SYSTEMS AND METHODS OF FABRICATION AND USE OF NbFeSb P-TYPE HALF-HEUSLER THERMOELECTRIC MATERIALS

Abstract

Discussed herein are half-Heusler thermoelectric materials including niobium, iron, antimony, and titanium that are formed by ball-milling and hot-pressing the ball-milled power to obtain various thermoelectric properties and an average grain size above 1 μm.

Description

BACKGROUND

Half-Heuslers (HHs) are intermetallic compounds which have great potential as high temperature thermoelectric materials for power generation. However, the dimensionless thermoelectric figure-of-merit (ZT) of HHs is lower than that of the most state-of-the-art thermoelectric materials. HHs are complex compounds, including MCoSb (p-type) and MNiSn (n-type), where M can be Ti or Zr or Hf or combination of two or three of the elements. They form in cubic crystal structure with a F4/3m (No. 216) space group. These phases are semiconductors with 18 valence electron count (VEC) per unit cell and a narrow energy gap. The Fermi level is slightly above the top of the valence band. The HH phases have a useful Seebeck coefficient with moderate electrical conductivity. The performance of thermoelectric materials depends on ZT, defined by ZT=(S²σ/κ)T, where σ is the electrical conductivity, S the Seebeck coefficient, κ the thermal conductivity, and T the absolute temperature. Half-Heusler compounds may be good thermoelectric materials due to their high power factor (S²σ). It has been reported that the MNiSn phases are promising n-type thermoelectric materials with exceptionally large power factors and MCoSb phases are promising p-type materials. In recent years, different approaches have been reported that have improved the ZT of half-Heusler compounds.

BRIEF SUMMARY OF THE DISCLOSURE

Embodiments described herein comprise a combination of features and characteristics intended to address various shortcomings associated with certain prior devices, compositions, systems, and methods. The various features and characteristics described above, as well as others, will be readily apparent to those of ordinary skill in the art upon reading the following detailed description, and by referring to the accompanying drawings.

In an embodiment, a method of fabricating a thermoelectric component, comprising: forming, an ingot of a first alloy comprising niobium, iron, antimony, and titanium according to a formula Nb_1-xTi_xFeSb; ball-milling the ingot to form a ball-milled powder; hot-pressing the ball-milled powder between 1100 K and 1400 K to form a thermoelectric component comprising a power factor above about 60 μW cm⁻¹ K⁻² from about 300K to about 775K.

In an embodiment, a thermoelectric device comprising: a hot-pressed component of ball-milled powder according to a formula Nb_1-xTi_xFeSb, wherein the component comprises a power factor above about 60 μW cm⁻¹ K⁻² from about 300 K to about 775 K and an average grain size above 1 μm.

In an embodiment, a thermoelectric device comprising: a hot-pressed component of ball-milled powder according to a formula Nb_1-xTi_xFeSb, wherein a power factor above about 60 μW cm⁻¹ K⁻² from about 300K to about 775K, and wherein x is from 0.02 to 0.7.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1A-1H show the thermoelectric properties of Nb_0.95Ti_0.05FeSb fabricated according to certain embodiments of the present disclosure.

FIGS. 2A-2D are scanning electron microscopy (SEM) images that show significant enlargement of the grain sizes of thermoelectric materials at increasing hot-pressing temperatures for samples fabricated according to certain embodiments of the present disclosure.

FIGS. 3A and 3B are graphs illustrating the effect of grain size on (FIG. 3A) lattice thermal conductivity and (FIG. 3B) hole mobility for samples fabricated according to certain embodiments of the present disclosure.

FIGS. 4A-F illustrate the thermoelectric property dependence of temperature for Nb_1-xTi_xFeSb with x=0, 0.04, 0.05, 0.06, 0.07, 0.1, 0.2, and 0.3 for samples fabricated according to certain embodiments of the present disclosure.

FIGS. 5A-5D show a TE property comparison for Nb_0.95Ti_0.05FeSb, fabricated according to certain embodiments of the present disclosure, as compared to reference sample data.

FIG. 6 illustrates the contribution of different phonon scattering mechanisms to the lattice thermal conductivity of Nb_1-xTi_xFeSb at room temperature for samples fabricated according to certain embodiments of the present disclosure.

FIGS. 7A-7B show the calculated and measured (FIG. 7A) output power density and (FIG. 7B) conversion efficiency of samples Nb_1-xTi_xFeSb with x=0.05 and 0.2 for samples fabricated according to certain embodiments of the present disclosure.

FIG. 8 is a Pisarenko plot of thermoelectric materials fabricated according to certain embodiments of the present disclosure.

FIG. 9 is an image of an etched surface of Nb_0.95Ti_0.05FeSb fabricated according to certain embodiments of the present disclosure.

FIGS. 10A-10D are statistic grain sizes analyses of Nb_0.95Ti_0.05FeSb pressed at 1123 K (FIG. 10A), 1173 K (FIG. 10B), 1273 K (FIG. 10C) and 1373 K (FIG. 10D) for samples fabricated according to certain embodiments of the present disclosure.

FIG. 11A is a cumulative grain size distribution of Nb_0.95Ti_0.05FeSb pressed at 1123 K, 1173 K, 1273 K and 1373 K according to certain embodiments of the present disclosure.

FIG. 11B is a Williamson-Hall plot of grain size vs. strain for thermoelectric materials fabricated according to certain embodiments of the present disclosure

FIG. 11C is a graph of the Hall carrier concentration (n_H) variation with the amount of dopants in Nb_1-xTi_xFeSb fabricated according to certain embodiments of the present disclosure.

FIG. 11D is a differential scanning calorimetry (DCS) graph of Nb_0.95Ti_0.05FeSb manufactured according to certain embodiments of the present disclosure.

FIGS. 12A-12C show thermoelectric properties for Nb_0.95Ti_0.05FeSb hot-pressed at 1373 K according to certain embodiments of the present disclosure.

FIG. 13 is an x-ray diffraction (XRD) pattern of Nb_1-xTi_xFeSb fabricated according to certain embodiments of the present disclosure.

FIGS. 14A and 14B illustrate the effect of grain size on lattice thermal conductivity and hole mobility for Nb_1-xTi_xFeSb fabricated according to certain embodiments of the present disclosure.

