Zn, and Te were weighed for the synthesis of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4) and loaded into a carbon-coated fused silica tube. The tube was flame-sealed under a vacuum of ~1.33 x 10^-3 Pa. The tube was then heated to 1323 K for 6 h, held at the same temperature for 12 h, and then water-quenched followed by vacuum annealing at 873 K for 3 days. The cast ingot samples were manually ground with a mortar and pestle in an Ar-filled glove box, and the ground powder was finally sieved under 45 μm.

2.2 Spark plasma sintering.

The ground bulk powders of the samples of

(14)

6

Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4) were loaded into a graphite die in an Ar-filled glove box and cold-compacted manually. The graphite die was taken out of the glove box and placed in a spark plasma sintering system (SPS-211Lx, Fuji Electronic Industrial Co., Japan). The sample chamber was evacuated to a vacuum of ~1.4  10⁻² Torr, and the powder samples in a graphite die were consolidated at ~823 K for 5 min by spark plasma sintering under an axial pressure of 50 MPa in vacuum.

2.3 Powder X-ray diffraction (XRD) and Infrared spectroscopy.

Powder XRD analysis was performed using a SmartLab Rigaku powder X-ray diffractometer (Cu Kα radiation) operating at 40 kV and 30 mA. Fourier transform infrared spectroscopy (FT-IR) spectra were recorded using a Vertex

(15)

7

70 Bruker FT-IR spectrometer equipped with a DLATGS detector and a KBr substrate with a multilayer coating beam splitter.

2.4 First-principles electronic structure calculations.

To figure out the effect of ZnTe substitution for Pb in PbTe, we carried out first-principles electronic structure calculations within the density functional theory formalism.

We utilized the projector augmented wave method²⁹ with the plane wave basis set, implemented in Vienna Ab-initio Simulation Package (VASP)^{30, 31}. For exchange correlation functional, we employed the generalized gradient approximation within the Perdew-Burke-Ernzerhof formalism³². Spin-orbit interaction (SOI) was included to accurately describe the effect of large SOI of Pb. To describe

~1% ZnTe substitution and ~2% Na doping in PbTe, we used 3 × 3 × 3 supercell of the conventional unit cell which

(16)

8

accommodates 216 atoms. For ZnTe substitution, one Zn atom was substituted for one Pb atom, and for Na doping two Pb atoms was replaced by two Na atoms. Totally, four configurations (pristine, ZnTe substitution, Na doping, and ZnTe substitution with Na doping) were considered, and their internal coordinates were optimized with force criterion of 0.01 eV/Å. Experimental lattice parameters were used in the calculations.

2.5 Electrical transport properties.

The samples after SPS were cut for electrical and thermal property measurements using a diamond saw and polished with ethanol using a polishing machine under N2

atmosphere. The polished samples are rectangular-shaped with dimensions of ~3 mm  3 mm  9 mm. The longer direction coincides with the direction in which the electrical

(17)

9

conductivity is measured. The electrical conductivity σ and the Seebeck coefficient S were measured simultaneously under a He atmosphere from room temperature to ~700 K on a ULVAC-RIKO ZEM-3 instrument system.

2.6 Thermal conductivity.

Thermal diffusivity (D) was directly measured under an Ar atmosphere, and the specific heat (CP) was indirectly derived using a standard sample (Pyroceram) as a function of temperature from room temperature to ~700 K using a flash diffusivity method in a Netzsch LFA 457 MicroFlash instrument. In a flash diffusivity method, the front face of a disk-shaped sample is irradiated by a short laser burst, and the resulting rear face temperature rise is recorded and analyzed by an IR detector. The thermal conductivity (κ) was calculated from the equation κ = DCPρ, where ρ is the

(18)

10

density of the sample.

2.7 Hall measurement.

The Hall coefficient (RH) were measured in the temperature range of ~300 K to ~673 K using a LakeShore 8407 Hall effect measurement system in a magnetic field of 1.5 T under Ar flowing.

