nanomaterials

Review
Insights into One-Dimensional Thermoelectric Materials:
A Concise Review of Nanowires and Nanotubes
Giovanna Latronico 1 , Hossein Asnaashari Eivari 2, *, Paolo Mele 3 and Mohammad Hussein Naseef Assadi 4,5, *

1 National Research Council of Italy Institute of Condensed Matter Chemistry and Technologies for
Energy (CNR-ICMATE), Via G. Previati 1/E, 23900 Lecco, Italy
2 Physics Department, Faculty of Science, University of Zabol, Zabol 98613-53856, Iran
3 College of Engineering, Shibaura Institute of Technology, Omiya Campus, 307 Fukasaku, Minuma-ku,
Saitama City 337-8570, Japan
4 RIKEN Center for Emergent Matter Science (CEMS), Wako 351-0198, Japan
5 Chemistry Department, Faculty of Engineering and Natural Sciences, Istinye University, Sarıyer,
Istanbul 34396, Türkiye
* Correspondence: [email protected] (H.A.E.); [email protected] (M.H.N.A.)

Abstract: This brief review covers the thermoelectric properties of one-dimensional materials, such
as nanowires and nanotubes. The highly localised peaks of the electronic density of states near the
Fermi levels of these nanostructured materials improve the Seebeck coefficient. Moreover, quantum
confinement leads to discrete energy levels and a modified density of states, potentially enhancing
electrical conductivity. These electronic effects, coupled with the dominance of Umklapp phonon scat-
tering, which reduces thermal conductivity in one-dimensional materials, can achieve unprecedented
thermoelectric efficiency not seen in two-dimensional or bulk materials. Notable advancements
include carbon and silicon nanotubes and Bi3 Te2 , Bi, ZnO, SiC, and Si1−x Gex nanowires with sig-
nificantly reduced thermal conductivity and increased ZT. In all these nanowires and nanotubes,
efficiency is explored as a function of the diameter. Among these nanomaterials, carbon nanotubes
offer mechanical flexibility and improved thermoelectric performance. Although carbon nanotubes
theoretically have high thermal conductivity, the improvement of their Seebeck coefficient due to
their low-dimensional structure can compensate for it. Regarding flexibility, economic criteria, ease
of fabrication, and weight, carbon nanotubes could be a promising candidate for thermoelectric
Citation: Latronico, G.; Asnaashari
power generation.
Eivari, H.; Mele, P.; Assadi, M.H.N.
Insights into One-Dimensional
Keywords: computational materials science; thermal conductivity; 1D thermoelectric nanomaterials;
Thermoelectric Materials: A Concise
figure of merit; nanowire; carbon nanotubes; flexibility
Review of Nanowires and Nanotubes.
Nanomaterials 2024, 14, 1272. https://
doi.org/10.3390/nano14151272

Academic Editor: Nikos Tagmatarchis 1. Introduction

Received: 20 June 2024 When the dimensions of a material decrease, due to quantum confinement, its prop-
Revised: 23 July 2024 erties change significantly. Accordingly, the focus of materials science over the past few
Accepted: 24 July 2024 decades has been on how material properties can be manipulated by controlling dimen-
Published: 29 July 2024 sionality, as the potential applications are endless. Some possible applications include
magnetic [1,2], optical [3,4], electronic [5,6], bioactive [7,8], mechanical [9,10], and ther-
moelectric (TE) [11,12] properties, of which, TE properties are being focused on in this
minireview. Other instances encompass photocatalytic hydrogen production [13], artificial
Copyright: © 2024 by the authors.
ligaments [14], printable mesoscopic solar cells [15], and flexible biomimetic e-whiskers [16].
Licensee MDPI, Basel, Switzerland.
Thermoelectricity is a property of materials that causes the conversion of a temperature
This article is an open access article
difference into electricity and vice versa, thanks to two effects, respectively, named the
distributed under the terms and
Seebeck and Peltier effects. The first is the basis of thermoelectric generators (TEGs), which
conditions of the Creative Commons
are simple and robust devices with no moving parts, making them perfect candidates for
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
energy harvesting and powering wireless devices. In the current energy and pollution crisis
4.0/).
and the exponential increase in the number of off-grid IoT devices, thermoelectricity could

Nanomaterials 2024, 14, 1272. https://doi.org/10.3390/nano14151272 https://www.mdpi.com/journal/nanomaterials

Nanomaterials 2024, 14, 1272 2 of 18

be one of the critical technologies in the future [17]. The Peltier effect has applications in
silent cooling, where the current flowing in the device generates a temperature difference
between the two sides of the module.
The thermoelectric figure of merit, ZT, is a dimensionless measure of the efficiency of
a thermoelectric material; the higher its value, the higher the efficiency of the material. ZT
is defined as:
ZT = (S2 σT)/(κ e + κ l ), (1)
where S is the Seebeck coefficient, σ is the electrical conductivity, κ e and κ l are electron ther-
mal conductivity and phonon thermal conductivity, and T is the absolute temperature [18].
The primary current limitation on the large-scale application of this technology is
its low efficiency. The arduous challenge lies in combining in the same solid material
the electronic properties of a crystal and the thermal ones of glass, following the phonon
glass electron crystal (PGEC) concept, decoupling the numerator and the denominator of
Equation (1). The Wiedemann–Franz law puts constraints on the dependence of electrical
conductivity and electron thermal conductivity:

κ e = L0 σT, (2)

where L0 is the Lorenz number (2.45 × 10−8 W S−1 K−2 ). Consequently, it is impossible to
independently manipulate the electronic contributions σ and κ e arbitrarily. An alternative
route to increase ZT through a reduction in thermal conductivity is to work on the phonon
component κ l , which can be mathematically expressed as:

κ l = 1 /3 Cv vΛ, (3)

where Cv is the specific heat at constant volume, v is the sound velocity, and Λ is the
phonon mean free path. Since phonon propagation is highly sensitive to surface scattering
via reduced Λ, nanoengineering 1D materials presents an unprecedented opportunity to
improve thermoelectric efficiency.
Significant efforts are underway to improve TEGs. In this framework, new materials
and new dopants are of great interest, but dimensionality has played a considerable
role in the recent progress in the field [19,20]. Compared to bulk materials, significant
improvements in the thermoelectric figure of merit, ZT, have been recently reported in
nanostructures. The advantage of low dimensionality can be explained by the reduction
in κ l and the improvement in the device’s power factor, S2 σ. For thermoelectricity, using
low-dimensional materials introduces additional freedom through the length scale. This
unique characteristic enables the deliberate manipulation of the materials’ electronic and
thermal properties, resulting in enhanced thermoelectric performance.
Reduced dimensionality leads to electron and phonon quantum confinement, resulting
in discrete energy levels and a modified density of states (DOS) [21,22]. Electronically, this
confinement can enhance electrical conductivity, depending on the alignment of energy
levels and carrier-scattering mechanisms [23]. In nanowires and nanotubes, carriers are
confined to move along a single axis, which can increase carrier mobility by reducing
scattering rates, provided the structure is chemically pure and defect free. However, surface
scattering and edge effects can significantly impact electrical conductivity, causing consid-
erable anisotropy [24]. The enhancement of the Seebeck coefficient in low-dimensional
materials can be attributed to the localisation of the DOS in one-dimensional (1D) sub-
bands, expressed as S(E) ∝ [δ(DOS)/δE]/[DOS] [25]. For phonons, Umklapp scattering, a
process where phonon momentum is not conserved within the first Brillouin zone, becomes
more significant. This process impedes the involved phonon’s heat-carrying capacity, re-
ducing thermal conductivity [26]—albeit with caveats [27]. Additionally, in nanowires and
nanotubes, phonons frequently encounter the surfaces of the wires or tubes, leading to
significant surface scattering. These scatterings reduce the mean free path of phonons, fur-
ther lowering thermal conductivity [28]. Compared to bulk materials or two-dimensional
nanomaterials, a significant advantage of one-dimensional structures is the flexibility of the
Nanomaterials 2024, 14, 1272 3 of 18