FIGS. 15A and 15B illustrate the power factor and calculated power output density for Nb_1-xTi_xFeSb fabricated according to certain embodiments of the present disclosure.

FIG. 16 is a graph of thermoelectric current (I_TE)-voltage (V_TE) curve and power output at hot side temperature 773 K for thermoelectric materials fabricated according to embodiments of the present disclosure.

FIGS. 17A-17E are graphs of thermoelectric properties of n-type NbFe_1-yCo_ySb fabricated according to certain embodiments of the present disclosure.

FIGS. 18A and 18B illustrate the temperature dependency of thermoelectric materials fabricated according to certain embodiments of the present disclosure.

DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS

The following discussion is directed to various exemplary embodiments. However, one of ordinary skill in the art will understand that the examples disclosed herein have broad application, and that the discussion of any embodiment is meant only to be exemplary of that embodiment, and not intended to suggest that the scope of the disclosure, including the claims, is limited to that embodiment.

The drawing figures are not necessarily to scale. Certain features and components herein may be shown exaggerated in scale or in somewhat schematic form and some details of conventional elements may not be shown in interest of clarity and conciseness.

In the following discussion and in the claims, the terms “including” and “comprising” are used in an open-ended fashion, and thus should be interpreted to mean “including, but not limited to . . . .”

“Achieving High Power Factor and Output Power Density in p-type half-Heuslers Nb1-xTixFeSb,” Proceedings of the National Academy of Sciences of the United States of America, vol. 133, no. 48 (October 2016), is incorporated by reference in its entirety herein.

It is estimated by some that about 20 to 50% of industrial energy input is lost as waste heat. By converting some of the waste heat into useful electrical power, the consumption of fossil fuels could be significantly reduced, which slows down the increase of the atmospheric carbon dioxide that is a major concern for the environment. Thermoelectric (TE) materials are considered unique in converting heat into electricity. The conversion efficiency by TE materials is usually characterized by,

$\begin{array}{cc}\eta =\frac{T_H-T_C}{T_H}·\frac{\sqrt{1+\stackrel{\_}{\mathrm{ZT}}}-1}{\sqrt{1+\stackrel{\_}{\mathrm{ZT}}}+\frac{T_C}{T_H}}& \left(1\right)\end{array}$

where ZT is the average thermoelectric figure of merit (ZT) between the hot side temperature (T_H) and cold side temperature (T_C) of a TE material, and is defined as:

$\begin{array}{cc}\mathrm{ZT}=\frac{\mathrm{PF}}{{\kappa }_{\mathrm{tot}}}T& \left(2\right)\\ \mathrm{PF}=S^2\sigma & \left(3\right)\\ {\kappa }_{\mathrm{tot}}={\kappa }_L+{\kappa }_e+{\kappa }_{\mathrm{bip}}& \left(4\right)\end{array}$

where PF, T, κ_tot, S, σ, κ_L, κ_e, and κ_iv, are the power factor, absolute temperature, total thermal conductivity, Seebeck coefficient, electrical conductivity, lattice thermal conductivity, electronic thermal conductivity, and bipolar thermal conductivity, respectively. Higher ZT corresponds to higher conversion efficiency.

However, this conversion approach sometimes harms the electron transport due to too much scattering of the carriers, which reduces electrical conductivity and the power factor. The output power density w may be determined by power factor.

$\begin{array}{cc}\omega =\frac{{\left(T_H-T_C\right)}^2}{4L}\stackrel{\_}{\mathrm{PF}}& \left(5\right)\end{array}$

where L is the leg length of TE material. Even though leg length can be shortened, it is not without limit due to stress, parasitic loss from contacts, mechanical reliability, etc. Therefore, higher power factor is desirable for higher output power density.

Half-Heusler (HH) compound is crystallized in the space group F43m. Among the various compounds, ZrNiSn based n-type and ZrCoSb based p-type materials are widely studied due to their presently acceptable ZT ˜1 at 873 to 1073 K^26-28, low cost, high power density, excellent mechanical properties, and non-toxicity.

HH possesses high power factors, which is desirable for power generation. As discussed herein, an Nb_1-xTi_xFeSb system was evaluated with x=0.04, 0.05, 0.06, 0.07, 0.1, 0.2, and 0.3 using arc melting, ball milling and hot-pressing (HP) at 1123, 1173, 1273, and 1373 K. It turns out higher HP temperature enhances the carrier mobility, leading to a record PF of ˜106 μW cm⁻¹ K⁻² at 300 K in Nb_0.95Ti_0.05FeSb. The high PF in the resulting compounds may stem from the increased carrier mobility. With such a high PF, a record output power density of about 22 W cm⁻² with a leg length of about 2 mm is experimentally achieved with T_C=293 K and T_H=868 K. On the other hand, the thermal conductivity is almost unchanged regardless of the grain sizes. Thus, by using higher HP temperature of 1373 K and changing the Ti concentration, higher mobility, PF, and ZT than the reported results for the same compositions were achieved.

Example Experimental Procedure

Sample Preparation:

In one example, 15 grams of raw elements (Nb pieces, 99.9% and Sb broken rods, 99.9%, Atlantic Metals & Alloy; Fe granules, 99.98% and Ti foams, 99.9%, Alfa Aesar) were weighed under an argon environment inside a glove box according to stoichiometry. The elements were arc melted to first form an ingot. To enhance the ingot's uniformity, the ingots were melted 3-4 times and flipped over each time. To get nano powders, the ingot were ball milled (SPEX 8000M Mixer/Mill) for 3 hours under Ar protection by two stainless steel balls with diameter of 12.7 mm. In alternate embodiments, the ball-milling may occur from 10 minutes to 70 hours, and may occur in one or more cycles. The powders were then consolidated into disks by hot pressing at 80 MPa and 2 minutes at 1123, 1173, 1273, and 1373 K with a temperature rising rate of ˜100 K/min.

Sample Characterization.