(19)

11

PART 3. RESULTS AND DISCUSSION

3.1 Powder X-ray diffraction (XRD) and Infrared spectroscopy analysis.

Powder X-ray diffraction (PXRD) patterns of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4) samples are shown in Figure 1. The samples of x = 0 and 1 are a single phase crystallizing in the NaCl-type structure whereas those of x = 2 and 4 show reflection peaks originating from ZnTe at 25°, 42°, and 50°. Their intensities increase with increasing ZnTe content. This observation indicates that a maximum solubility limit of ZnTe in PbTe is less than 1%, consistent with a previous study.^{21, 22} Indeed, the lattice hardly changes upon ZnTe addition. The refined lattice parameters of the samples of x = 0, 1, 2, and 4 are 6.453(1), 6.454(1), 6.453(1), and 6.455(1) Å, respectively. If ZnTe is soluble enough in the PbTe matrix, the lattice would contract

(20)

12

under the same Na concentration because the effective ionic radius of Zn²⁺ (0.74 Å) in octahedral geometry³³ is much smaller than that of Pb²⁺ (1.19 Å) and Na⁺ (1.02 Å).

Infrared absorption spectra of the Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4) support the PXRD results (Figure 2). Doping Na into pure PbTe makes the band gap shifted from ~0.26 eV to <0.1 eV. However, the effect of ZnTe addition is negligible. The absorption spectra of Na-doped samples are similar to p-type NaPb18-xSnxMTe20 (M = Bi, Sb),³⁴ Pb1- xSnxTe,³⁵ and Ag1-x(Pb1-ySny)mSb1-zTem+236. These spectra are also heavily masked by the high carrier concentrations (>10²⁰ cm^-3) in these samples. In comparison, we measured infrared absorption spectra of the samples of PbTe-x%

ZnTe (x = 1, 2, 4) without Na doping and all samples exhibit spectroscopically observable energy band gaps between 0.25 and 0.26 eV, which is nearly same as that of pure PbTe.

(21)

13

Figure 1. Powder X-ray diffraction patterns of samples of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4) and the peaks of ZnTe phase are marked by asterisk.

20 30 40 50 60 70 80

** ** * (511)

(422)

(420)(331)

(400)

(222)(311)

(220)

(200) x = 0

x = 1 x = 2 x = 4

Pb_0.985Na_0.015Te-x% ZnTe

Intensity ( a.u.)

2  (deg.)

(111)

* ZnTe

(22)

14

Figure 2. infrared absorption spectra of Pb1-xZnxTe (x = 0, 0.01, 0.02, 0.04) and Pb0.985Na0.015Te-x% ZnTe (x = 0, 2).

0.1 0.2 0.3 0.4 0.5 0.6

 /S (a.u.)

E (eV)

PbTe Pb_0.99Zn_0.01Te Pb_0.98Zn_0.02Te Pb_0.96Zn_0.04Te Pb_0.985Na_0.015Te

Pb_0.985Na_0.015Te-2% ZnTe

(23)

15

3.2 Electronic band structures of

Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4)

To understand the effect of ZnTe substitution on TE properties of PbTe system, we performed first principles electronic structure calculations within the density functional theory (DFT) formalism for PbTe and its Na and Zn-doped phases. The DFT electronic band structures with spin-orbit coupling of PbTe and PbTe-1% ZnTe show that Zn substitution does not induce the localized defect level between the valence band maximum (VBM) and conduction band minimum (CBM) (Figure 3a and b). Instead, the localized states of Zn emerge in the high energy level of E - EF at 0.6 eV, far above CBM. The incorporation of Zn also results in small distortions at the band edge of the VB (Figure 3c) to give a relatively large band splitting at VBM compared to little modification at the CBM. Recent reports show that a high mole

(24)

16

fraction of SrTe (>5 %) alloying with PbTe create convergence of a primary valence band along the L line and a secondary band along the Σ line and consequently enhance TE performance.²⁰ Similar band distortion occurs with the Zn substitution. Our calculations for PbTe and PbTe-1% ZnTe also show the presence of the corresponding bands that having a maximum at R point and one between Г and R points, indicated by a black arrow. The energy difference between the first and the second highest valence bands (ΔEval) is 0.11 eV, which is close to that of the lightly doped PbTe-SrTe but larger than that of the heavily-alloyed PbTe-SrTe²⁰. When Zn substitutes Pb, the maximum of the second highest valence band is shifted to Г point, and its energy level is elevated. As a result, ΔEval is reduced by 0.01 eV by 1% ZnTe doping.