resulting devices, which can be bent without delamination or cracking [29]. However, it
is important to note that 2D thermoelectric materials have unique niche applications due
to their relatively easy synthesis via exfoliation from layered bulk materials [30,31]. Un-
doubtedly, one-dimensional materials still offer many possibilities for engineering highly
efficient and practical thermoelectric devices.
So far, researchers have reported several types of one-dimensional nanostructures that
have shown improved efficiency in thermoelectric applications. These include superlattice
structures [32], silicon nanowires (NWs) [33], both carbon and non-carbon nanotubes
(NTs) [34], and nanocomposite structures [35]. Numerous publications have explored
this subject, and interested readers can find more comprehensive information in these
excellent reviews [34,36–38]. This article provides a brief overview of the performance of
standard classes of one-dimensional thermoelectric materials, explicitly focusing on the
computational aspects of thermal conductivity, which are arguably the most challenging to
manage due to numerical complexity [11,39].

2. Non-Flexible Inorganic 1D Thermoelectric Materials

In 1993, Hicks and Dresselhaus [40], using analytical and statistical calculations,
predicted that quantum effects could provide a new way to design thermoelectric materials.
As indicated in Figure 1a, in the case of nanowires, the calculated ZT increases significantly
with decreasing wire diameter below the thermal de Broglie wavelength of4the
Nanomaterials 2024, 14, x FOR PEER REVIEW carriers.
of 18
This increase is because of a change in the density of states. In addition, in 1D structures,
electron movements are confined to a single dimension parallel to the nanowire. Such
confinement leads to notoscattering
propelled researchers seriouslyoffconsider
the surface. Therefore, the mobility
Si nanostructures of carriers in the
as up-and-coming
nanowire
candidatesdirection remains applications.
for thermoelectric unchanged.

Figure (a)Theoretical
1.(a)
Figure 1. TheoreticalZTZT values
values vs. vs. diameter
diameter foranisotropic
for an an anisotropic conducting
conducting nanowirenanowire of a square
of a square
cross sectionfabricated
cross section fabricatedalong
alongthethe lattice
lattice x, y,x,and
vector’s
vector’s y, and z directions.
z directions. (b)progression
(b) The in ZT in ZT
The progression
improvement through
improvement throughthetheyears. Adapted
years. Adapted withwith
permission
permission Refs. [37,40]
from from . Copyrights
Refs. [37,40]. 1993 1993
Copyrights
American Physical Society and 2013 Springer Nature.
American Physical Society and 2013 Springer Nature.

Nomura et al. fabricated fishbone-type silicon phononic crystal (PnC) nanostructures

via electron beam lithography and measured the thermal conductivity via micro-time-
domain thermoreflectance [42]. Their experiment showed that the PnC had lower thermal
conductivity than a nanowire with a corresponding width (Figure 2b). More specifically,
for an effective width of 145 nm, the thermal conductivity was lowered from around 65
Nanomaterials 2024, 14, 1272 4 of 18

Furthermore, the increased phonon scattering from the surface of a thin wire leads to a
reduction in the structure’s thermal conductivity and, hence, an increase in ZT. Accordingly,
Hicks and Dresselhaus predicted a ZT as high as 14 for a Bi2 Te3 nanowire of 5 Å diameter.
The outcomes of these theoretical predictions definitely demonstrate the immense potential
of quantum nanowire structures as thermoelectric nanomaterials with a high ZT. As
depicted in Figure 1b, the groundbreaking works by Hicks and Dresselhaus—credited with
seminal papers exploring the figure of merit for confined spaces in two-dimensional [41]
and one-dimensional [40] systems—ushered in a new era of thermoelectric research. These
publications marked a significant milestone in the field, propelling the investigation of
TE materials towards the realm of nanoengineering, subsequently leading to remarkable
advancements in their thermoelectric figure of merit [11,12].
Hochbaum et al. reported the synthesis of large-area, wafer-scale Si nanowires [33].
The result of the thermal conductivity measurement in their work is shown in Figure 2a.
The thermal conductivity of bulk Si was about 150 W m−1 K−1 . However, nanowires with
diameters of ~50 nm exhibited a much smaller thermal conductivity of ~4 W m−1 K−1 ,
which approached the amorphous limit for Si. Additionally, the Seebeck coefficient and the
electrical conductivity of the wires remained comparable to those of doped bulk Si. As a
result, the nanowires had a ZT of 0.6. The significant reduction in thermal conductivity
was attributed to nanowires’ surface roughness, which strongly screens a broad spectrum
of phonons and alters the phonon transmission through the confined structures. Silicon is
environmentally friendly and abundant. Further, the current silicon-based microelectronic
industry is highly developed. Consequently, the significant ZT enhancement observed in
Si NWs and their numerous inherent advantages have propelled researchers to seriously
consider Si nanostructures as up-and-coming candidates for thermoelectric applications.
Nomura et al. fabricated fishbone-type silicon phononic crystal (PnC) nanostructures
via electron beam lithography and measured the thermal conductivity via micro-time-
domain thermoreflectance [42]. Their experiment showed that the PnC had lower thermal
conductivity than a nanowire with a corresponding width (Figure 2b). More specifically,
for an effective width of 145 nm, the thermal conductivity was lowered from around
65 W m−1 K−1 to ~45 W m−1 K−1 . Nonetheless, comparing the theoretical band engineer-
ing analysis and the thermal conductivity measurements on several fishbone-like PnC
nanostructures with different fin-width-to-period ratios raised some discrepancies, which
could be explained by the increase in the travelling distance of the phonons within the fin
part of the nanostructure.
Zhang et al. [43] were able to fabricate through a metal-assisted chemical etching
method some high-density and large-area, vertically aligned porous silicon nanowire arrays
(SiNWAs) on the two sides of silicon substrates, called sandwich structured composites
(SSCs), with control over porosity and length. They investigated the SSCs’ thermoelectric
properties to obtain the ZT values of the corresponding SiNWAs at room temperature. They
observed that ZT increases, together with the length and porosity of the structures (up to
547 m2 g−1 ). The peak value of the figure of merit was 0.493 for a large-porosity SiNWA,
which is 77 times higher than that of bulk silicon (0.0064), thanks also to a remarkable
Seebeck coefficient, equal to 513 µV K−1 .
By introducing a small hole at the centre of the Si NW, i.e., constructing a silicon
nanotube (NT) structure, Chen et al. [44] demonstrated that the room-temperature thermal
conductivity of the structure could be reduced by 35%, as shown in Figure 2c. Holey Si
was fabricated by dry-etching silicon on an insulator wafer. In a similar work, the lattice
thermal conductivity of holey Si was reported to be close to that of amorphous SiO2 at
room temperature, and its ZT value was reported to be 0.4 [45]. However, competitive ther-
moelectric performance in 1D Si nanostructures still required a further reduction in thermal
conductivity. Subsequently, a follow-up experimental study [46] reported an excellent
power factor and low thermal conductivity of 1.8 W m−1 K−1 , close to the amorphous limit,
for silicon nanowires. The work presented a silicon-based thermoelectric generator made
of a collection of highly p-doped silicon nanowires, remarkably enhancing the electrical
Nanomaterials 2024, 14, 1272 5 of 18

Nanomaterials 2024, 14, x FOR PEER REVIEW 5 of 18

power density. The authors argued that the generator is suitable for applications exploiting
small ∆T values.

