Access to this full-text is provided by Springer Nature.
Content available from Nature Communications
This content is subject to copyright. Terms and conditions apply.
ARTICLE
Towards tellurium-free thermoelectric modules for
power generation from low-grade heat
Pingjun Ying1,7, Ran He1,7, Jun Mao 2, Qihao Zhang1, Heiko Reith 1, Jiehe Sui3, Zhifeng Ren2✉,
Kornelius Nielsch 1,4,5 ✉& Gabi Schierning 1,6✉
Thermoelectric technology converts heat into electricity directly and is a promising source of
clean electricity. Commercial thermoelectric modules have relied on Bi
2
Te
3
-based com-
pounds because of their unparalleled thermoelectric properties at temperatures associated
with low-grade heat (<550 K). However, the scarcity of elemental Te greatly limits the
applicability of such modules. Here we report the performance of thermoelectric modules
assembled from Bi
2
Te
3
-substitute compounds, including p-type MgAgSb and n-type Mg
3
(Sb,
Bi)
2
, by using a simple, versatile, and thus scalable processing routine. For a temperature
difference of ~250 K, whereas a single-stage module displayed a conversion efficiency of
~6.5%, a module using segmented n-type legs displayed a record efficiency of ~7.0% that is
comparable to the state-of-the-art Bi
2
Te
3
-based thermoelectric modules. Our work demon-
strates the feasibility and scalability of high-performance thermoelectric modules based on
sustainable elements for recovering low-grade heat.
https://doi.org/10.1038/s41467-021-21391-1 OPEN
1Leibniz Institute for Solid State and Materials Research, Dresden, Germany. 2Department of Physics and Texas Center for Superconductivity at the
University of Houston (TcSUH), University of Houston, Houston, TX, USA. 3National Key Laboratory for Precision Hot Processing of Metals, School of
Materials Science and Engineering, Harbin Institute of Technology, Harbin, China. 4Institute of Applied Physics, Technical University of Dresden,
Dresden, Germany. 5Institute of Materials Science, Technical University of Dresden, Dresden, Germany. 6Department of Physics, Experimental Physics,
Bielefeld University, Bielefeld, Germany.
7
These authors contributed equally: Pingjun Ying, Ran He. ✉email: zren@uh.edu;k.nielsch@ifw-dresden.de;gabi.
schierning@uni-bielefeld.de
NATURE COMMUNICATIONS | (2021) 12:1121 | https://doi.org/10.1038/s41467-021-21391-1 | www.nature.com/naturecommunications 1
1234567890():,;
Content courtesy of Springer Nature, terms of use apply. Rights reserved
More than 60% of the energy generated by burning fossil
fuels is dissipated as waste heat, of which more than
half is low-grade heat with temperatures <550 K1,2.
Effective harnessing this “cooler”heat to generate electricity is
vital for alleviating the burden on the energy supply and redu-
cing the emission of greenhouse gases. Although potential
technologies, such as the organic Rankine cycle3,4,thermo-
galvanic cells5, and thermo-osmotics6are being explored, these
are limited by their low efficiencies, short lifetimes, and diffi-
culty in system integration. In comparison, thermoelectric (TE)
technology stands out owing to its solid-state nature, which
guarantees ultra-long operational lifetime, and is particularly
attractive for heat-to-electricity conversion7.Thebroader
applicability of TE technology relies on the availability of high-
performance materials and modules that operate efficiently
below 550 K.
The energy conversion efficiency of a TE material is governed
by the dimensionless figure of merit zT,defined as zT =S2σT/κ
tot
,
where S,σ,T, and κ
tot
are the Seebeck coefficient, the electrical
conductivity, the absolute temperature, and the total thermal
conductivity, respectively. Among various materials tested to
date, Bi
2
Te
3
-based materials have unparalleled TE properties and
have thus been the focus of laboratory-scale demonstrations and
commercial devices that operate below 550 K with typical con-
version efficiencies of about 3–6%8–10. However, the wider
applicability of Bi
2
Te
3
-based commercial modules is severely
limited by the scarcity of Te with a concentration of <0.001 ppm
in the Earth’s crust11 and an annual production of less than 500
metric tons12. Therefore, it is imperative to develop TE modules
from other, more abundant materials while retaining high per-
formance at temperatures below 550 K.