This band convergence is consistent with that of the previous study²⁰ with taking account of the spin-orbit interaction. We further considered a hole doping effect by incorporation of Na

(25)

17

along with ZnTe. Figures 3(d) and (e) show that electronic band structures of PbTe-2% Na and PbTe-1% ZnTe-2% Na, respectively. Doping Na introduces hole carriers into PbTe, but significant band distortions does not occur, especially at the first and second highest valence bands. When Zn is additionally introduced, the similar change on the second highest valence band to the case of PbTe-1% ZnTe is observed. As a result, ΔEval decreases by 0.014 eV for the Pb0.990Na0.010Te-2% ZnTe. Co-doping of Na and Zn causes the larger decrease in ΔEval, possibly due to additional interaction between Zn and Na impurities in PbTe.

(26)

18

Figure 3. Electronic band structures are depicted for (a) pristine PbTe, (b) Zn-substituted PbTe, (d) Na-doped PbTe, and (e) Na-doped and Zn-substituted PbTe. (c) and (f) show the band structures in the area marked by red rectangles in enlarged scale for undoped and Na-doped cases, respectively.

(27)

19

3.3 Thermal properties of

Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4)

The temperature-dependent electrical conductivity (σ) of the Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4) decreases with increasing temperature, indicative of heavily degenerate doping (Figure 4). The σ value decreases with increasing ZnTe content mainly due to the reduction in hole mobility, as explained below: ~2700, ~1900, ~1700, and ~1600 S cm^-1 for x = 0, 1, 2, and 4 at 300K, respectively. Then, it follows a nearly quadratic temperature-dependent power law of σ ~ T^-δ (δ = 2.4 – 2.6), which is a typical behavior for p-type PbTe materials because of dominating acoustic phonon scattering at high temperature.³⁷ The values of δ are - 2.48(5), -2.50(4), -2.51(6), and -2.54(5) for x = 0, 1, 2, and 4, respectively. The temperature-dependent Seebeck coefficient (S) is positive over the entire temperature range

(28)

20

of the measurement, indicative of p-type conduction. The S value at 300 K are ~68, ~67, ~70, and ~72 μV K^-1 for x = 0, 1, 2, and 4, respectively, with nearly linear temperature dependence.

The values of the Hall coefficients (RH) are positive over the entire temperature range of the measurement, consistent with the positive sign of S and confirming p-type nature. The RH value is temperature dependent: for x = 2, it increases from ~0.054 cm³ C^-1 at 300 K to a maximum of ~0.065 cm³ C^-1 at 400 K and subsequently decreases to ~0.036 cm³ C^-1 at 673 K. This behavior is common for p-type PbTe and SnTe materials with multi-valence bands. In these compounds, the light L band shifts to lower energy whereas the heavy Σ band nearly remains with increasing temperature. As a result, the energy difference between L and Σ band (ΔEL-Σ) becomes smaller and hole carriers between the two bands are repopulated, causing the strong temperature dependence of

(29)

21

the RH.^38-40 In contrast, a typical semiconductor with only a single band contributing to the charge transport exhibits the nearly temperature-independent RH. The peak temperature of the RH (TH) corresponds to a critical temperature at which the contribution from the two valence bands of L and Σ to hole transport is comparable, thereby implying that the lower TH

correlates with the smaller ΔEL-Σ (Figure 5).^{38, 40, 41} The TH

decreases from 423 K for x = 0 to 400 K for x = 1, indicative of the reduction in ΔEL-Σ supported by theoretical calculation results mentioned above, but negligibly changes with further addition of ZnTe (x ≥ 2) probably due to the low solubility limit of ZnTe in PbTe. Note that pure p-type PbTe typically shows TH ~ 425 K regardless of hole concentration.

A similar behavior was observed in the PbTe-SrTe system, in which SrTe shows the very low solubility in PbTe.¹⁷ When SrTe is heavily alloyed with PbTe up to 8% via non- equilibrium synthesis, a decreasing trend of TH is seen

(30)

22

because of valence band convergence caused by the high solubility of SrTe with PbTe.²⁰

The temperature-dependent hole concentration (Np) was determined from the relationship Np = 1/(e∙RH), where e is the electronic charge, considering a simple parabolic band and single carrier conduction (Figure 6). The Np values for x = 0, 1, 2, and 4 at 300 K are ~1.56 × 10²⁰, ~1.15 × 10²⁰, ~1.15

× 10²⁰, and ~1.06 × 10²⁰ cm^-3, respectively, indicating that the incorporation of Zn cation in PbTe affect the Np value negligibly because of the same valence of Zn and Pb cations.