Figure
Figure2.2.(a)
(a)The
Thedependence
dependence of of
thermal
thermal conductivity
conductivityon temperature
on temperaturefor Sifor
nanowires with different
Si nanowires with different
diameters scratched from wafers of different resistivities. The curve with open squares is the thermal
diameters scratched from wafers of different resistivities. The curve with open squares is the thermal
conductivity of bulk amorphous Si. (b) The thermal conductivity of a fishbone-like Si nanostructure
conductivity of bulk amorphous Si. (b) The thermal conductivity of a fishbone-like Si nanostructure
compared to that of similar-width nanowires. (c) The dependence of thermal conductivity on
compared tofor
temperature that of similar-width
Si nanowires (Si NWs)nanowires. (c) The dependence
and Si nanotubes (Si NTs). Theof thermal conductivity
cross-section areas for Sion
NTstemper-
ature
and Si for
NWsSi are
nanowires
7.30 nm2(Si
andNWs) and
7.37 nm Si nanotubesIn(Si
2, respectively. (c),NTs). The cross-section
the right-axis labels show areas for Si NTs and
the reduction
in
Si κNWs
in nanotubes nm2 and 7.37
are 7.30 compared nmin2 ,same-width
to that respectively.nanowires.
In (c), theThe insets in labels
right-axis (a,c) show
showthethe
nanowire
reduction in
and nanotube structures. Adapted with permission from Refs. [33,42,44]
κ in nanotubes compared to that in same-width nanowires. The insets in (a,c) showand . Copyrights 2008 the2014
nanowire
Nature Springerstructures.
and nanotube and 2010 American
Adapted Chemical Society. from Refs. [33,42,44]. Copyrights 2008 and 2014
with permission
Nature Springer and 2010 American Chemical Society.
By introducing a small hole at the centre of the Si NW, i.e., constructing a silicon
nanotube
Apart(NT)
fromstructure, Chen et al. in
Si-based nanowires, [44] demonstrated
a recent that the
experimental work room-temperature
[47], a nanostructure-
thermal conductivity of the structure could be reduced by 35%, as shown in Figure 2c.
integrated chip composed of CdSSe nanowires was reported to be an excellent photo-
Holey Si was fabricated by dry-etching silicon on an insulator wafer. In a similar work,
thermoelectric generator. The Seebeck coefficient and the thermoelectric output voltage
the lattice thermal conductivity of holey Si was reported to be close to that of amorphous
of the chip were reported to be −152.4 Mv K−1 and 10.8 mV, respectively. Moreover, the
SiO2 at room temperature, and its ZT value was reported to be 0.4 [45]. However,
authors reported a significant elevation in the output voltage, up to 45 mV, upon illumina-
 Nanomaterials 2024, 14, 1272 6 of 18

tion by white light. The elevation of the voltage was attributed to the high concentration of
photo-generated charge carriers.
Alongside these experimental breakthroughs, computational investigations have regu-
larly provided new insights into one-dimensional thermoelectricity. Neophytou et al. [48]
conducted a simplified tight-binding density functional investigation to study thermoelec-
tricity in Si NWs using 20 orbitals for Si atoms and accounting for the spin–orbit coupling.
Their models consisted of Si NW of different diameters constructed from cylindrical cuts
through Si’s zincblende lattice. Their results demonstrated that for Si NWs with diameters
below 7 nm, the power factor and ZT values can be enhanced by up to 100% compared
to bulk values. Another theoretical investigation by Sansoz [49], with non-equilibrium
molecular dynamics calculations with the Stillinger–Weber potential [50], showed that
nanoengineering the Si NW’s outer surface can substantially reduce the lattice thermal
conductivity. According to the author’s prediction, the smooth-surfaced and circular Si NW
had a thermal conductivity of 36 W m−1 K−1 . In contrast, the Si NW with a saw-toothed
surface with (111)/(100) facet terminations had a thermal conductivity of 16 W m−1 K−1 ,
which is 10 times smaller than the prediction for bulk Si. Moreover, using Monte Carlo
simulations for phonon transport within the diffuse and ballistic limits, Jean et al. [51]
studied heat conduction in Si nanowires that were 2 µm long and 115 nm wide. They
particularly investigated the role of constrictions, a pinch-like narrowing in the middle
of Si NWs. It was clear that the thermal conductivity of the NW can be lowered and
modulated by up to 90% through a fine adjustment of the steepness of the constrictions, its
diameter ratio to the nanowire’s width, its length, and other parameters. The minimum κ
was predicted to be less than 10 µW m−1 K−1 .
Additionally, the thermoelectric properties of one-dimensional Bi nanowires and
nanotubes were shown to depend on diameter critically. Theoretical investigations using a
semiclassical transport model indicated that the ZT of n-type Bi NWs with a diameter of
5 nm can exceed 6 at 77 K [52]. Moreover, by combining the effective mass envelope wave
function and the Boltzmann transport equation, the theoretical results in Figure 3 show the
effect of the diameter and thickness of n-type Bi nanotubes on their power factor, electron
Nanomaterials 2024, 14, x FOR PEER REVIEW
thermal conductivity, lattice thermal conductivity, and maximum ZT at 77 K. As a final 7 o
result, panel d indicates that when the thickness and diameter of Bi NTs decrease to 2 nm
and 10 nm, the value of their ZT can increase to 6 [53].

Figure3.3.Variations
Figure Variations in calculated
in the the calculated optimum
optimum power power factor
factor (a), (a), thermal
electron electronconductivity
thermal conductivity
(b),
lattice thermal conductivity (c), and optimum thermoelectric ZT (d)
lattice thermal conductivity (c), and optimum thermoelectric ZT (d) with thickness for with thickness for n-type
n-type Bi
nanotubeswith
nanotubes with different
different diameters
diameters oriented
oriented along
along the the trigonal
trigonal directiondirection at 77 K.with
at 77 K. Adapted Adapted w
permission
permission from
from [53].[53]
Ref.Ref. . Copyright
Copyright 2010 American
2010 American InstituteInstitute of Physics.
of Physics.

Further theoretical calculations using equilibrium molecular dynamics and

Green–Kubo formula [54,55] showed that compared to bulk zinc oxide, the ZT value
ZnO nanowires increases by 30 times, reaching ~0.06, when the diameter of the w
decreases to 8 Å [56]. All electron density functional calculations with generali
Nanomaterials 2024, 14, 1272 7 of 18

Further theoretical calculations using equilibrium molecular dynamics and the Green–
Kubo formula [54,55] showed that compared to bulk zinc oxide, the ZT value of ZnO
nanowires increases by 30 times, reaching ~0.06, when the diameter of the wire decreases
to 8 Å [56]. All electron density functional calculations with generalised gradient approxi-
mation [57] and the Boltzmann transport model on n-type silicon–germanium (Si1−x Gex )
nanowires ((110) direction with 2.3 nm2 rectangular cross section) predicted that the ZT
value can be increased by 4.3 times over that of Si of the same dimensionality [58]. In
corroboration with this prediction regarding Si-based nanowires, an experimental inves-
tigation showed that the figure of merit of β-SiC nanowires (diameters of 60~100 nm) is
about 120 times higher than the maximum ZT value of bulk SiC, reaching values up to
0.12 [59].