In recent years, Mg-based materials, including n-type Mg
3
(Sb,
Bi)
2
13–17 and p-type MgAgSb18–21 have attracted great attention
from the TE community because of the nontoxic nature, abun-
dance of their constituent elements, and their high zT of ~1.0 at
temperatures <550 K. Moreover, these materials exhibit excellent
mechanical robustness and compatible TE properties between the
n-type and p-type TE materials22,23. Previous reports showed
excellent performances of these materials at the device level. For
example, Kraemer, et al. reported a ~8.5% efficiency of single-leg
MgAgSb operating between 293 K and 518 K24. Mao, et al.
improved the cooling performance by using Mg
3
(Sb,Bi)
2
to
replace the Bi
2
(Te,Se)
3
for n-type legs17. These work merit the
great potential of these materials in replacing the Bi
2
Te
3
for low-
grade heat recovery applications. However, for successful delivery,
it is essential to employ synthesis routines that are potentially
scalable for these Te-free TE modules, as well as to address their
device-level issues such as geometry optimization, brazing pro-
cess, and contact optimization, etc.23. Till now, despite their
promise, the assembly of these substitute compounds into power-
generation modules has not been reported.
Herein, we synthesized p-type MgAgSb and n-type Mg
3
(Sb,Bi)
2
compounds using direct mechanical alloying followed by rapid
current-assisted sintering. Note that for n-type Mg
3
(Sb,Bi)
2
,we
used less than 0.2% Te as the dopant in this work to secure the
material properties since its performance was widely validated, so
that the module is not completely free of Te. However, Te is not
an essential dopant with available alternatives, such as Sc, Nd, Y,
etc., yielded similar TE performances according to several recent
studies25–27. We reproduced these high-performance materials
with a synthesis routine that is potentially scalable. Such up-
scaling potential is critically important for heat-recovery appli-
cations. Subsequently, these high-zT compounds were translated
into high-performance TE modules. We realized a high conver-
sion efficiency of ~6.5% and ~7.0% under a temperature differ-
ence of ~250 K in a single-stage module and a segmented module,
respectively. Our efficiency is comparable to those reported for
Bi
2
Te
3
-based modules. This work marks a feasible, sustainable
alternative to Bi
2
Te
3
-based TE modules and will spur the appli-
cation of TE technology in converting low-grade heat to
electricity.
Results
Scalable preparation of thermoelectric materials and modules.
High-performance TE materials with simple synthesis are favored
for module fabrication. However, the synthesis of Mg-based
compounds usually involves procedures that are either compli-
cated, expensive, or time-consuming. For example, the synthesis
of Mg
3
(Sb,Bi)
2
compounds usually involves complicated proces-
sing routines including melting (such as arc melting, induction
melting, or traditional melting), pre-annealing, powerization,
sintering (such as spark plasma sintering, hot pressing, or
induction pressing), and post-annealing17,28–31. In another
example, MgAgSb, being in αphase at room temperature,
changes to the βphase at ~573 K, and to the γphase at ~633 K18.
Whereas only the αphase has the requisite high zT, phase-pure α-
MgAgSb is difficult to obtain using traditional melting techniques
unless a time-consuming annealing process is followed20.
An alternative synthesis routing were reported to overcome
such limitations for synthesizing the n- and p-type legs using only
three steps: weighing, mechanical alloying, and rapid sintering
(Fig. 1a, b; “Methods”)17,21. Following these reports, we here
Fig. 1 Fabrication of Te-free TE materials and modules. a Element weighing and mechanical alloying to prepare powder of the n- and p-type TE
compounds. bOne-step spark plasma sintering of materials and contact layers to prepare the TE legs. cAssembling of the Te-free TE module, including
polishing, cutting, loading, positioning, and brazing the TE legs to the pre-circuited AlN ceramic plates.