The Pisarenko plot^{20, 39} shows that correlation between S and Np of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4) falls near the theoretical Pisarenko line of p-type PbTe with the consideration of two valence-band structure (Figure 7). This observation suggests that the valence band convergence caused by the incorporation of ZnTe is insufficient to enhance the Seebeck coefficient because of the low solubility of ZnTe

(31)

23

in PbTe, as in the case of the lightly alloyed PbTe-SrTe samples.¹⁷

(32)

24

Figure 4. Temperature dependence of electrical conductivity σ (left) and Seebeck coefficient S (right) of Pb0.985Na0.015Te -x% ZnTe (x = 0, 1, 2, 4).

300 400 500 600 700 0

1000 2000 3000

 (S cm -1 )

T (K)

x = 0 x = 1 x = 2 x = 4

50 100 150 200 250

300

S (  V K -1 )

(33)

25

Figure 5. Temperature dependence of Hall coefficient RH of samples of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4).

300 400 500 600 700 2

4 6 8

R H (10 -2 cm 3 C -1 )

T (K)

x = 0 x = 1 x = 2 x = 4

(34)

26

Figure 6. Temperature dependence of hole concentration Np of

samples of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4).

300 400 500 600 700 8

12 16 20

24

N p (10 19 cm -3 )

T (K)

x = 0 x = 1 x = 2 x = 4

(35)

27

Figure 7. Room temperature Seebeck coefficient S as a function of hole concentration (Np) of samples of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4). The solid line is the theoretical Pisarenko plot for p-type PbTe^{20, 39}.

60 80 100 120 140

Pisarenko line of p-type PbTe x = 0, this work

x = 1, this work x = 2, this work x = 4, this work

10

²⁰

S (  V K -1 )

N p (cm -3 )

10

¹⁹

(36)

28

The temperature-dependent hole mobility μH was estimated from the equation μH = σ/(Np∙e) (Figure 8). The μH values for x = 0, 1, 2, and 4 at 300 K are ~94, ~81, ~68, and ~59 cm² V^-1 s^-1, respectively. They decrease with increasing ZnTe content because ZnTe precipitates in the PbTe play a role as the scatterers of hole carrier. The temperature-dependence of μH follows a power law of μH

~ T^-λ (λ = 3.3 – 3.9) in the range from 400 K to 673 K, which is similar to μH ~ T^-4 of p-type PbTe-SrTe materials.²⁰ Rapid drop of the μH at high temperature is typical for p-type PbTe-based materials and is attributed to contribution of the heavy holes in the Σ valence band.²⁰ N-type PbTe materials tend to show different dependence of μH ~ T^-2.⁴²

Figure 9 shows the temperature-dependent power factors (PF, σS²) of the samples. The PF values for x = 0, 1, 2, and 4 at 300 K are ~12.4, ~8.5, ~7.7, and ~8.7 μW

(37)

29

cm^-1 K^-2, respectively. The PF increases with temperature to reach a maximum, followed by decrease. For example, the PF value for x = 2 increases from ~7.7 μW cm^-1 K^-2 at 300 K to a maximum of ~31.6 μW cm^-1 K^-2 at ~610 K, and subsequently decreases to ~27.0 μW cm^-1 K^-2 at ~700 K.

The PF values for x = 0, 1, 2, and 4 reach to ~29.9, ~27.1,

~27.0, and ~26.4 μW cm^-1 K^-2at ~700 K, respectively, which are higher than ~21 μW cm^-1 K^-2 for PbTe-2%

CdTe-1.5% Na and ~22 μW cm^-1 K^-2 at ~700 K for PbTe-2% HgTe-1% Na.^{25, 27}

(38)

30

Figure 8. Temperature dependence of (a) hole mobility μH of samples of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4).

300 400 500 600 700 0

40 80 120

_Pb

0.985Na_0.015Te-x% ZnTe

 H (cm 2 V -1 s -1 )

T (K)

x = 0 x = 1 x = 2 x = 4

(39)

31

Figure 9. Temperature dependence of power factor PF (=

σS²) of samples of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4).

300 400 500 600 700 10

20 30 40

PF (  W cm -1 K -2 )

T (K)

x = 0 x = 1 x = 2 x = 4

(40)

32

3.4 Thermal transport property of

Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4)

The temperature dependence of the total thermal conductivity κtot and the lattice thermal conductivity κlatt of the samples are shown in Figures 10 and 11, respectively.

The κtot of the samples follows a power law of κtot ~ T^-α (α = 1.4 – 1.5). For x = 2, the κtot value is ~3.76 W m^-1 K^-1 at 300 K and decreases to ~1.10 W m^-1 K^-1 at ~700 K.