A PANalytical multipurpose diffractometer with an X'celerator detector (PANalytical X'Pert Pro) was used to characterize the phases. Morphology and elemental ratios of the samples were characterized by scanning electron microscopy (SEM, LEO 1525) and Electron Probe Micro-Analysis (EPMA, JXA-8600), respectively. Transmission electron microscopy (TEM, JEOL 2100F) was used to observe the detailed microstructures.

Property Measurement.

The thermal conductivity was calculated as a product of the thermal diffusivities, specific heat, and density, which were measured by a laserflash machine (LFA457, Netzsch), a differential scanning calorimeter (DSC 404 C, Netzsch), and an Archimedes' kit, respectively. Samples in bar shape with sizes around 2 mm×2 mm×10 mm were used for measuring the electrical conductivity and the Seebeck coefficient in a ZEM-3 (ULVAC) machine. Hall concentrations (n_H) were measured using a physical properties measurement system (PPMS, Quantum Design) under ±3 Tesla magnetic induction. The Hall mobility μ_H was calculated using μ_H=σ/(n_Hq). The uncertainty for electrical conductivity is 3%, Seebeck coefficient 5%, thermal conductivity 7% (comprising uncertainties of 4% for thermal diffusivity, 5% for specific heat, and 3% for density), so the combined uncertainty for power factor and ZT are 10% and 12%, respectively. To increase the readability, no error bars are present on the curves.

Results and Discussion

Enhanced Power Factor with Higher Hot Pressing Temperature

FIGS. 1A-1H show the thermoelectric properties of Nb_0.95Ti_0.05FeSb with hot pressing temperatures at 1123, 1173, 1273, and 1373 K. In particular, FIG. 1A shows the power factor, FIG. 1B shows the electrical conductivity (a), FIG. 1C shows the Seebeck coefficient (S), FIG. 1D shows the total thermal conductivity (κ_tot), FIG. 1E shows the lattice and bipolar thermal conductivity (κ_L+κ_bip), FIG. 1F shows the bipolar thermal conductivity (κ_bip), FIG. 1G shows the figure of merit (ZT) from 200 to 1000K, and FIG. 1H shows ZT with T from 300 to 573 K. As discussed herein, the power factor may be expressed as “PF” or as “S²σ.” The dashed lines in FIGS. 1B and 1E represents the T^−3/2 and T⁻¹ relation, respectively. The dashed line in FIG. 1C shows the calculated Seebeck coefficient using the SPB model.

FIG. 1A shows a power factor (PF) above at below 1000 K with an increasing hot pressing temperature. The peak value reaches ˜106 μW cm⁻¹ K⁻² at room temperature, for the sample hot-pressed at about 1373 K.

FIG. 1B shows that the high power factor may be at least in part due to the improved electrical conductivity. As shown in FIG. 1B, the electrical conductivity (o) is above 1.5 10⁵ Ohm⁻¹m⁻¹ at hot-pressing temperatures from 1123 K to 1373 K. Above 673 K, the electrical conductivity converges and follows the T^−−3/2 law, showing the acoustic phonons dominate carrier scattering.

The Seebeck coefficient in FIG. 1C is illustrated via a dashed line which is the calculated Seebeck coefficient. FIG. 1C uses the single parabolic band (SPB) model (taking the scattering parameter r=−0.5 since acoustic phonon scattering dominates),

$\begin{array}{cc}S=\left(\frac{k_B}{e}\right)\left[\frac{2F_1\left(\eta \right)}{F_0\left(\eta \right)}-\eta \right]& \left(6\right)\\ F_n\left(\eta \right)={\int }_0^{\infty }\frac{{\chi }^n}{1+e^{\chi -\eta }}d\chi & \left(7\right)\end{array}$

where η is related with n_H through

$\begin{array}{cc}{n}_H=4{\pi \left(\frac{2m_h^{*}k_BT}{h^2}\right)}^{3/2}\frac{F_{1/2}\left(\eta \right)}{r_H}& \left(8\right)\\ r_H=\frac{3}{2}\frac{F_{-1/2}\left(\eta \right)F_{1/2}\left(\eta \right)}{2F_0^2\left(\eta \right)}& \left(9\right)\end{array}$

where η, n_H, m*_h, and r_H are the reduced Fermi energy, the Hall carrier concentration, density of states (DOS) effective mass of holes, and the Hall factor, respectively. n_H is obtained through Hall measurement (Table 1) and m*_h is obtained by fitting the Pisarenko relation in FIG. 8. F_n(η) is the Fermi integral of order n. The good agreement between the calculated and experimental values shows the adequacy of SPB model in describing the hole transport. In FIG. 8, the Pisarenko relationship is shown with DOS effective mass m*_h=7.5 m₀ for holes, showing the validation of SPB model.

Meanwhile, the thermal conductivity shown in FIG. 1D is a little bit lower for samples hot pressed at the lower hot pressing temperatures, which, however, is mainly due to the difference in the electronic thermal conductivity as a result of the electrical conductivity

$\begin{array}{cc}{\kappa }_e=L\sigma T& \left(10\right)\\ L={\left(\frac{k_B}{e}\right)}^2\left[\frac{3F_2\left(\eta \right)}{F_0\left(\eta \right)}-{\left(\frac{2F_1\left(\eta \right)}{F_0\left(\eta \right)}\right)}^2\right]& \left(11\right)\end{array}$

where L is the Lorenz number. κ_L+κ_bip are plotted in FIG. 1e by subtracting the κ_e from κ_tot. It shows the lattice thermal conductivity is barely affected by the hot pressing temperature. In addition, at above 773 K there occurs slightly deviation from the T⁻¹ curve, showing some minor bipolar effect even though the Seebeck coefficient seems to not show such an effect. The κ_bip are calculated in FIG. 1F using a three-band model. The highest value reaches ˜0.4 W m⁻¹ K⁻¹ at 973 K, a small value comparing to the κ_L.

As a result, the enhanced power factor and almost fixed thermal conductivity led to higher ZT values, especially at temperatures below 573 K (FIGS. 1G and 1H) where the power factor shows larger differences (FIG. 1A).