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-21391-1
2NATURE COMMUNICATIONS | (2021) 12:1121 | https://doi.org/10.1038/s41467-021-21391-1 | www.nature.com/naturecommunications
Content courtesy of Springer Nature, terms of use apply. Rights reserved
employed mechanical alloying not only because it is a lower-cost
way to realize large-scale production, but also because it allows
for accurate stoichiometry that ensures high reproducibility,
which is necessary to scale up production. This is especially
essential for this work since the compounds studied here are rich
in Mg, Bi, and Sb, which would otherwise largely evaporate if
traditional melting techniques were used. The phases of the TE
materials were characterized by X-ray diffraction (XRD). The
XRD patterns (Supplementary Fig. 1a, b) indicated high purity for
the n-type Mg
3.3
Bi
1.498
Sb
0.5
Te
0.002
(denoted as n-Sb0.5) and
Mg
3.3
Bi
1.298
Sb
0.7
Te
0.002
(denoted as n-Sb0.7), and the p-type
MgAg
0.97
Sb
0.99
(abbreviated as “p-MgAgSb”). We then undertook
scanning electron microscopy (SEM) and energy-dispersive X-ray
spectroscopy (EDX) elemental mapping of n- and p-type legs
with contact layers. The elemental distribution was nearly
uniform (Supplementary Fig.1c, d) and we could find no obvious
interaction between the TE materials and the contact layers.
Using the mechanically-alloyed powder samples, we then
fabricated the TE legs for module assembly by sintering in one
step the TE powder together with contact-layer powder on
both sides. Note that such one-step sintering was also applied for
segmented n-type legs. Herein, based on the previous reports,
we selected Fe and Ag as the contact layers for the n- and p-type
legs, respectively17,24. Finally, the sintered disks were diced and
polished into the desired geometry, placed in the proper
positions, and brazed to pre-circuited ceramic plates AlN in a
vacuum furnace (Fig. 1c). Due to the simplicity, our approach will
potentially reduce the assembling time greatly when compared to
the traditional routines.
Thermoelectric properties and module optimization.We
measured the transport properties of the TE materials, including
the electrical conductivity, the Seebeck coefficient, and the ther-
mal conductivity. The compounds synthesized in this work,
including n-Sb0.5, n-Sb0.7, and p-MgAgSb, possess similar
properties when compared to previous reports (Fig. 2) despite the
straight forward synthesis procedure17,20. For p-MgAgSb, we
obtained a peak zT of ~1.0 at 423 K and an average zT of ~0.9 at
temperatures ranging from room temperature to 548 K (Fig. 2d).
For the n-type materials, whereas n-Sb0.5 showed a higher zT
up to 423 K, n-Sb0.7 exhibited better performance at higher
temperatures (423 K to 548 K). The peak zT values reach 0.9 (at
423 K) and 1.2 (at 548 K) in n-Sb0.5 and n-Sb0.7, respectively.
Based on the zT profiles of n-Sb0.5 and n-Sb0.7, we postulated
that a segmented leg could maximize the average zT of the n-type
materials. According to the transport properties these com-
pounds, we employed finite element simulation to assist in
designing the geometrical configuration of the TE modules. With
a hot-side temperature (T
hot
) of 548 K and a cold-side
temperature (T
cold
) of 293 K, we evaluated the maximum
conversion efficiency as a function of the working current (I),
the ratio of the cross-sectional areas between the p- and n-type
legs (A
p
/A
n
), and the height ratio of the two n-type materials (H
n-
Sb0.7
/H
n-Sb0.5
) in a segmented leg (Fig. 3a and Supplementary
Fig. 2). We found that the ratio of H
n-Sb0.7
/H
n-Sb0.5
was optimal
over a large range from 0.75 to 1.75 (Supplementary Fig. 2). In
addition, the ratio of A
p
/A
n
was found to have a limited impact
on efficiency (Fig. 3a). These results suggest that the module
performance is not sensitive to the geometric factors, which is
beneficial since it tolerates certain deviations in the TE-leg
fabrication process without degrading the efficiency. Accordingly,
we fabricated TE modules with segmented n-type legs with
the ratio of A
p
/A
n
being unity to facilitate the device assembling.