Those for x = 0, 1, 2, and 4 at 300 K are ~4.34, ~3.87,

~3.76, and ~3.98 W m^-1 K^-1, respectively. The electrical thermal conductivity κelec was estimated from κelec = LσT, where L is the Lorenz number. The L values as a function of temperature were obtained from temperature-dependent Seebeck coefficient data on the assumption of a single parabolic band (SPB) model.⁴³ The κlatt value was estimated by subtracting the κelec value from the κtot value: κlatt =

(41)

33

κtot – κelec (= LσT). The temperature dependence of the κlatt follows a power law of ~ T^-β (β = 1.6 – 1.8) and is significantly different from the typical dependence of T^-1 observed for normal semiconductors with Umklapp scattering as the major mechanism for phonon scattering.

Such deviation from T^-1 results from bipolar contribution in p-type PbTe materials.^{40, 44} The lowest κlatt of ~0.69 W m^-

1 K^-1 was observed at ~700 K for x = 2 and the corresponding values for x = 0, 1, and 4 are ~0.79, ~0.82, and ~0.84 W m^-1 K^-1, respectively, which are 25 – 50%

higher than that of Pb0.98Na0.02Te-10% SrTe (0.5 – 0.6 W m^-1 K^-1 at ~700 K) with the hierarchical architecturing.²⁰ Further understanding on thermal transport properties requires the compositional information at the micro- and nanoscale as well as the microstructural information including the size and distribution of precipitates, grain size, and coherency of interface between matrix and precipitates.

(42)

34

A detailed transmission electron microscopy (TEM) study is in progress.

(43)

35

Figure 10. Temperature dependence of total thermal conductivity κtot of samples of Pb0.985Na0.015Te-x% ZnTe (x

= 0, 1, 2, 4).

300 400 500 600 700 1

2 3 4

 tot (W m -1 K -1 )

T (K)

x = 0 x = 1 x = 2 x = 4

(44)

36

Figure 11. Temperature dependence of lattice thermal conductivity κlatt of samples of Pb0.985Na0.015Te-x% ZnTe (x

= 0, 1, 2, 4).

300 400 500 600 700 1

2 3

 la t (W m -1 K -1 )

T (K)

x = 0 x = 1 x = 2 x = 4

(45)

37

3.5 Thermoelectric figure of merit ZT of Pb_0.985Na_0.015Te-x% ZnTe (x = 0, 1, 2, 4)

The TE figure of merit ZT as a function of temperature increases with increasing temperature (Figure 12), which is typical behavior of PbTe-based materials. The ZT for x = 0, 1, 2, and 4 at ~700 K is ~1.53, ~1.51, ~1.73, and ~1.47, respectively. Pb0.985Na0.015Te-2% ZnTe exhibits the highest ZT of ~1.73 at ~700 K, mainly due to a combination of relatively higher PF and lower κlatt, compared to ZT ~1.53 at

~700 K for Na-doped PbTe with no incorporation of ZnTe. In comparison, the ZT value at ~700 K for Pb0.985Na0.015Te-2%

ZnTe is higher than ZT ~1.17 for PbTe-2% CdTe-1.5% Na²⁷ and ZT ~ 1.59 for PbTe-2% HgTe-1% Na²⁷, but is relatively lower than ZT ~ 1.9 for Pb0.98Na0.02Te-8% SrTe²⁰.

(46)

38

Figure 12. Temperature dependence of the TE figure of merit ZT of samples of Pb0.985Na0.015Te-x% ZnTe (x = 0, 1, 2, 4).

300 400 500 600 700 0.0

0.5 1.0 1.5 2.0

ZT

T (K)

x = 0 x = 1 x = 2 x = 4

(47)

39

PART 4. CONCLUSIONS

The effect of alloying ZnTe with PbTe on TE properties were studied to realize a high ZT through valence band convergence as alloying SrTe with PbTe. The Hall coefficient data of Na-doped PbTe-ZnTe materials reveal that the energy separation ΔEL-Σ between the light-hole L band and the heavy-hole Σ band become narrower even at a low solubility (< 1 %) of ZnTe in PbTe. However, there is no considerable enhancement in the Seebeck coefficient, suggesting that the higher solubility of ZnTe with PbTe is required for valence band convergence. The enhanced ZT of

~1.73 for the SPS-processed Pb0.985Na0.015Te-2% ZnTe is mainly attributed to a relatively higher power factor of ~27.0 μW cm^-1 K^-2 at ~700 K and a lower lattice thermal conductivity of ~0.69 W m^-1 K^-1 at ~700 K, compared to

(48)

40

Pb0.985Na0.015Te (ZT ~ 1.53). Developing synthetic procedure for obtaining the higher solubility of ZnTe in PbTe is necessary to enhance a ZT value of p-type PbTe-ZnTe system by valence band convergence.