3. Flexible Thermoelectric Devices with 1D Carbon Nanotubes

Due to inherent brittleness and rigidity, conventional bulk TE materials, which are
based on inorganic semiconductors, cannot establish contact with the curved surfaces of
heat sources and sinks well enough. This rigidity imposes limitations on their application
in modern electronics. So, it is desired to fabricate flexible TE materials with high thermo-
electric performance. Carbon nanotubes (CNTs), shown in Figure 4a,b, have some potential
utility in overcoming this limitation. CNTs are attractive because they are flexible, and
their source materials are abundant and nontoxic [60]. They can also be fabricated using
high-throughput solution-phase synthesis techniques, and they have high specific energy
(i.e., W g−1 ) enabled by their low mass [61]. Also, the thermal properties of CNT arrays
are tuneable. There are detailed reviews of CNT TE materials, including theoretical and
experimental efforts [36], so the coverage here is concise.
Commonly, three methods are used to synthesise CNTs [62]: electric arc discharge,
chemical vapour deposition (CVD), and pulsed laser vaporisation. Figure 4c shows a
general comparison between Seebeck coefficients of highly oriented pyrolytic graphite
(HOPG), multi-walled CNTs (MWCNTs), and single-walled CNTs (SWCNTs). All samples
were subjected to ambient atmosphere and light for a long time so they could be considered
sufficiently oxygen doped. The figure shows that SWCNTs and MWCNTs have room-
temperature Seebeck coefficients of approximately 45 µV K−1 and 17 µV K−1 , respectively.
These values are much higher than that of HOPG. Figure 4c indicates the enhancement of
the Seebeck coefficient with a reduction in the size of the CNTs. The sign of the Seebeck
coefficient of SWCNTs and MWCNTs is reported to be positive, indicating that oxygen-
doped nanotubes are p-type.
Although CNTs have high thermal conductivity [63], which is unacceptable for thermo-
electric applications, a computational study demonstrated that a semiconducting SWCNT
can be optimised to make its thermoelectric performance competitive [64]. The study
revealed that p-type SWCNTs show higher power factors originating from their more sig-
nificant electrical conductivity, which was attributed to a relatively larger DOS around the
valence band maximum and longer hole relaxation times. It also showed that the ZT value
of the CNT at elevated temperatures can be remarkably enhanced to 1.9 by a reduction in
its thermal conductivity using isotopic substitution and hydrogen sorption. Consequently,
superior electronic transport properties of earth-abundant and environmentally friendly
CNTs, coupled with an effective reduction in their thermal conductivity, could guarantee
viable thermoelectric performance. Moreover, based on the electronic structure, the SWCNT
can be either metallic (m-SWCNT) or semiconductor (s-SWCNT). Statistical reports indicate
that one-third of all possible SWCNTs show metallic behaviour, while two-thirds behave
as semiconductors. This ratio is maintained in almost all SWCNT synthesis procedures.
Efforts have been made to develop methods to separate as-prepared SWCNT samples
based on diameter, electronic structure, and chirality [36].
The thermopower dependence on the electronic structure and diameter of CNTs is
depicted in Figure 4d. The figure is a part of the results of theoretical calculations carried
out by Hung et al. [65] for s-SWCNTs using a method in which the Boltzmann trans-
 of the Seebeck coefficient of SWCNTs and MWCNTs is reported to be positive, indicating
that oxygen-doped nanotubes are p-type.
Nanomaterials 2024, 14, 1272 8 of 18 for
Although CNTs have high thermal conductivity [63], which is unacceptable
thermoelectric applications, a computational study demonstrated that a semiconducting
SWCNT can be optimised to make its thermoelectric performance competitive [64]. The
port
studyformalism
revealedwasthatcombined with an extended
p-type SWCNTs show highertight-binding model.originating
power factors The thermopower
from their
trend (Figure 4d) demonstrated a strong relationship between enhanced
more significant electrical conductivity, which was attributed to a relatively thermopower
larger andDOS
reduced nanotube diameter. Benchmarking α values against the bandgaps reported in
around the valence band maximum and longer hole relaxation times. It also showed that
Figure 4e indicated that the thermopower of a given s-SWNT is also related to its bandgap.
the ZT value of the CNT at elevated temperatures can be remarkably enhanced to 1.9 by
Accordingly, for a tube with a small diameter, the thermopower could exceed 2000 µV K−1
a reduction in its thermal conductivity using isotopic substitution and hydrogen sorption.
at room temperature. This value is about 10 times higher than that of commonly used
Consequently,materials.
thermoelectric superiorAs reported
electronicby the
transport
authors, properties of earth-abundant
the large thermopower of nanotubesand
environmentally
is friendly CNTs, coupled
related to their one-dimensionality and thewith
largean effectiveofreduction
bandgaps in their
s-SWNTs with small thermal
di-
conductivity, could guarantee viable thermoelectric performance. Moreover,
ameters. More recently, Taborowska et al. [66] experimentally demonstrated the potential based on the
electronic structure,
application of carbon the SWCNT(CNTs)
nanotubes can beforeither metallic (m-SWCNT)
thermoelectric or semiconductor
purposes through substantial (s-
SWCNT).doping
molecular Statistical
andreports indicate
the inclusion that one-third
of ZnO nanowires.ofThis
all possible
treatmentSWCNTs
enhancedshow metallic
the electri-
behaviour,
cal whileoftwo-thirds
conductivity behave ascarbon
pure single-walled semiconductors. This ratioarrays
nanotube (SWCNT) is maintained
by a factorin of
almost
3.
all SWCNT synthesis
Additionally, the powerprocedures. Efforts
factor improved have
from been
34.9 m−1 K
µWmade to−develop
2 for unmodified
methodsSWCNTs
to separate
as-prepared
to −1 K−2 for
42.91 µW mSWCNT samples based with
CNTs treated on diameter,
molecularelectronic
doping andstructure, and chirality
ZnO nanowire inclusion.[36].