Note that the segmented n-type legs were prepared by using
the same one-step sintering routine that was used for the single-
stage module (Fig. 1c) where an H
n-Sb0.7
/H
n-Sb0.5
ratio of ~1.5 was
selected. The selected H
n-Sb0.7
/H
n-Sb0.5
ratio yielded a temperature
profile that fully exploits the zT profiles of the n-type materials
(Fig. 2d and Supplementary Fig. 2).
To further evaluate the module quality, we characterized the
contact resistivity at the n-type/Fe and p-type/Ag junctions,
which was found to be 26.6 µΩ∙cm2and 6 µΩ∙cm2, respectively
(Fig. 3b). Although these values remain higher than those for the
benchmark Bi
2
Te
3
/Ni (1–5µΩ∙cm2)32 that was realized upon
optimizations for more than half a century, the overall interfacial
Fig. 2 Temperature-dependent thermoelectric properties of the fabricated materials. a Absolute Seebeck coefficient |α|, belectrical conductivity σ,
ctotal thermal conductivity κ
tot
, and dfigure of merit zT. n-Sb0.5: Mg
3.3
Bi
1.498
Sb
0.5
Te
0.002
(green triangles); n-Sb0.7: Mg
3.3
Bi
1.298
Sb
0.7
Te
0.002
(blue
diamonds); p-MgAgSb: MgAg
0.97
Sb
0.99
(red rings). The error bars represent the corresponding measurement uncertainties from the commercial devices.
NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-21391-1 ARTICLE
NATURE COMMUNICATIONS | (2021) 12:1121 | https://doi.org/10.1038/s41467-021-21391-1 | www.nature.com/naturecommunications 3
Content courtesy of Springer Nature, terms of use apply. Rights reserved
resistance impacts the efficiency of our module inconsiderably.
This was demonstrated by comparing the simulated internal
resistance, open-circuit voltage, output powers, and conversion
efficiency with and without the contact resistance of the single-
stage module (Supplementary Fig. 3) and segmented module
(Supplementary Fig. 4).
Based on the simulation results, we assembled single-stage and
segmented TE modules, each with 2 p-type and 2 n-type TE legs,
with individual leg dimensions of 2 × 2 × 6.5 mm3. The p-type
legs were MgAg
0.97
Sb
0.99
for all modules, but the compositions of
the n-type legs were altered in different modules, including
single-stage n-Sb0.5 and n-Sb0.7, and segmented n-Sb0.5/Sb0.7.
The output power (P), output heat flow (Q), and conversion
efficiency of these modules were characterized as functions of the
current (I) under a series of thermal loads (ΔT), in which the T
hot
was varied from 323 K to 523 K and T
cold
was maintained at
~293 K (Supplementary Fig. 5). As shown in Fig. 4, a high
conversion efficiency (η
max
) of ~6.5% was realized in the single-
stage modules for a temperature difference (ΔT) of 250 K.
Moreover, the module with segmented legs boosted the η
max
to
~7.0% for the same ΔT. The measured efficiencies were lower
than the simulation results where the respective values are ~8.8%
and ~9.3% for n-Sb0.7 single-stage and n-Sb0.5/Sb0.7 segmented
module. As shown in the supporting information, the reduced
measurement efficiency originates from an enlarged output heat
flow, whereas the output power (Supplementary Figs. 3-4) were
almost identical between simulations and measurements. The
larger output heat flow in measurement suggests the potential
existence of a thermal bypass, possibly due to the insufficient
vacuum level or because of the direct thermal radiation from the
hot side to the cold side in the Mini-PEM measurement setup
since the small filling factor in our module (~16%). Our state-of-
the-art modules are comparable to the Bi
2
Te
3
-based ones, and
could potentially be improved by a better thermal management.