FIGS. 2A-2D are scanning electron microscopy (SEM) images that show significant enlargement of the grain sizes with increasing hot-pressing temperatures. Statistically (see FIGS. 9, 10A-10D, and 11) the average grains sizes are found to be ˜0.3, ˜0.5, ˜3.0, and ˜4.5 μm for samples pressed at 1123 K (FIG. 2A), 1173 K (FIG. 2B), 1273 K (FIG. 2C), and 1373 K (FIG. 2D), respectively. In various embodiments, the average grain size may be larger than 1.0 μm, for example, up to 10 μm. As discussed herein, the average grain size may refer to an average of the maximum dimensions of each grain.

Table 1 (below) shows the room temperature Hall carrier concentration (n_H), Hall mobility (ρ_H), Hall factor (r_H), deformation potential (E_def), and relative density of Nb_1-xTi_xFeSb for different hot pressing temperatures. The room temperature (RT) Hall measurement shows a ˜73% enhancement of Hall mobility (μ_H) of samples pressed at 1373 K over 1123 K. Since the lattice thermal conductivity (FIG. 1E) changes little with the grain sizes, it is very interesting to investigate why the enlarged grain size performs so differently to the electron transport.

	TABLE 1
					relative
	nH	μH		Edef	density
	(1020 cm−3)	(cm2 V−1 s−1)	rH	(eV)	(%)
x = 0.04, 1373 K	6.3	25.2	1.12	12.5	99.1
x = 0.05, 1123 K	7.2	15.2	1.12	—*	99.4
x = 0.05, 1173 K	7.7	20.2	1.12	—*	98.9
x = 0.05, 1273 K	7.7	24.3	1.12	11.8	98.6
x = 0.05, 1373 K	8.1	26.3	1.12	11.5	99.0
x = 0.06, 1373 K	9.3	27.3	1.11	11.6	99.3
x = 0.07, 1373 K	10.7	26.7	1.10	11.6	98.9
x = 0.10, 1373 K	15.2	24.0	1.08	12.7	99.3
x = 0.20, 1373 K	25.7	15.2	1.06	13.7	99.1
x = 0.30, 1373 K	30.3	9.3	1.06	17.6	98.9
*Data are not given since the mobility are strongly affected by grain boundary scattering

Effect of Grain Size on Lattice Thermal Conductivity and Carrier Mobility

The lattice thermal conductivity (κ_L) of samples pressed at 1373 K is obtained by subtracting κ_e and κ_bip (FIG. 1F) from κ_tot. For analyzing, Debye-Callaway model was used and the phonon scattering was split into 4 different sources: Three phonon process (3P), grain boundary (GB), point defects (PD), and electron-phonon interaction (EP). The complete results combining 3P, GB, PD, and EP are discussed below.

Table 2 lists parameters in Debye-Callaway model for Nb_1-xTi_xFeSb.

	PD	3P	EP	v
x	Γ(10−3)	ε	B (1011 s−1)	γ1	C × nH−4/3 (m4 s−1)	(m s−1)
0	0	80	8.06	0.69	3.86 × 10−25	3879
0.04	4.5		8.16			3871
0.05	5.6		8.19			3869
0.06	6.6		8.22			3867
0.07	7.6		8.24			3865
0.1	10.6		8.32			3859
0.2	19.0		8.59			3838
0.3	25.4		8.90			3818

FIGS. 3A and 3B are graphs illustrating the effect of grain size on (FIG. 3A) lattice thermal conductivity and (FIG. 3B) hole mobility (n_H). The symbols are measurement data and lines are calculated values. The effect of grain boundary scattering is shown in FIG. 6 and discussed below. The calculated lattice thermal conductivity decreases only ˜9% with the grain size decreases by more than one order of magnitude (e.g., from 4.5 μm to 0.3 μm). Experimentally, κ_L decreases consistently (˜2%). The insensitivity of the phonon transport to the grain size may indicate that the phonon mean free path is much less than 0.3 μm. In the example in FIG. 3A, the

For analyzing the mobility, firstly it was assumed a constant n_H from room temperature through 1000 K, which is widely observed in half-Heulser systems, thus μ_H varies in the same trend as the conductivity (σ) through

σ=qn_Hμ_H  (12)

At high temperatures, the dominant electron scattering is from the acoustic phonon (AP),

$\begin{array}{cc}{\mu }_{\mathrm{AP}}=\frac{F_{-1/2}\left(\eta \right)}{2F_0\left(\eta \right)}{\mu }_0\mathrm{where},& \left(13\right)\\ {\mu }_0=\frac{e{\mathrm{\pi \hslash }}^4}{\sqrt{2}{\left(k_BT\right)}^{3/2}}\frac{v_l^2d}{{E_{\mathrm{def}}^2\left(m_b^{*}\right)}^{5/2}}& \left(14\right)\\ m_b^{*}=m_h^{*}{\left(N_v\right)}^{-2/3}& \left(15\right)\end{array}$

v_l is the longitudinal phonon velocity, calculated according to the elastic constants⁴⁹, d is the density, m*_h is the density of states (DOS) effective mass for holes and N_v is the valley degeneracy. E_def is the deformation potential that is obtained by extrapolating the high temperature mobility back to room temperature using the T^−3/2 law. E_def is found ˜12 eV for Nb_0.95Ti_0.05FeSb, a relatively small value comparing with previously reported TE systems showing the weaker electron-phonon interaction in HH systems⁴⁴ and thus beneficial for the electron transport and the high power factor.

For the mobility at below 573 K, the grain boundary scattering was evaluated,

$\begin{array}{cc}{\mu }_{\mathrm{GB}}={\mathrm{De}\left(\frac{1}{2\pi m_b^{*}k_BT}\right)}^{1/2}\exp \left(-\frac{E_B}{k_BT}\right)& \left(16\right)\end{array}$

where D is the grain size and E_B is the barrier energy of grain boundary, which is set as a common fitting parameter for Nb_0.95Ti_0.05FeSb. Thus μ_H could be obtained through the Matthiessen's rule,

μ_H⁻¹=μ_AP⁻¹+μ_GB⁻¹  (17)

With E_B˜0.1 eV, the fitting result (FIG. 3B), verified the effects of grain boundary scattering. In addition, it shows the grain boundary scattering become stronger when size is smaller: from 0.5 to 0.3 μm, the mobility decreases by ˜30%, but from 4.5 to 3.0 μm, the mobility drops only 5%. When taken together, FIGS. 3A and 3B show the grain boundary may be more effective in electron scattering than phonon scattering.