(49)

41

REFRERENCES

1. L. E. Bell, Science, 2008, 321, 1457-1461.

2. W. G. Zeier, A. Zevalkink, Z. M. Gibbs, G. Hautier, M. G. Kanatzidis and G. J. Snyder, Angew. Chem., Int. Ed., 2016, 55, 6826-6841.

3. G. Tan, L. D. Zhao and M. G. Kanatzidis, Chem. Rev., 2016, 116, 12123-12149.

4. J. R. Sootsman, H. Kong, C. Uher, J. J. D'Angelo, C. I. Wu, T. P.

Hogan, T. Caillat and M. G. Kanatzidis, Angew. Chem., Int. Ed., 2008, 47, 8618-8622.

5. J. Q. He, J. R. Sootsman, S. N. Girard, J. C. Zheng, J. G. Wen, Y. M.

Zhu, M. G. Kanatzidis and V. P. Dravid, J. Am. Chem. Soc., 2010, 132, 8669-8675.

6. S. N. Girard, J. Q. He, X. Y. Zhou, D. Shoemaker, C. M. Jaworski, C.

Uher, V. P. Dravid, J. P. Heremans and M. G. Kanatzidis, J. Am.

Chem. Soc., 2011, 133, 16588-16597.

7. K. Biswas, J. Q. He, G. Y. Wang, S. H. Lo, C. Uher, V. P. Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2011, 4, 4675-4684.

8. J. Q. He, S. N. Girard, J. C. Zheng, L. D. Zhao, M. G. Kanatzidis and V. P. Dravid, Adv. Mater., 2012, 24, 4440-4444.

(50)

42

9. S. N. Girard, T. C. Chasapis, J. Q. He, X. Y. Zhou, E. Hatzikraniotis, C. Uher, K. M. Paraskevopoulos, V. P. Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2012, 5, 8716-8725.

10. M. Ohta, K. Biswas, S. H. Lo, J. Q. He, D. Y. Chung, V. P. Dravid and M. G. Kanatzidis, Adv. Energy Mater., 2012, 2, 1117-1123.

11. S. H. Lo, J. Q. He, K. Biswas, M. G. Kanatzidis and V. P. Dravid, Adv. Funct. Mater., 2012, 22, 5175-5184.

12. S. A. Yamini, T. Ikeda, A. Lalonde, Y. Z. Pei, S. X. Dou and G. J.

Snyder, J. Mater. Chem. A, 2013, 1, 8725-8730.

13. R. J. Korkosz, T. C. Chasapis, S. H. Lo, J. W. Doak, Y. J. Kim, C. I.

Wu, E. Hatzikraniotis, T. P. Hogan, D. N. Seidman, C. Wolverton, V.

P. Dravid and M. G. Kanatzidis, J. Am. Chem. Soc., 2014, 136, 3225-3237.

14. D. Wu, L. D. Zhao, X. Tong, W. Li, L. J. Wu, Q. Tan, Y. L. Pei, L.

Huang, J. F. Li, Y. M. Zhu, M. G. Kanatzidis and J. Q. He, Energy Environ. Sci., 2015, 8, 2056-2068.

15. K. F. Hsu, S. Loo, F. Guo, W. Chen, J. S. Dyck, C. Uher, T. Hogan, E. K. Polychroniadis and M. G. Kanatzidis, Science, 2004, 303, 818-821.

16. K. Biswas, J. Q. He, Q. C. Zhang, G. Y. Wang, C. Uher, V. P. Dravid and M. G. Kanatzidis, Nat. Chem., 2011, 3, 160-166.

(51)

43

17. K. Biswas, J. He, I. D. Blum, C.-I. Wu, T. P. Hogan, D. N. Seidman, V. P. Dravid and M. G. Kanatzidis, Nature, 2012, 489, 414-418.

18. J. P. Heremans, V. Jovovic, E. S. Toberer, A. Saramat, K. Kurosaki, A. Charoenphakdee, S. Yamanaka and G. J. Snyder, Science, 2008, 321, 554-557.