In principle, our work demonstrated the feasibility of an Te-free
TE modules for extended applications due to their remarkable
sustainability.
Discussion
Te-free TE modules could provide a clean and effective way of
converting low-temperature waste heat to electricity. However,
their practical applicability has been hindered by difficulties in
synthesizing these Te-free materials on large scale, and high-
performance TE modules based on such materials have not been
successfully fabricated. We were able to synthesize high zT n-type
Mg
3
(BiSb)
2
and p-type MgAgSb by using a scalable routine that
combined mechanical alloying and current-assisted sintering. By
allowing a much simpler synthesis of such TE materials, our
approach presents a substantial advance in shortening the
synthesis period and reducing the elemental loss, which is espe-
cially essential for large-scale synthesis. The synthesized TE
materials were subsequently translated into single-stage and
segmented TE modules with conversion efficiency reaching
~6.5% and ~7.0%, respectively, for a temperature difference of
~250 K.
On the other hand, numerous challenges have to be overcome
before realizing the ultimate substitution of Bi
2
Te
3
module by a
Te-free one, since the former has been investigated for more than
half a century yet the latter is in its infancy. The required studies
include but are not restricted to (1) upscale to the level of kilo-
grams without degrading the TE properties, possibly by using
planetary ball milling; (2) thermal cycle test to examine the device
reliability at elevated temperatures; (3) long-term stability (in
years) under actual operating conditions and different atmo-
spheres such as current load and temperature gradient; (4)
techniques for packing and sealing to overcome potential
instabilities under atmosphere at elevated temperatures.
Despite the aforementioned challenges, this work thus realizes
high-performance TE modules free from Bi
2
Te
3
that are capable
of harvesting low-grade (<550 K) waste heat. The efficiency
demonstrated in this work exceeds that of the best Bi
2
Te
3
-based
modules. Subsequent enhancements are possible upon further
advance the material properties and optimize the filling factor of
Fig. 3 Module optimization. a Simulated efficiency (η) with respect to the A
p
/A
n
ratio and the working current (I) for the single-stage module (n-Sb0.7)
under T
hot
=543 K and T
cold
=293 K. bMeasured electrical contact resistivity at the n-type/Fe and p-type/Ag junctions. The red circles and blue
diamonds represent the scanning resistance across the junctions, the solid horizontal lines indicate the contact resistivity.
Fig. 4 Performance of the Te-free segmented modules. Comparison of the
measured conversion efficiency (η
max
, lines with symbols) among the Te-
free modules in this work under a series of temperature difference (ΔT).
The η
max
of the Bi
2
Te
3
-based modules from the literature (dashed lines)
were also plotted for comparison8,9,33–35. The inset shows the photograph
of a segmented Te-free module. The error bars represent the measurement
uncertainties from the commercial device.
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-21391-1
4NATURE COMMUNICATIONS | (2021) 12:1121 | https://doi.org/10.1038/s41467-021-21391-1 | www.nature.com/naturecommunications
Content courtesy of Springer Nature, terms of use apply. Rights reserved
the modules. The use of abundantly available elements and the
ease of fabrication render our modules a notable substitute for the
Bi
2
Te
3
-based modules in low-grade-heat recovery. This will
potentially spur the application of thermoelectric technology for
power generation from low-grade heat.
Methods
Synthesis of n- and p-type materials. High-purity powder Bi (99.9%), Sb
(99.99%), Te (99.99%), Mg (99.8%), and Ag (99.9%) were weighed out in the atomic
ratios of Mg
3.3
Bi
1.498
Sb
0.5
Te
0.002
(denoted as n-Sb0.5), Mg
3.3
Bi
1.298
Sb
0.7
Te
0.002
(deno-
ted as n-Sb0.7), and MgAg
0.97
Sb
0.99
(denoted as p-MgAgSb). For each sample, the
weighed elements were loaded into a hardened steel ball-milling jar in a glove box
under an argon atmosphere with an oxygen and water level below 1.0 ppm and then
ball-milled for 20 hours using a SPEX 8000D machine. The ball-milled powders were
subject to field-assisted sintering (FAST, FCT System GmbH) together with the
contact powders. Powders of iron (Fe, purity 99.8%) and silver (Ag, purity 99.9%)
were selected as the contact layers for n-type and p-type materials, respectively. The n-
type materials were sintered in a graphite die under a pressure of 50 MPa at 1023K
for 3 minutes, and the p-type materials were sintered in a tungsten carbide (WC) die
under a pressure of 120MPa at 553 K for 3min.