It is noted that 1) for samples pressed at 1123 K through 1373 K, the lattice strain decreases much less (from 0.048% to 0.04%) (see FIG. 11B below)), 2) the change in carrier mobility is less than 80% (FIG. 3B) and could be attributed to the grain size variation, and 3) the n_H changes only ˜6% (Table 1), much less than the at least 100% caused by disorders.

TE Properties of p-Type Nb_1xTi_xFeSb

FIGS. 4A-F illustrate the thermoelectric property dependence of temperature for Nb_1-xTi_xFeSb with x=0, 0.04, 0.05, 0.06, 0.07, 0.1, 0.2, and 0.3 that were hot-pressed from about 1000° C. to about 1100° C. FIG. 4A shows the electrical conductivity, FIG. 4B shows the Seebeck coefficient, FIG. 4C shows the power factor, FIG. 4D shows the total thermal conductivity, FIG. 4E shows the lattice and bipolar thermal conductivity, and FIG. 4F shows the ZT. In FIG. 4A, the dashed line and the inset show the ˜T^−3/2 relation and the conductivity of undoped NbFeSb, respectively. While discreet values of x such as x=0, 0.04, 0.05, 0.06, 0.07, 0.1, 0.2, and 0.3 are discussed herein, x may be up to 0.7 depending upon the embodiment.

The electrical conductivity (σ) (FIG. 4A) of undoped NbFeSb increases with temperature, showing typical semiconductor behaviors. The band gap could be estimated from the high temperature electrical conductivity using the following relation,

$\begin{array}{cc}\ln \left(\sigma \right)\propto E_g\frac{-1}{2k_BT}& \left(18\right)\end{array}$

The estimated value of the electrical conductivity is ˜0.51 eV, very close to the DFT calculation (0.53 eV).

For the highly doped samples, σ obeys the T^−3/2 law, showing the dominating acoustic phonon scattering. Meanwhile, σ increases with x up to 0.2 because of the increased n_H; further increase x to 0.3, σ decreased due to the decreased μ_H (Table 1). The Seebeck coefficient is lower with higher n_H (FIG. 4B), showing a good agreement with the Pisarenko relation with a density of states (DOS) effective mass m*_h=7.5 m₀ (FIG. 8).

$\begin{array}{cc}S=\frac{8\pi k_B^2}{3{\mathrm{eh}}^2}m_h^{*}{T\left(\frac{\pi }{3n}\right)}^{\frac{2}{3}}& \left(19\right)\end{array}$

The p-type Nb_1-xTi_xFeSb possesses both very high conductivity and Seebeck coefficient, leading to much higher power factor at a wide temperature range (˜100 μW cm⁻¹ K⁻² at RT and ˜50 μW cm⁻¹ K⁻² at 973 K in Nb_0.95Ti_0.05FeSb, Nb_0.94Ti_0.06FeSb, and Nb_0.93Ti_0.07FeSb (FIG. 4C). These values are higher than previously reported values where 1123 K was used for sintering.

FIGS. 5A-5D show a TE property comparison, FIG. 5A shows the power factor;

FIG. 5B shows the ZT of the Nb_1-xTi_xFeSb systems from this work and previously reported data (reference samples); FIG. 5C illustrates the power factor of the Nb_1-xTi_xFeSb systems from this work and previously reported data (reference samples); and FIG. 5D illustrates the ZT among Nb_0.95Ti_0.05FeSb fabricated according to embodiments of the present disclosure as compared to previously reported values of constantan and YbAl₃ with peak power factors exceeding 100 μW cm⁻¹ K⁻² as shown in FIG. 5C. As shown in FIG. 5A, higher pressing temperature is associated with a higher PF.

Turning back to FIG. 4D, the total thermal conductivities (κ_tot) are shown. For an undoped sample, the value is ˜17.3 W m⁻¹ K⁻¹ at RT and 7.2 W m⁻¹ K⁻¹ at 973 K. Ti doping effectively suppresses the thermal conductivity, especially the lattice part (FIG. 4E). As a result, the peak ZT reaches 1.1 for Nb_0.8Ti_0.2FeSb (FIG. 4F) at 973 K. Higher values could be reasonably expected if the measurement goes to higher temperature, which, unfortunately, exceeds the limit of the employed measurement systems. These values are better than the previously reported values illustrated in FIG. 5B. In addition, for Nb_0.9Ti_0.1FeSb and Nb_0.7Ti_0.3FeSb, the ZT value also reaches 1 at 973 K.

FIGS. 5C and 5D compare the power factor and ZT of a few materials with very high power factors previously reported for YbAl₃ single crystal, nanostructured Nb_0.95Ti_0.05FeSb (according to an embodiment of the present disclosure) and constantan, all possess peak power factor of at least 100 μW cm⁻¹ K⁻². Although the power factor in Nb_0.95Ti_0.05FeSb is not the highest, the ZTs are much higher than the other two materials due to the relatively lower thermal conductivity. It is noted that the “repeat” indication on these figures indicates that the experiments were repeated for those values/compositions as indicated.

To analyze the room temperature lattice thermal conductivity, the Callaway model was used, incorporating different scattering mechanisms. The 3P, GB, PD, and EP processes are considered, as discussed herein, fitting parameters are listed in Table 3.

TABLE 3
Parameters in Klemens model for Nb1−xTixFeSb
	PD	3P	EP	v
x	Γ(10−3)	ε	B (1011 s−1)	γ1	C × nH−4/3 (m4 s−1)	(m s−1)
0	0	80	8.06	0.69	3.86 × 10−25	3879
0.04	4.5		8.16			3871
0.05	5.6		8.19			3869
0.06	6.6		8.22			3867
0.07	7.6		8.24			3865
0.1	10.6		8.32			3859
0.2	19.0		8.59			3838
0.3	25.4		8.90			3818

FIG. 6 illustrates the contribution of different phonon scattering mechanisms to the lattice thermal conductivity of Nb_1-xTi_xFeSb at room temperature, these samples were formed via hot-pressing from about 1000° C. to about 1100° C. The indications of 3P, GB, PD, and EP represent three phonon process, grain boundary scattering, point defect scattering, and electron-phonon interaction, respectively. For grain boundary scattering, the grain size was set as 4.5 μm. The result is shown in FIG. 6. Indeed, the grain boundary scattering is much weaker comparing to the other scattering processes. Also notice the electron-phonon (EP) interaction in this system.