Figure4.4.(a,b)
Figure (a,b)Schematic
Schematicrepresentation
representationofofsingle-walled
single-walled and
and multi
multi (four)-walled
(four)-walled carbon
carbon nanotubes.
nanotubes.
(c) Dependence of the Seebeck coefficient α on temperature for air-saturated single-walled
(c) Dependence of the Seebeck coefficient α on temperature for air-saturated single-walled nanotubes
nanotubesmulti-walled
(SWNTs), (SWNTs), multi-walled nanotubes
nanotubes (MWNTs), and(MWNTs), and highly
highly oriented oriented
pyrolytic pyrolytic
graphite (HOPG). graphite
EA
(HOPG). EA and PLV refer to synthesis methods, i.e., electric arc and pulsed laser vaporisation. (d)
and PLV refer to synthesis methods, i.e., electric arc and pulsed laser vaporisation. (d) Optimum
Optimum thermopower (Seebeck coefficient α) values vs. SWNT diameter for s-SWNTs at room
thermopower (Seebeck coefficient α) values vs. SWNT diameter for s-SWNTs at room temperature.
temperature. (e) The Kataura plot shows the family pattern of the SWNT bandgap vs. diameter.
(e) The Kataura plot shows the family pattern of the SWNT bandgap vs. diameter. Adapted with
Adapted with permission from Refs. [62,65]. Copyrights 2006 Springer and 2015 American Physical
permission from Refs. [62,65]. Copyrights 2006 Springer and 2015 American Physical Society.
Society.
Experimentally, Zhou et al. reported a straightforward method for fabricating a
Theand
compact thermopower dependence
flexible thermoelectric ondevice,
(TE) the electronic
shown instructure
Figure 5a,and diameter
leveraging of CNTs is
large-area
depicted in Figure 4d. The figure is a part of the results of theoretical calculations
continuous synthesis of CNT films and localised doping technology [67]. Figure 5b shows carried
out by Hung et al. [65] for s-SWCNTs using a method in which the
the schematics of the fabrication process. The prepared module showed low internalBoltzmann transport
formalismand
resistivity wasdidcombined withgold
not require an extended tight-binding
or silver top model. Theonthermopower
electrode deposition each TE device trend
(Figure
or 4d)interconnections
metallic demonstrated betweena strongn-relationship between
and p-type legs, enhanced
further thermopower
simplifying the assemblyand
reducedThis
process. nanotube
technique diameter. Benchmarking
effectively avoids designαcomplications
values against the to
related bandgaps reported in
contact resistance
Figure 4e indicated that the thermopower of a given s-SWNT is also related to itsmodule
and can be easily scaled up for large-scale production. The output voltages of the bandgap.
at various steady temperature differences are depicted in Figure 5c. Figure 5d shows
the voltage–current and power–current curves of the module under the condition in
Nanomaterials 2024, 14, 1272 9 of 18

which the hot-side temperature is TH = 330 K and the steady-state temperature difference
is ∆T = 27.5 K. At the mentioned temperatures, the power density of the module was
estimated to be 167 µWcm−2 . In this instance, because of the relatively high thermal
conductivity, the CNT-based TE modules seem suitable for working at low-temperature
differences. Furthermore, in a more recent experimental work [68], the fabrication of a
flexible thermoelectric device based on a Bi2 Te3 –carbon nanotube hybrid was reported. The
hybrid structure shows good flexibility under bending. Thermoelectrically, it exhibits a10ZT
Nanomaterials 2024, 14, x FOR PEER REVIEW of 18
value of about 0.23 at 330 K and a maximum output power density of about 0.93 mWcm−2
under a steady temperature difference of 25 K at room temperature.

Figure
Figure 5.5. (a) AA thermoelectric
thermoelectricmodulemodulemademadeofofalternating
alternatingp- p-
and
andn-type doped
n-type dopedCNTs
CNTsoperating
operating
betweenhot
between (THH)) and
hot(T andcold
cold(T(TCC))reservoirs.
reservoirs.(b)
(b)Schematics
Schematicsofofthe
thefabrication
fabrication process
process forfor sequenced
sequenced p- p-
n-type TE
and n-type TEcomponents
componentsbased basedon onlarge-area
large-area continuously
continuously synthesised
synthesised CNTCNT films
films andand localised
localised
doping technology.
doping technology. (c) (c) The
Thevoltage
voltagegenerated
generatedbetween
betweenthethetwotwo ends
ends of of
thethe module
module in different
in different
steady-temperaturedifferences.
steady-temperature differences.(d) (d)The
Thevoltage–current
voltage–current curve
curve andand
thethe power–current
power–current curve
curve of the
of the
module at a hot-side temperature of 330 K and a temperature difference of
module at a hot-side temperature of 330 K and a temperature difference of 27.5 K. Adapted from27.5 K. Adapted from
Ref. [67] . Copyright 2017 the authors (CC
Ref. [67]. Copyright 2017 the authors (CC BY 4.0). BY 4.0).

Evans etdepending
Notably, al. [76] conducted non-equilibrium
on their structure molecular
form (individual, dynamics
films, bundled, simulations
bucky papers,using
etc.) and method of synthesis, CNTs’ thermal conductivity varies from
Tersoff and Lennard Jones potentials, accounting for both covalent and dispersion the nearly thermal
insulation,
interaction,with a thermal
for two types ofconductivity
(10, 10) CNT of bundles, W m−1crossbar
about 0.1namely K−1 , to such high values
and parallel as as
arrays,
6600 W in − 1
m Figure − 1
K [36,69].
shown 6b. TheyThis variability
found is partly due to
that at atmospheric thermal crossbar
pressure, energy transport through
and parallel arrays
ahave
CNTabeing predominantly
considerably governed bythermal
low out-of-plane the phonon conductionofmechanism
conductivity 0.028 and [70,71].
0.55 W Even
m−1 K−1,
in metallic CNTs, heat transport is dominated by the collective vibration
respectively. These κ values are orders of magnitude lower than the axial thermal of atoms [72].
Subsequently, topological defects in a CNT, such as simple or Stone–Wales-type
conductivity of isolated CNTs (observed to be 3000~6000 W m⁻¹ K⁻¹ [74]). When vacan-
cies and extraneous
comparing the two impurities
bundles, one(other forms
should of carbon,
note residual
that because catalyst,
of the etc.),
unified scatter theand
orientation
phonons and decrease their mean free path. In addition, CNT–CNT coupling in bundled
much larger contact area among the CNTs, the thermal conductivity of the parallel array
samples can decrease thermal conductivity by about an order of magnitude compared to
at low pressure is still a few times larger than that of the crossbar array.
isolated CNTs [73]. Generally, in a stand-alone CNT, thermal conductivity increases with
Reverse non-equilibrium molecular dynamics (MD) simulation based on the
AIREBO potential [77] conducted by Boroushak et al. [78] showed a reduction in thermal
conductivity when the outer sides of single-walled (10, 10) and double-walled (5, 5)@(10,
10) CNTs were functionalised with polyethylene (PE) groups. The calculations showed
that thermal conductivity decreases with increasing weight percentage of the functional
Nanomaterials 2024, 14, 1272 10 of 18