The phase purity and crystal structure of the samples were examined by X-ray
diffraction (XRD, Bruker D8, Co radiation) and their microstructures were
analyzed by scanning electron microscopy (SEM). The sample homogeneity was
characterized by energy-dispersive X-ray spectroscopy (EDX). The temperature-
dependent Seebeck coefficient (S) and electrical conductivity (σ) were measured by
the standard four-probe method (LSR-3, Linseis). The temperature-dependent
thermal diffusivity (λ) was measured by a laser flash method under a helium
atmosphere (LFA 1000, Linseis). The density (ρ) of the samples was measured by
the Archimedes method, and the heat capacity (C
p
) was obtained from previous
reports17,20. The thermal conductivity (κ
tot
) was calculated according to the
relation κ
tot
=λ∙ρ∙C
p
. The measurement uncertainties are 2%, 5%, and 7% for σ,
S, and κ
tot
, respectively, which yield an error in zT of ~18%.
Thermoelectric module fabrication and characterization. The sintered n- and p-
type bulk samples with contact layers were cut into legs using a diamond wire saw
in the dimension of 2.0 × 2.0 × 6.5 mm3. The TE legs (with the contacts), electrodes,
and the ceramic substrates were bonded in a single step. The bonding was enabled
by curing the silver paste at 548 K for 30 minutes in a high-vacuum tube furnace.
The dimension of the module is 10 × 10 mm2in cross-section and 9.3 mm in height
as a combination of legs (6.5 mm), electrodes (0.8 mm × 2), and ceramic plates
(0.6 mm × 2). Copper wires were soldered onto the cold-side electrodes for current
and voltage measurements. The electrical output power (P) and the conversion
efficiency (η) were measured under vacuum (< 5 × 10−2Pa) by a Mini PEM
(Advance Riko). To reduce the thermal contact loss, a graphite sheet (0.1 mm
thickness) and thermal silicone grease (ST1002, Slont) were sandwiched between
the module and the heater by a 60 N compression force. The hot-side temperature
(T
hot
) of the thermoelectric element varied from 323 K to 548 K, whereas the cold-
side temperature (T
cold
) was maintained at ~293 K. The radiative heat loss was
compensated by a built-in program. An efficiency uncertainty of ~10% is applied
based on the calibration results of a standard module that was provided by the
company.
Finite element simulation. COMSOL Multiphysics with Heat Transfer Module
was used to perform the three-dimensional finite-element simulations of the
power-generation characteristics for the thermoelectric module. A geometrical
model with the same dimensions as the experimental thermoelectric element was
used to calculate the electric power and heat flow outputs. Fourth-order polynomial
fittings of the temperature-dependent Seebeck coefficient, electrical conductivity,
and thermal conductivity for both n- and p-type materials were used as material
properties in the simulations.
Data availability
All data generated or analyzed during this study are included in the published article and
its Supplementary Information. The data that support the findings of this study are
available from the corresponding author upon reasonable request.
Received: 23 October 2020; Accepted: 13 January 2021;
References
1. Wang, D., Ling, X., Peng, H., Liu, L. & Tao, L. Efficiency and optimal
performance evaluation of organic Rankine cycle for low grade waste heat
power generation. Energy 50, 343–352 (2013).
2. Schierning, G. Bring on the heat. Nat. Energy 3,92–93 (2018).
3. Bao, J. & Zhao, L. A review of working fluid and expander selections for
organic Rankine cycle. Renew. Sust. Energ. Rev. 24, 325–342 (2013).