Power Output

Due to the higher power factor, higher power output is expected. FIGS. 7A and 7B show the calculated and measured output power density (FIG. 7A) and conversion efficiency of samples Nb_1-xTi_xFeSb with x=0.05 and 0.2 (FIG. 7B). The samples used for the measurements in FIGS. 7A and 7B were hot-pressed from about 1000° C. to about 1100° C. during formation.

The efficiency (η_max), and output power density (ω) under a large temperature gradient accounting for the Thomson effect are calculated and are shown in FIGS. 7A and 7B. With T_C=293 K and leg length L=2 mm. the calculated ω and η_max are, respectively, ˜28 Wcm⁻² and 8.8% in Nb_0.95Ti_0.05FeSb when T_H is 868 K (FIG. 7A). FIG. 7A illustrates that the measured values for both x=0.05 and x=0.2 are above 10 W cm⁻² at above about 600 K.

It has to be considered that the output power density is inversely related to the leg length (equation (5)). By normalizing the leg length to 2 mm and comparing the values of power density among different reports (see Table 4), a high value for bulk thermoelectric materials was achieved which is desirable for power generation. It is noted that the power density numbers in parentheses are normalized to a leg length of 2 mm using equation (5).

TABLE 4
Comparison among several previously reported output power density of various TE systems.
				ω of	ω of
Measurement	Materials	L		module	materials	Previously
system	(p-type/n-type)	(mm)	TC/TH (K)	(W cm−2)	(W cm−2)	Reported?
single leg	Nb0.95Ti0.05FeSb/—	2	293/868	—	21.5	no
single leg	Nb0.8Ti0.2FeSb/—	2	293/823	—	14.6	no
single leg	MgAg0.965Ni0.005Sb0.99/—	2	293/518	—	0.51	yes
unicouple	Nb0.6Ti0.4FeSb0.95Sn0.05/	2	373/873	—	6.5	yes
	Hf0.25Zr0.75NiSn0.99Sb0.01
unicouple	Hf0.44Zr0.44Ti0.12CoSb0.8Sn0.2/	2	373/873	—	5.5	yes
	Hf0.25Zr0.75NiSn0.99Sb0.01
Module	Nb0.86Hf0.14FeSb/	10	336/991	2.2 (11)*	—	yes
	Hf0.6Zr0.4NiSn0.98Sb0.02
Module	PbTe(MnTe)0.02Na0.04/	2.8	303/873	1.35 (1.89)	5.62 (7.87)	yes
	PbTe(Pbl2)0.002
Module	Mm0.30Fe1.46Co2.54Sb12.05/	4	313/733	0.46 (0.92)	1.15 (2.3)	yes
	Yb0.09Ba0.05La0.05Co4Sb12
Module	La0.7Ba0.1Ga0.1Ti0.1Fe3Co1Sb12/	7.6	323/873	1.28 (4.86)	—	yes
	Yb0.3Ca0.1Al0.1Ga0.1In0.1Co3.75Fe0.25Sb12

Table 4 comprises data from electron probe micro-analysis (EPMA) measurements on various samples fabricated according to embodiments of the present disclosure.

Nominal          EPMA
Nb0.96Ti0.04FeSb Nb0.95Ti0.04Fe1.03Sb0.98
Nb0.95Ti0.05FeSb Nb0.94Ti0.06Fe1.01Sb0.99
Nb0.94Ti0.06FeSb Nb0.93Ti0.07Fe1.01Sb0.99
Nb0.93Ti0.07FeSb Nb0.92Ti0.08Fe0.99Sb1.01
Nb0.9Ti0.1FeSb   Nb0.90Ti0.11Fe1.00Sb0.99
Nb0.8Ti0.2FeSb   Nb0.81Ti0.19Fe1.02Sb0.99
Nb0.7Ti0.3FeSb   Nb0.68Ti0.31Fe1.02Sb0.98

By increasing the hot pressing temperature up to 1373 K in half-Heuslers Nb_1-xTi_xFeSb (x=0.04, 0.05, 0.06, 0.07, 0.1, 0.2, and 0.3), the hot-pressed grain size is increases, resulting in higher carrier mobility and ultimately higher electrical conductivity and much higher power factor. The highest power factor of ˜106 μW cm⁻¹ K⁻² has been achieved in Nb_0.95Ti_0.05FeSb. The high power factor resulted in an record output power density of ˜22 W cm⁻² measured with the cold side and hot side temperatures of 293 K and 868 K, respectively. Such a high output power density is beneficial for large-scale thermoelectric power generation applications.

FIG. 9 is an image of an etched surface of Nb_0.95Ti_0.05FeSb fabricated according to certain embodiments of the present disclosure. In particular, FIG. 9 illustrates a sample of Nb_0.95Ti_0.05FeSb pressed at 1373 K that has been first polished by alumina sandpapers with a roughness of about 1 μm. The samples were subsequently fine-polished by about a 50 nm SiO₂ suspension on a Vibromet machine, resulting in a surface roughness in the order of 10 nm. The smooth surface was then etched for 10 to 15 seconds using etchant containing 50 ml distilled water, 1 ml HF acid (38%), 2 ml H₂O₂ (35%).

FIGS. 10A-10D are statistic grain sizes analyses of Nb_0.95Ti_0.05FeSb pressed at 1123 K (FIG. 10A), 1173 K (FIG. 10B), 1273 K (FIG. 10C) and 1373 K (FIG. 10D). The average size is area-weighed, thus larger grains have larger weight when averaging. The average grain size increases with increasing hot-pressing temperature, with grain sizes ranging up to 0.9 μm at a hot pressing temperature of 1123 K in FIG. 10A, and an average grain size of 0.28 μm. Grains formed subsequent to the 1173 K hot-pressing temperature in FIG. 10B are up to 1.8 μm with an average size of about 0.55 μm. FIG. 10C illustrates the distribution of grain sizes for grains formed subsequent to the 1273 K hot-pressing temperature are up to about 8 μm with an average size of about 0.2.92 μm, and FIG. 10D illustrates the distribution of grain sizes formed subsequent to the 1373 K hot-pressing temperature, the average grain size is about 4.49 μm and the grains range in size up to about 14 μm.