the CNT length up to the mean free path of phonons, which is about ~500 nm for MWCNTs
and SWCNTs [74]. Furthermore, thermal conductivity is also predicted to decrease with
increasing CNT diameter. In this regard, using the first-order perturbation theory with
constant relaxation time and the three-phonon Umklapp process, Cao et al. found a nearly
inverse power relationship between the CNT’s diameter and its thermal conductivity [75]
(Figure 6a).
Evans et al. [76] conducted non-equilibrium molecular dynamics simulations using
Tersoff and Lennard Jones potentials, accounting for both covalent and dispersion interac-
tion, for two types of (10, 10) CNT bundles, namely crossbar and parallel arrays, as shown
in Figure 6b. They found that at atmospheric pressure, crossbar and parallel arrays have a
considerably low out-of-plane thermal conductivity of 0.028 and 0.55 W m−1 K−1 , respec-
tively. These κ values are orders of magnitude lower than the axial thermal conductivity
of isolated CNTs (observed to be 3000~6000 W m−1 K−1 [74]). When comparing the two
bundles, one should note that because of the unified orientation and much larger contact
area among the CNTs, the thermal conductivity of the parallel array at low pressure is still
a few times larger than that of the crossbar array.
Reverse non-equilibrium molecular dynamics (MD) simulation based on the AIREBO
potential [77] conducted by Boroushak et al. [78] showed a reduction in thermal conduc-
tivity when the outer sides of single-walled (10, 10) and double-walled (5, 5)@(10, 10)
CNTs were functionalised with polyethylene (PE) groups. The calculations showed that
thermal conductivity decreases with increasing weight percentage of the functional group,
i.e., a ~50 decrease in κ for ~25% of the polymer weight ratio. According to these results,
the thermal conductivity of the SWCNT is more sensitive to the change in the weight
percentage of functional groups. Moreover, using atomistic Green’s function calculations,
Chalopin et al. [79] found that the phonon thermal conductivity of densely compacted
pellets composed of almost 1 µm long carbon nanotubes has an upper bound of about
5 W m−1 K−1 , which is about a thousand times lower than that of an individual CNT or
graphene. They also predicted that the upper κ bound is not dependent on the nanotube
diameter at room temperature.
An experimental study on sheets of different lengths of MWCNTs was reported by
Aliev et al. [80]. As shown in Figure 6c, in the temperature range of 75~150 K, the length
of MWCNT sheets does not influence the thermal conductivity value in a significant way
(Figure 6c). However, above 150 K, longer samples had higher thermal conductivities.
They also found that at T = 295 K, the thermal conductivity value along the films with
a length of about 0.37 mm is 50 W m−1 K−1 , while across the films, the conductivity
value is 2.1 W m−1 K−1 . Figure 6d shows that in multilayer systems, thermal conductivity
decreases with increasing numbers of layers. The authors attributed the observed low
thermal conductivity value to the respective nanotubes’ internal defects and the boundaries’
phonon scattering in bundles forming the basis of MWCNT sheets.
To summarise, comparing the theoretical and experimental findings discussed before,
it becomes evident that computational predictions can confidently be considered an upper
bound for thermal conductivity. For example, Berber et al. conducted molecular dynamics
simulations using the Tersoff potential to calculate the thermal conductivity of isolated
(10, 10) CNTs, resulting in κ = 6600 W m−1 K−1 [81]. However, most experimental reports
find κ to be significantly lower. This discrepancy arises from the idealised structural
models used in simulations. In reality, most synthetic methods, especially top-down
approaches, produce materials with numerous defects [82,83]. These defects introduce
various phonon-scattering mechanisms, reducing thermal conductivity. Notably, when
simulations account for these structural complexities, the agreement between theory and
experiment improves [76,78].
 isolated (10, 10) CNTs, resulting in κ = 6600 W m⁻¹ K⁻¹ [81]. However, most experimental
reports find κ to be significantly lower. This discrepancy arises from the idealised
structural models used in simulations. In reality, most synthetic methods, especially top-
down approaches, produce materials with numerous defects [82,83]. These defects
introduce various phonon-scattering mechanisms, reducing thermal conductivity.
Nanomaterials 2024, 14, 1272 11 of 18
Notably, when simulations account for these structural complexities, the agreement
between theory and experiment improves [76,78].

Figure 6.
Figure 6. (a)
(a) Theoretical
Theoreticalcalculations
calculationsshowing
showing thethe
relationship between
relationship thermal
between conductivity
thermal and and
conductivity
CNT diameter parameterised by increasing the first component of the (n, 0) vector. The inset shows
CNT diameter parameterised by increasing the first component of the (n, 0) vector. The inset shows
how folding graphene produces a (n, m) CNT. (b) Pressure dependence of cross-plane thermal
how folding of
conductivity graphene produces
single-walled a (n,
(10, 10) m) CNT.
carbon (b) Pressure
nanotubes arranged dependence
in a parallelofbundle
cross-plane thermal
structure,
conductivity of single-walled
shown in the upper-left inset, and (10, 10) carbon
a crossbar nanotubes
array, arranged
shown in the in a inset.
lower-right parallel bundle structure,
(c) Measurement
of thermal
shown conductivity
in the upper-leftas a function
inset, and a of temperature
crossbar array, along
shown mono-walled carbon inset.
in the lower-right nanotube sheets of
(c) Measurement
different
of thermallengths: 7.6 mmas(open
conductivity circles),of5.4
a function mm (open squares),
temperature and 0.37 mm
along mono-walled (solid nanotube
carbon circles). (d)
sheets
Thermal conductivity of mono-walled carbon nanotube sheets versus several superimposed layers
of different lengths: 7.6 mm (open circles), 5.4 mm (open squares), and 0.37 mm (solid circles).
(d) Thermal conductivity of mono-walled carbon nanotube sheets versus several superimposed
layers with parallel alignment. Adapted with permission from Refs. [75,76,80]. Copyrights 2004
American Physical Society, 2007 Elsevier, and 2012 American Institute of Physics.

4. Outlook and Perspective

As seen in Figure 7a, the window for carrier concentration (or any other experimentally
adjustable parameter) where ZT is maximum for any given material is narrow. Traditionally,
experimentalists have relied on judicious and assiduous sets of experiments, aggressively
engineering candidate materials to fine-tune design parameters for the best TE performance
that falls within this narrow range [17,18]. As seen in the literature reviewed before, though
there have been numerous independent theoretical investigations, synergetic reliance on
computational insight has often been scarce in the rational design of thermoelectric ma-
terials. In this regard, theory and experiment have mainly progressed in parallel. The
limitation in collaboration stems from a series of computational bottlenecks preventing
widescale and dependable computational aid. These bottlenecks were especially severe in
calculating the electron relaxation time [12] and the phonon contribution to thermal con-
ductivity [11]. Both quantities are complex to calculate and rely on many approximations,
diminishing their predictive relevance to experimental works.
It is important to note that the outstanding works reviewed here have been conducted
over the past three decades. Therefore, the computational limitations should be acknowl-
edged accordingly. Moving forward, however, using new data analytic tools that were
previously unavailable seems to be indispensable. Noticeably, in the past few years, reliance
on data analytic methods in materials discovery has increased exponentially, encompassing
many aspects [84–86]. Here, we would like to focus on an approach particularly applicable
 for the best TE performance that falls within this narrow range [17,18]. As seen in the
literature reviewed before, though there have been numerous independent theoretical
investigations, synergetic reliance on computational insight has often been scarce in the
rational design of thermoelectric materials. In this regard, theory and experiment have
mainly progressed in parallel. The limitation in collaboration stems from a series of
Nanomaterials 2024, 14, 1272 12 of 18
computational bottlenecks preventing widescale and dependable computational aid.
These bottlenecks were especially severe in calculating the electron relaxation time [12]
and the phonon contribution to thermal conductivity [11]. Both quantities are complex to
to low-dimensional
calculate thermoelectrics:
and rely on the proliferation
many approximations, of machine
diminishing learningrelevance
their predictive for calculating
to
the electron relaxation
experimental works. time and the phonon contribution to thermal conductivity.

Figure 7.
Figure 7. (a)
(a) The
TheSeebeck
Seebeckcoefficient (SCoeff
coefficient ), electrical
(SCoeff conductivity
), electrical (σ), and
conductivity thermal
(σ), conductivity
and thermal (κ)
conductivity
of polycrystalline Pb 1−xSnxTe as a function of carrier concentration (n) at 300 K. ZT achieves its peak
(κ) of polycrystalline Pb1−x Snx Te as a function of carrier concentration (n) at 300 K. ZT achieves
value within a narrow range of n, highlighted in yellow. Varying the temperature shifts these curves
its peak value within a narrow range of n, highlighted in yellow. Varying the temperature shifts
in a complex and nontrivial way. (b) The proposed workflow for high-throughput nanomaterial
these curves in a complex and nontrivial way. (b) The proposed workflow for high-throughput
nanomaterial screening for finding new ZT heights, relying on materials informatics and machine
learning techniques. Panel (a) adapted from Ref. [12]. Copyright 2020 the authors (CC BY 4.0).