19. Y. Z. Pei, X. Y. Shi, A. LaLonde, H. Wang, L. D. Chen and G. J.

Snyder, Nature, 2011, 473, 66-69.

20. G. Tan, F. Shi, S. Hao, L. D. Zhao, H. Chi, X. Zhang, C. Uher, C.

Wolverton, V. P. Dravid and M. G. Kanatzidis, Nat. Commun, 2016, 7, 12167.

21. B. J. Sealy and A. J. Crocker, J. Mater Sci., 1973, 8, 1731-1736.

22. B. J. Sealy and A. J. Crocker, J Mater. Sci., 1973, 8, 1737-1743.

23. A. J. Crocker, J. Mater. Sci., 1968, 3, 534-539.

24. M. Lee and C. U. Kim, Physica B, 2001, 304, 267-275.

25. K. Ahn, M. K. Han, J. Q. He, J. Androulakis, S. Ballikaya, C. Uher, V.

P. Dravid and M. G. Kanatzidis, J. Am. Chem. Soc., 2010, 132, 5227-5235.

26. Y. Z. Pei, A. D. LaLonde, N. A. Heinz and G. J. Snyder, Adv. Energy Mater., 2012, 2, 670-675.

27. K. Ahn, K. Biswas, J. He, I. Chung, V. Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2013, 6, 1529-1537.

(52)

44

28. P. K. Rawat, B. Paul and P. Banerji, ACS Appl. Mater. Inter., 2014, 6, 3995-4004.

29. P. E. Blochl, Phys. Rev. B, 1994, 50, 17953-17979.

30. G. Kresse and J. Hafner, J. Phys.-Condens. Mat., 1994, 6, 8245- 8257.

31. G. Kresse and J. Furthmuller, Phys. Rev. B, 1996, 54, 11169- 11186.

32. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865-3868.

33. R. D. Shannon, Acta Crystallogr. A, 1976, 32, 751-767.

34. A. Gueguen, P. F. P. Poudeu, C. P. Li, S. Moses, C. Uher, J. Q. He, V. Dravid, K. A. Paraskevopoulos and M. G. Kanatzidis, Chem.

Mater., 2009, 21, 1683-1694.

35. M. Orihashi, Y. Noda, L. D. Chen and T. Hirai, Mater. T. Jim., 2000, 41, 1196-1201.

36. K. Ahn, H. Kong, C. Uher and M. G. Kanatzidis, J. Solid State Chem., 2016, 242, Part 2, 34-42.

37. R. S. Allgaier and W. W. Scanlon, Phys. Rev., 1958, 111, 1029- 1037.

38. Y. Pei, A. D. LaLonde, N. A. Heinz, X. Shi, S. Iwanaga, H. Wang, L.

Chen and G. J. Snyder, Adv. Mater., 2011, 23, 5674-5678.

(53)

45

39. L. D. Zhao, H. J. Wu, S. Q. Hao, C. I. Wu, X. Y. Zhou, K. Biswas, J. Q.

He, T. P. Hogan, C. Uher, C. Wolverton, V. P. Dravid and M. G.

Kanatzidis, Energy Environ. Sci., 2013, 6, 3346-3355.

40. G. J. Tan, F. Y. Shi, S. Q. Hao, H. Chi, T. P. Bailey, L. D. Zhao, C.

Uher, C. Wolverton, V. P. Dravid and M. G. Kanatzidis, J. Am. Chem.

Soc., 2015, 137, 11507-11516.

41. H. Wang, Z. M. Gibbs, Y. Takagiwa and G. J. Snyder, Energy Environ. Sci, 2014, 7, 804-811.

42. A. F. Joffe and L. S. Stilbans, Rep. Prog. Phys., 1959, 22, 167-203.

43. A. F. May, E. S. Toberer, A. Saramat and G. J. Snyder, Phys. Rev.

B, 2009, 80, 125205.

44. G. J. Tan, F. Y. Shi, J. W. Doak, H. Sun, L. D. Zhao, P. L. Wang, C.

Uher, C. Wolverton, V. P. Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2015, 8, 267-277.