FIG. 11A is a cumulative grain size distribution of Nb_0.95Ti_0.05FeSb pressed at 1123 K, 1173 K, 1273 K and 1373 K that reflects the data illustrated in FIGS. 10A-10D.

FIG. 11B is a Williamson-Hall plot of grain size vs. strain for thermoelectric materials fabricated according to certain embodiments of the present disclosure. The grain size and strain from x-ray diffraction (XRD) spectra were analyzed using the full width at half maximum (FWHM) broadening using equation (20),

$\begin{array}{cc}\mathrm{FW}\left(S\right)*\cos \left(\theta \right)=\left(\mathrm{FW}-\mathrm{FW}\left(I\right)\right)*\cos \left(\theta \right)=\frac{0.94\lambda }{D_{\mathrm{xrd}}}+4\varepsilon \sin \left(\theta \right)& \left(20\right)\end{array}$

D_xrd is the effective grain size, ϵ is the strain. λ and θ are the wavelength and diffraction angles, respectively. FW(I) is the instrumental broadening, which is calibrated using a standard Si sample. Thus a plot of FW(S)*cos(θ) with respect to sin(θ) should be a straight line of which the intercept and slope are related with the size and strain, respectively.

FIG. 11C is a graph of the Hall carrier concentration (n_H) variation with the amount of dopants in Nb_1-xTi_xFeSb fabricated according to certain embodiments of the present disclosure. The dashed line shows the linear relationship of n_H with x when the latter is smaller than 10%. As shown in FIG. 11C, an increase in dopant has a linear relationship with an increase in n_H and continues to increase when x is greater than 10%.

FIG. 11D is a differential scanning calorimetry (DSS) graph of Nb_0.95Ti_0.05FeSb samples manufactured according to certain embodiments of the present disclosure. FIG. 11D illustrates DSC measurements up to 1573 K. A phase transition occurs at ˜1400 K and the melting point is around 1523 K. Thus, higher temperatures during hot pressing temperature may be beneficial for the TE performance.

FIGS. 12A-12C show thermoelectric properties for Nb_0.95Ti_0.05FeSb hot-pressed at 1373 K according to certain embodiments of the present disclosure. FIG. 12A illustrates the Seebeck coefficient and FIG. 12B shows the resistivity of individual branches in the valance band of Nb_0.95Ti_0.05FeSb pressed at 1373 K. Subscript L and H means the light and heave bands, respectively. FIG. 12C shows the bipolar thermal conductivity among different bands. κ_C-VL, κ_C-VH and κ_VL-VH are the bipolar thermal conductivity between conduction-V_L band; conduction-V_H band; and V_L band-V_H band, respectively.

FIG. 13 is an x-ray diffraction (XRD) pattern of Nb_1-xTi_xFeSb fabricated according to certain embodiments of the present disclosure. The values for x, as noted in FIG. 13, range from 0 to 0.3.

FIG. 14A illustrates the effect of grain size on the lattice thermal conductivity (κ_L), which increases with increasing grain size, it is appreciated that the 1373 K data represented the measured values and that the trend lines for 4.5 μm and 0.3 μm are the calculated values.

FIG. 14B illustrates the hole mobility μ_H which increases with increasing grain size. It is appreciated that the 1123, 1173, 1273, and 1373 K data represented are the measured values and that the trend lines for 4.5 μm, 3.0 μm, 0.5 μm, and 0.3 μm are the calculated values.

FIGS. 15A and 15B illustrate the power factor (FIG. 15A) and calculated power output density (FIG. 15B) for Nb_1-xTi_xFeSb fabricated according to certain embodiments of the present disclosure. The cold side temperature used for the calculations was 323 K.

Thus, a higher hot pressing temperature up to 1373 K generated a higher carrier mobility and a larger grain size, which lead to a much higher power factor (PF) of ˜106 μW cm⁻¹ K⁻² in p-type half-Heusler Nb_0.95Ti_0.05FeSb. With such a high power factor, a record output power density of ˜22 W cm⁻² may be calculated. The power output measurement confirms that ZT determines the conversion efficiency and PF determines the output power density. Thus PF is equally as desirable as ZT when choosing TE materials for power generation purposes.

FIG. 16 is a graph of thermoelectric current (I_TE)-voltage (V_TE) curve and power output at hot side temperature 773 K for thermoelectric materials fabricated according to embodiments of the present disclosure, including hot-pressing between 1000° C. and 1100° C. For efficiency and power density, the thermoelectric current (I_TE)-voltage (V_TE) measurement were performed at various hot side temperature (T_H) while the cold side temperature (T_C) is maintained at 293 K by a combined cooling from flowing liquid with a thermoelectric module (TEC). The Nb_0.95Ti_0.05FeSb and Nb_0.8Ti_0.2FeSb were cut and polished to a size of about 1.3×1.3 mm² in cross section and about 2 mm in height. Brazing materials (Ag_0.78Cu_0.22, liquidus ˜680° C.) were applied on both sides of the polished piece to form a mechanically strong bonding with low electrical resistance to the hot and cold junctions. During measurement the single leg system was vacuumed to below 10⁻⁴ mbar and surrounded by a heated copper shield for minimizing the convection and radiation loss.

In another example, an Nb-based half-Heusler material was fabricated according to the formula NbFe_1-yCo_ySb, including hot-pressing between 1000° C. and 1100° C. FIGS. 17A-17E are graphs of thermoelectric properties of n-type NbFe_1-yCo_ySb fabricated according to certain embodiments of the present disclosure. In particular, these figures illustrate the electrical conductivity (σ) (FIG. 17A), the Seebeck coefficient (S) (FIG. 17B), the power factor (PF) (FIG. 17C), the total thermal conductivity (κ_tot) (FIG. 17D) and the figure of merit (ZT) (FIG. 17E) of n-type NbFe_1-yCo_ySb. The total thermal conductivity as shown in FIG. 17D is below about 10.5 from about 480K to about 1000K.