Calculating the electron relaxation time is challenging because it is influenced by

scattering from various mechanisms, including vacancies, interstitial atoms, impurity
atoms, grain boundaries, and phonons [87]. Consequently, designing a completely ab initio
model to obtain the relaxation time is practically impossible. Therefore, it is frequently
approximated as a constant value, disregarding the effects of carrier concentration (doping
level) and temperature. When applied to high-throughput surveys, such approximations
are bound to result in overly optimistic predictions. However, Hirosawa et al. recently made
significant progress in accurately determining the electron relaxation time using machine
learning projection trained on experimental data [88]. This technique was successfully
applied to Ta4 SiTe4 , producing a reasonable match between theory and experiment [89]. At
this stage, the transferability of such relaxation time to new materials is yet to be examined.
Using classical molecular dynamics has been computationally the most straightfor-
ward and efficient way of estimating the lattice contribution to thermal conductivity, espe-
cially for materials with complex morphologies. However, the lack of suitable force fields
that produce accurate trajectories for the desired set of elements within an extensive tem-
Nanomaterials 2024, 14, 1272 13 of 18

perature range has always posed a limitation. Recently, machine-learning-based force fields
(MLFFs) have offered some hope in this regard [90–92]. MLFFs, trained on high-quality
density functional data (e.g., as implemented in VASP 6 at https://vasp.at/, accessed
17 July 2024), ensure that the force fields capture the complex interactions between atoms
more accurately than empirical potentials. As a result, the calculated thermal conductivity
using MLFFs is closer to experimental values [93,94]. Additionally, once trained, MLFFs can
perform simulations much faster than ab initio methods for supercells containing tens of
thousands of atoms. This speed advantage allows for the simulation of larger systems and
longer time scales, allowing the investigation of dilute dopants, low-symmetry structures,
and complex interfaces.
The powerful tools described before, combined with impressive materials databases,
such as the Materials Project (with approximately 150,000 entries) [95], AtomWork-Adv
(containing around 400,000 crystal structures, 500,000 materials properties, and 48,000
phase diagrams) [96], and 2DMatPedia (which includes several thousand two-dimensional
materials) [97], can automate high-confidence, high-throughput searches for thermoelectric
materials, as schematically shown in Figure 7b. Therefore, it is highly anticipated that
future discoveries in thermoelectric nanostructures will depend on a renewed synergy
between theory and experiment.
Experimentally, common synthesis routes for one-dimensional materials include elec-
trospinning. This highly efficient top-down method uses electrostatic force to create contin-
uous nanofibres. Moreover, solution-phase methods, like hydrothermal and solvothermal
techniques, facilitate chemical reactions under controlled conditions of pH, time, pressure,
temperature, and additives. Other techniques include electrochemical deposition methods,
such as chemical vapour deposition for semiconductor nanowires, atomic layer deposition
for precise thickness control, and confined or oriented template synthesis methods like
reactive sputtering [29,98,99]. Future experimental development should address concerns
with scalability and uniformity by improving the precise control required over synthesis
conditions and reducing the cost and environmental impact by eliminating toxic elements
and reagents.
Ensuring the stability of the presented 1D materials at high temperatures and their
integration into devices without degrading their thermoelectric properties is a significant
challenge. In particular, carbon nanotubes degrade with increasing temperature, primarily
through oxidation, leading to carbon dioxide and other gases forming, causing structural
damage. An optical absorbance technique investigating the temperature dependence
of vertically aligned single-walled carbon nanotube degradation revealed that significant
oxidisation starts at T = ~400 ◦ C [100]. Multi-walled carbon nanotubes were found to oxidise
at even lower temperatures of ~200 ◦ C [101]. Given these limitations, the thermoelectric
application of CNTs is limited to temperatures near ambient.
Finally, we would like to draw the readers’ attention to the necessity of a wholesome
approach to assessing thermoelectric materials and devices. Though a thermodynamically
important factor, the maximum attainable figure of merit ZT usually fails to illustrate a com-
plete picture regarding practical application concerns. For instance, economic constraints
often set a trade-off between efficiency and cost when expensive elements or techniques
are required to increase ZT. Efforts have been made to address the incomplete nature of
the ZT metric by developing an analytical framework based on the cost per watt (USD/W)
metric, demonstrating that minimising the cost per watt for thermoelectric materials and
systems can sometimes, but not always, align with maximising ZT [102].
Furthermore, in some breakthrough thermoelectric materials, the maximum ZT occurs
within a narrow temperature range. For example, in Bi0.5 Sb1.5 Te3 [103] and Cu-doped
Bi2 Te2.7 Se0.3 [103], ZT peaks around ~1.9 and ~1.2 at temperatures centred at 300 K and
400 K, respectively. However, outside narrow temperature ranges within ±20 K, ZT drops
rapidly by several folds, leading to reliability issues in real-life applications. Consequently,
a comprehensive approach is essential for assessing thermoelectric materials and devices,
Nanomaterials 2024, 14, 1272 14 of 18

as the maximum figure of merit ZT often fails to address practical application concerns,
including economic constraints and temperature-dependent reliability issues.

5. Conclusions
This brief review demonstrated how reducing the dimensions of materials can signifi-
cantly alter their properties due to quantum confinement, enabling the circumvention of the
main limitations in thermoelectric technology. One-dimensional nanostructures improve
thermoelectric performance by reducing phonon thermal conductivity and enhancing the
power factor. Among the materials presented, Si nanowires with reduced thermal con-
ductivity and maintained electrical properties have shown significant ZT enhancements,
making them promising candidates for thermoelectric devices when integrated with other
Si-based semiconducting components. However, carbon nanotubes offer flexibility and
high specific energy, broadening the use cases for thermoelectric materials. The thermal con-
ductivity in carbon nanotubes can be reduced to competitive values of ~10 W m−1 K−1 by
introducing defects and preparing the nanotubes in the bundle, fibre, film, and bucky-paper
forms. Combined with one of the highest power factors among one-dimensional structures,
the ZT value for CNT-based TE devices can reach up to 0.42 at 300 K. Building on these
significant achievements to date and the latest computational advancements, including
machine learning techniques for determining the electron relaxation time and machine-
learning-based force fields for estimating thermal conductivity, the future of thermoelectric
research is poised to increasingly depend on theory-guided experiments.