4. Vélez, F. et al. A technical, economical and market review of organic Rankine
cycles for the conversion of low-grade heat for power generation. Renew. Sust.
Energ. Rev. 16, 4175–4189 (2012).
5. Duan, J. et al. Aqueous thermogalvanic cells with a high Seebeck coefficient for
low-grade heat harvest. Nat. Commun. 9, 5146 (2018).
6. Straub, A. P., Yip, N. Y., Lin, S., Lee, J. & Elimelech, M. Harvesting low-grade
heat energy using thermo-osmotic vapour transport through nanoporous
membranes. Nat. Energy 1, 16090 (2016).
7. Bell, L. E. Cooling, heating, generating power, and recovering waste heat with
thermoelectric systems. Science 321, 1457–1461 (2008).
8. Hao, F. et al. High efficiency Bi
2
Te
3
-based materials and devices for
thermoelectric power generation between 100 and 300 C. Energy Environ. Sci.
9, 3120–3127 (2016).
9. Hu, X., Nagase, K., Jood, P., Ohta, M. & Yamamoto, A. Power generation
evaluated on a bismuth telluride unicouple module. J. Electron Mater. 44,
1785–1790 (2015).
10. Kaibe H., et al. Development of thermoelectric generating cascade modules
using silicide and Bi-Te. In: Proc. 23 rd International Conference on
Thermoelectrics.–Australia (2004).
11. Taylor, S. R. Abundance of chemical elements in the continental crust—a new
table. Geochim Cosmochim. Ac 28, 1273–1285 (1964).
12. Goldfarb R. J. Tellurium: providing a bright future for solar energy (US
Geological Survey, 2015).
13. Agne, M. T. et al. Heat capacity of Mg
3
Sb
2
,Mg
3
Bi
2
, and their alloys at high
temperature. Mater. Today Phys. 6,83–88 (2018).
14. Mao, J. et al. Anomalous electrical conductivity of n-type Te-doped
Mg
3.2
Sb
1.5
Bi
0.5
.Mater. Today Phys. 3,1–6 (2017).
15. Shuai, J. et al. Recent progress and future challenges on thermoelectric Zintl
materials. Mater. Today Phys. 1,74–95 (2017).
16. Xin, J. et al. Growth and transport properties of Mg
3
X
2
(X=Sb, Bi) single
crystals. Mater. Today Phys. 7,61–68 (2018).
17. Mao, J. et al. High thermoelectric cooling performance of n-type Mg
3
Bi
2
-
based materials. Science 365, 495–498 (2019).
18. Kirkham, M. J. et al. Ab initio determination of crystal structures of the
thermoelectric material MgAgSb. Phys. Rev. B 85, 144120 (2012).
19. Ying, P. et al. Hierarchical chemical bonds contributing to the intrinsically low
thermal conductivity in α‐MgAgSb thermoelectric materials. Adv. Funct.
Mater. 27, 1604145 (2017).
20. Ying, P. et al. High performance α-MgAgSb thermoelectric materials for low
temperature power generation. Chem. Mat. 27, 909–913 (2015).
21. Zhao, H. et al. High thermoelectric performance of MgAgSb-based materials.
Nano Energy 7,97–103 (2014).
22. Liu, Z. et al. Mechanical properties of nanostructured thermoelectric materials
alpha-MgAgSb. Scr. Mater. 127,72–75 (2017).
23. He, R., Schierning, G. & Nielsch, K. Thermoelectric devices: a review of
devices, architectures, and contact optimization. Adv. Mater. Technol. 3,
1700256 (2018).
24. Kraemer, D. et al. High thermoelectric conversion efficiency of MgAgSb-based
material with hot-pressed contacts. Energy Environ. Sci. 8, 1299–1308 (2015).
25. Shi, X. et al. Efficient Sc-doped Mg
3.05–x
Sc
x
SbBi thermoelectrics near room
temperature. Chem. Mat. 31, 8987–8994 (2019).