FIGS. 18A and 18B illustrate the temperature dependency of thermoelectric materials fabricated according to certain embodiments of the present disclosure, including hot-pressing between 1000° C. and 1100° C. In some examples, the thermoelectric material may further comprise cobalt according to a formula of Nb_1-xTi_xFeCoSb. FIG. 18A shows the temperature dependence of the thermal diffusivity of Nb_1-xTi_xFeCoSb with x=0, 0.04, 0.05, 0.06, 0.07, 0.1, 0.2, and 0.3 and FIG. 18B shows the specific heat of Nb_1-xTi_xFeCoSb with x=0, 0.04, 0.05, 0.06, 0.07, 0.1, 0.2, and 0.3.

While exemplary embodiments of the invention have been shown and described, modifications thereof can be made by one skilled in the art without departing from the scope or teachings herein. The embodiments described herein are exemplary only and are not limiting. Many variations and modifications of the compositions, systems, apparatus, and processes described herein are possible and are within the scope of the invention. Accordingly, the scope of protection is not limited to the embodiments described herein, but is only limited by the claims that follow, the scope of which shall include all equivalents of the subject matter of the claims. Unless expressly stated otherwise, the steps in a method claim may be performed in any order and with any suitable combination of materials and processing conditions.

Where numerical ranges or limitations are expressly stated, such express ranges or limitations should be understood to include iterative ranges or limitations of like magnitude falling within the expressly stated ranges or limitations (e.g., from about 1 to about 10 includes, 2, 3, 4, etc.; greater than 0.10 includes 0.11, 0.12, 0.13, etc.). For example, whenever a numerical range with a lower limit, R_l, and an upper limit, R_u, is disclosed, any number falling within the range is specifically disclosed. In particular, the following numbers within the range are specifically disclosed: R=R+k*(R_u−R_l), wherein k is a variable ranging from 1 percent to 100 percent with a 1 percent increment, i.e., k is 1 percent, 2 percent, 3 percent, 4 percent, 5 percent, . . . , 50 percent, 51 percent, 52 percent, . . . , 95 percent, 96 percent, 97 percent, 98 percent, 99 percent, or 100 percent. Moreover, any numerical range defined by two R numbers as defined in the above is also specifically disclosed. Use of broader terms such as “comprises,” “includes,” and “having” should be understood to provide support for narrower terms such as “consisting of,” “consisting essentially of,” and “comprised substantially of.”

While exemplary embodiments of the invention have been shown and described, modifications thereof can be made by one skilled in the art without departing from the scope or teachings herein. The embodiments described herein are exemplary only and are not limiting. Many variations and modifications of the compositions, systems, apparatus, and processes described herein are possible and are within the scope of the invention. Accordingly, the scope of protection is not limited to the embodiments described herein, but is only limited by the claims that follow, the scope of which shall include all equivalents of the subject matter of the claims. Unless expressly stated otherwise, the steps in a method claim may be performed in any order and with any suitable combination of materials and processing conditions.

Claims

1. A method of fabricating a thermoelectric component, comprising: forming, an ingot of a first alloy comprising niobium, iron, antimony, and titanium according to a formula Nb1-xTixFeSb;ball-milling the ingot to form a ball-milled powder;hot-pressing the ball-milled powder between 1100 K and 1400 K to form a thermoelectric component comprising a power factor above about 60 μW cm−1 K−2 from about 300K to about 775K.

2. The method of claim 1, wherein x is from about 0.02 to about 0.7.

3. The method of claim 1, further comprising: subsequent to forming an ingot, melting the ingot at least once and reforming the ingot via a second melting.

4. The method of claim 1, further comprising ball-milling the ingot from 10 minutes to 70 hours.

5. The method of claim 1, further comprising hot-pressing the ball-milled powder to form a plurality of grains of an average grain size from about 0.25 μm to about 14 μm.

6. The method of claim 1, further comprising hot-pressing the ball-milled powder to form a plurality of grains of an average grain size above about 1.0 μm.

7. The method of claim 1, further comprising hot-pressing the ball-milled powder to form the thermoelectric component with a ZT above 0.6 above about 650 K.

8. A thermoelectric device comprising: a hot-pressed component of ball-milled powder according to a formula Nb1-xTixFeSb, wherein the component comprises a power factor above about 60 μW cm−1 K−2 from about 300 K to about 775 K and an average grain size above 1 μm.

9. The device of claim 8, wherein x is from 0.02-0.7.

10. The device of claim 8, wherein the power factor of the component is above about 80 μW cm−1 K−2 from about 300 K to about 500 K.

11. The device of claim 8, further comprising an output power density (ω) of at least 10 Wcm−2 above about 600 K.

12. The device of claim 8, further comprising a lattice thermal conductivity (κL) less than 8 Wm−1K−1 above about 500 K.

13. The device of claim 8, further comprising a plurality of grains of an average grain size from about 2.5 μm to about 4.50 μm.

14. A thermoelectric device comprising: a hot-pressed component of ball-milled powder according to a formula Nb1-xTixFeSb, wherein a power factor above about 60 μW cm−1 K−2 from about 300K to about 775K, and wherein x is from 0.02 to 0.7.

15. The device of claim 14, further comprising a plurality of grains of an average grain size from about 0.25 μm to about 14 μm.

16. The device of claim 14, further comprising a plurality of grains of an average grain size from about 2.5 μm to about 4.50 μm.

17. The device of claim 14, wherein the hot-pressed component comprises a total thermal conductivity (κtot) of less than about 10 W m−1 K−1 from about 425 K to about 1000 K.

18. The device of claim 17, wherein the hot-pressed component comprises a total thermal conductivity (κtot) of less than about 10.5 Wm−1K−1 from about 480K to about 1000K.

19. The device of claim 14, wherein a Hall mobility (μH) of the hot-pressed component is above 16 cm2 V s−2 from about 300K to about 450K.

20. The device of claim 14, wherein x is from 0.1 to 0.5.