(54)

46

APPENDIX

#1. Bi2Te3 ball-milling

#1.1 Introduction

In nano size, electrical characteristics that have not been seen before are revealed. Through mechanical grinding of Bi2Te3, change of thermoelectric material properties can be arisen. However, because of the plate-like structure of Bi2Te3, there is a problem that the jar is coated when ball milling is performed (Figure 13). Therefore, milled Bi2Te3

could not be collected unless it was broken. Furthermore, the size of the powder does not change. To solve this problem, process control agent is introduced.

#1.2 Experimental section

Stoichiometric amounts of elemental Bi, Te were weighed for the synthesis of Bi2Te3 and loaded into a jar. Bi2Te3

(55)

47

Figure 13. Coating effect of ball milling.

(56)

48

powder was milled at 250 rpm for 6 h in Ar atmosphere using a planetary ball mill (PM100, Retch) and 10mm zirconia balls.

The ball to powder ratio was kept at 20:1. Phase identification of the samples were analyzed by X-ray diffraction (Dmax 2500, Rigaku with monochromatic Cu Kα radiation). The ball-milled Bi2Te3 was used as a starting material. After milling using 3mm ball up to 24 hours in 3 hour intervals, ethanol (99.5% Samchun) or ethylene glycol (99.8%

anhydrous, Sigma aldrich) was added and milled for 12 hours.

Powder was sintered with a graphite mold of 13 mm in diameter by SPS (SPS-211LX, FUJI) under 50 MPa at 673 K for 10 min. Particle size of the samples were analyzed by Field-Emission Scanning Electron Microscope(SUPRA 55VP, Carl Zeiss).

#1.3 Results and discussions

Bi2Te3 synthesis was successfully achieved by ball milling.

However there is no XRD peak broadening because of coating

(57)

49

Figure 14. Powder X-ray diffraction patterns of samples of Bi2Te3

(58)

50

effect (Figure 14). Using ethanol as a process control agent, the density of Bi2Te3 after spark plasma sintering was less than 80% of the theoretical, but the coating effect disappeared.

Not only that, but also the coating effect was eliminated when using ethylene glycol (Figure 15). Figure 16 shows the milled Bi2Te3 image before and after addition ethylene glycol. The dry milling did not decrease in particle size over time, but it was reduced in ball milling after PCA addition.

#1.4 Conclusions

Although the thermoelectric properties have not been confirmed, this experiment confirmed that ethylene glycol is effective to eliminate coating effect and works as a process control agent to reduce particle size.

(59)

51

Figure 15. the coating effect was eliminated when using ethylene glycol

(60)

52

Figure 16. SEM images of surface of the Bi2Te3 samples obtained from ball-milled powders for (a)6hr (b)12hr (c)18 hr (d)24hr milling time. (e)milling for 12hr after additing ethylene glocol

a. b.

c. d.

e.

(61)

53

#2. Iridate Pyrochlore synthesis

#2.1 Introduction

Magnetic compounds with a pyrochlore structure show a variety of properties. The Iridium–pyrochlore is the family of compounds R2Ir2O7 (R is rare earth). Also, a more generic term for the pyrochlore crystal structure is Fd3m. It provides a fascinating opportunity to investigate metal-insulator transition, superconductivity and geometrical frustration.

Before measuring these properties, the synthesis was proceeded.

#2.2 Experimental section

Polycrystalline samples of R2Ir2O7 (R = Pr, Sm) were prepared by a solid-state reaction. Mixtures of treated Pr6O11(99,996%, Alfa aesar), Sm2O3 (99.9%, Sigma Aldrich) and IrO2(99.9%, Sigma aldrich) in stoichiometric ratios were pressed into pellets, which were then inserted into a quartz

(62)

54

tube and sealed under a vacuum of ~1.33 x 10^-3 Pa. Tube was heated at 1273K-1373K for about 12hr with several times. If the reaction was no longer in progress, add rare earth oxide and react. Only in case Sm2Ir2O7, it reacted with air in powder form.

#2.3 Results and discussions

Powder X-ray diffraction data demonstrates that both Sm2O3 and IrO2 decreased and pure Sm2Ir2O7 was synthesized (Figure 17). However, there were unknown peaks at 25°and 31°in spite of various attempts (Figure 18).

#2.4 Conclusions.

High quality pyrochlore samples were made through this experiment. To verify the properties described in the introduction part, sputtering should be done with a sample of 80% or more density. For future works, hot isostatic pressure system will be used.

(63)

55

Figure 17. Powder X-ray diffraction patterns of samples of Sm2Ir2O7

(64)

56

Figure 18. Powder X-ray diffraction patterns of samples of Pr2Ir2O7