26. Li, J. et al. Point defect engineering and machinability in n-type Mg
3
Sb
2
-based
materials. Mater. Today Phys. 15, 100269 (2020).
27. Shi X. et al. Extraordinary n-Type Mg
3
SbBi thermoelectrics enabled by
Yttrium doping. Adv. Mater. 31, 1903387 (2019).
28. Zhang, J. et al. Discovery of high-performance low-cost n-type Mg
3
Sb
2
-based
thermoelectric materials with multi-valley conduction bands. Nat. Commun.
8, 13901 (2017).
29. Kuo, J. J. et al. Grain boundary dominated charge transport in Mg
3
Sb
2
-based
compounds. Energy Environ. Sci. 11, 429–434 (2018).
30. Zhang, J., Song, L., Borup,K. A., Jørgensen, M. R. V. & Iversen, B. B. New insight
on tuning electrical transport properties via chalcogen doping in n-type Mg
3
Sb
2
-
based thermoelectric materials. Adv. Energy Mater. 8, 1702776 (2018).
31. Tamaki, H., Sato, H. K. & Kanno, T. Isotropic conduction network and defect
chemistry in Mg
3+δ
Sb
2
-based layered Zintl compounds with high
thermoelectric performance. Adv. Mater. 28, 10182–10187 (2016).
32. Feng, S. et al. Reliable contact fabrication on nanostructured Bi
2
Te
3
-based
thermoelectric materials. Phys. Chem. Chem. Phys. 15, 6757–6762 (2013).
33. Zhu, B. et al. Realizing record high performance in n-type Bi
2
Te
3
-based
thermoelectric materials. Energy Environ. Sci. 13, 2106–2114 (2020).
34. Lu, X. et al. High‐efficiency thermoelectric power generation enabled by
homogeneous incorporation of MXene in (Bi, Sb)
2
Te
3
matrix. Adv. Energy
Mater. 10, 1902986 (2020).
35. Deng, R. et al. High thermoelectric performance in Bi
0.46
Sb
1.54
Te
3
nanostructured with ZnTe. Energy Environ. Sci. 11, 1520–1535 (2018).
NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-21391-1 ARTICLE
NATURE COMMUNICATIONS | (2021) 12:1121 | https://doi.org/10.1038/s41467-021-21391-1 | www.nature.com/naturecommunications 5
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Acknowledgements
We greatly acknowledge the financial support by the strategic project at IFW Dresden on
“Wireless sensor devices for high temperature applications”. Z.R. acknowledges the
Research Award from the Alexander von Humboldt Foundation, USA 1200926 USS, and
Q.Z. acknowledges the support from Alexander von Humboldt Foundation, CHN
1210297 HFST-P.
Author contributions
P.Y.,R.H.,Z.R.K.N.,andG.S.designedthework.P.Y.andH.R.assembledthe
modules and characterized the module performance. R.H. and P.Y. prepared the TE
materials and measured the transport properties. J.M. measured the contact resistance.
J.S. assembled the ceramic substrate with electrodes. P.Y. and Q.Z performed simu-
lations. P.Y., R.H., K.N., and G.S. wrote the manuscript. All authors edited the
manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Competing interests
The authors declare no competing interests.
Additional information
Supplementary information The online version contains supplementary material
available at https://doi.org/10.1038/s41467-021-21391-1.
Correspondence and requests for materials should be addressed to Z.R., K.N. or G.S.
Peer review information Nature Communications thanks Jean-Pierre Fleurial, Tsutomu
Kanno, Michihiro Ohta and the other, anonymous, reviewer(s) for their contribution to
the peer review of this work.
Reprints and permission information is available at http://www.nature.com/reprints
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
material in this article are included in the article’s Creative Commons license, unless
indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creativecommons.org/
licenses/by/4.0/.
© The Author(s) 2021
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-21391-1
6NATURE COMMUNICATIONS | (2021) 12:1121 | https://doi.org/10.1038/s41467-021-21391-1 | www.nature.com/naturecommunications
Content courtesy of Springer Nature, terms of use apply. Rights reserved
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