Author Contributions: Conceptualisation, M.H.N.A.; investigations, G.L., H.A.E., P.M. and M.H.N.A.;
formal analysis, G.L., H.A.E. and M.H.N.A.; supervision, P.M. and M.H.N.A.; funding acquisition,
M.H.N.A.; writing—original draft, G.L., H.A.E., P.M. and M.H.N.A. All authors have read and agreed
to the published version of the manuscript.
Funding: This work was also supported by the University of Zabol, Iran (grant no. UOZ-GR-9618-43).
Data Availability Statement: Data sharing is not applicable to this article, as no new data were
created in this study. All data reported here can be traced back to the cited literature.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Arabi, H.; Eivari, H.A. Applying a suitable route for preparation Fe3 O4 nanoparticles by Ammonia and investigation of their
physical and different magnetic properties. Int. J. Nano Dimens. 2014, 5, 297–303.
2. Huang, P.; Zhang, P.; Xu, S.; Wang, H.; Zhang, X.; Zhang, H. Recent advances in two-dimensional ferromagnetism: Materials
synthesis, physical properties and device applications. Nanoscale 2020, 12, 2309–2327. [CrossRef] [PubMed]
3. Rahdar, A.; Eivari, H.A.; Sarhaddi, R. Study of structural and optical properties of ZnS:Cr nanoparticles synthesized by co-
precipitation method. Indian J. Sci. Technol. 2012, 5, 1855–1858. [CrossRef]
4. Wu, J.; Ma, H.; Yin, P.; Ge, Y.; Zhang, Y.; Li, L.; Zhang, H.; Lin, H. Two-dimensional materials for integrated photonics: Recent
advances and future challenges. Small Sci. 2021, 1, 2000053. [CrossRef]
5. Eivari, H.A.; Ghasemi, S.A.; Tahmasbi, H.; Rostami, S.; Faraji, S.; Rasoulkhani, R.; Goedecker, S.; Amsler, M. Two-dimensional
hexagonal sheet of TiO2 . Chem. Mater. 2017, 29, 859–8603. [CrossRef]
6. Li, Z.; Yan, T.; Fang, X. Low-dimensional wide-bandgap semiconductors for UV photodetectors. Nat. Rev. Mater. 2023, 8, 587–603.
[CrossRef]
7. Sadeghi, B.; Garmaroudi, F.S.; Hashemi, M.; Nezhad, H.; Nasrollahi, A.; Ardalan, S.; Ardalan, S. Comparison of the anti-bacterial
activity on the nanosilver shapes: Nanoparticles, nanorods and nanoplates. Adv. Powder Technol. 2012, 23, 22–26. [CrossRef]
8. Khodadad, H.; Hatamjafari, F.; Pourshamsian, K.; Sadeghi, B. Microwave-assisted Synthesis of Novel Pyrazole Derivatives and
their Biological Evaluation as Anti-Bacterial Agents. Comb. Chem. High Throughput Screen. 2021, 24, 695–700. [CrossRef]
9. Sohbatzadeh, Z.; Eivari, H.A.; Fakhrabad, D.V. Formation energy and some mechanical properties of hydrogenated hexagonal
monolayer of GeC. Physica B 2018, 547, 88–91. [CrossRef]
10. Jiang, H.; Zheng, L.; Liu, Z.; Wang, X. Two-dimensional materials: From mechanical properties to flexible mechanical sensors.
InfoMat 2020, 2, 1077–1094. [CrossRef]
11. Eivari, H.A.; Sohbatzadeh, Z.; Mele, P.; Assadi, M.H.N. Low thermal conductivity: Fundamentals and theoretical aspects in
thermoelectric applications. Mater. Today Energy 2021, 21, 100744. [CrossRef]
Nanomaterials 2024, 14, 1272 15 of 18

12. Gutiérrez Moreno, J.J.; Cao, J.; Fronzi, M.; Assadi, M.H.N. A review of recent progress in thermoelectric materials through
computational methods. Mater. Renew. Sustain. Energy 2020, 9, 16. [CrossRef]
13. Yang, B.; Wang, Z.; Zhao, J.; Sun, X.; Wang, R.; Liao, G.; Jia, X. 1D/2D carbon-doped nanowire/ultra-thin nanosheet g-C3 N4
isotype heterojunction for effective and durable photocatalytic H2 evolution. Int. J. Hydrogen Energy 2021, 46, 25436–25447.
[CrossRef]
14. Li, C.; Jia, R.; Yang, Y.; Liao, G. A Hierarchical Helical Carbon Nanotube Fiber Artificial Ligament. Adv. Fiber Mater. 2023, 5,
1549–1551. [CrossRef]
15. Shao, W.; Cen, R.; Sheng, J.; Ma, D.; He, J.; He, J.; Ding, T.; Wu, W. 28.35 mW cm−2 power output assisted concentrating printable
CsPbBr3 solar cell-thermoelectric module tandem device using two-pronged strategy. Device 2024, 2, 100285. [CrossRef]
16. Chen, C.; Li, X.-L.; Zhao, S.; Song, Y.; Zhu, Y.; Wang, Q.; Zhong, C.; Chen, R.; Li, E.; Li, Z.; et al. A biomimetic e-whisker sensor
with multimodal perception and stimuli discrimination. Device 2023, 1, 100148. [CrossRef]
17. Zhang, Q.; Deng, K.; Wilkens, L.; Reith, H.; Nielsch, K. Micro-thermoelectric devices. Nat. Electron. 2022, 5, 333–347. [CrossRef]
18. Yan, Q.; Kanatzidis, M.G. High-performance thermoelectrics and challenges for practical devices. Nat. Mater. 2022, 21, 503–513.
[CrossRef]
19. Al-Fartoos, M.M.R.; Roy, A.; Mallick, T.K.; Tahir, A.A. Advancing Thermoelectric Materials: A Comprehensive Review Exploring
the Significance of One-Dimensional Nano Structuring. Nanomaterials 2023, 13, 2011. [CrossRef]
20. Wang, J.; Mu, X.; Sun, M. The Thermal, Electrical and Thermoelectric Properties of Graphene Nanomaterials. Nanomaterials 2019,
9, 218. [CrossRef]
21. Assadi, M.H.N. Na site doping a pathway for enhanced thermoelectric performance in Na1−x CoO2 ; the case of Gd and Yb
dopants. J. Phys. Condens. Matter 2020, 32, 125502. [CrossRef] [PubMed]
22. Assadi, M.H.N.; Mele, P.; Fronzi, M. Suppression of magnetism and Seebeck effect in Na0.875 CoO2 induced by SbCo dopants.
Mater. Renew. Sustain. Energy 2020, 9, 5. [CrossRef]
23. Ovchinnikov, A.A.; Ukrainskii, I.I. Introduction. In Electron-Electron Correlation Effects in Low-Dimensional Conductors and
Superconductors; Springer: Berlin/Heidelberg, Germany, 1991; pp. 1–9.
24. Youn, S.; Kim, J.; Lee, H.; Kim, D.H.; Bang, J.; Lee, W. Strong anisotropic transport properties of quasi-one-dimensional ZrTe3
nanoribbons. Nano Energy 2024, 127, 109771. [CrossRef]
25. Petsagkourakis, I.; Pavlopoulou, E.; Cloutet, E.; Chen, Y.F.; Liu, X.; Fahlman, M.; Berggren, M.; Crispin, X.; Dilhaire, S.; Fleury, G.
Correlating the Seebeck coefficient of thermoelectric polymer thin films to their charge transport mechanism. Org. Electron. 2018,
52, 335–341. [CrossRef]
26. Yang, J. Theory of Thermal Conductivity. In Thermal Conductivity: Theory, Properties, and Applications; Tritt, T.M., Ed.; Springer:
Boston, MA, USA, 2004; pp. 1–20.
27. Maznev, A.; Wright, O. Demystifying umklapp vs normal scattering in lattice thermal conductivity. Am. J. Phys. 2014, 82,
1062–1066. [CrossRef]
28. El Sachat, A.; Alzina, F.; Sotomayor Torres, C.M.; Chavez-Angel, E. Heat Transport Control and Thermal Characterization of
Low-Dimensional Materials: A Review. Nanomaterials 2021, 11, 175. [CrossRef] [PubMed]
29. Suman, P.H.; Jorgetto, A.O.; Romeiro, F.C.; Felix, A.A.; Morais, P.V.; Melquíades, M.O.; Orlandi, M.O. 1D Hybrid Tin Oxide
Nanostructures: Synthesis and Applications. In 1D Semiconducting Hybrid Nanostructures; Wiley-VCH: Weinheim, Germany, 2023;
pp. 97–125.
30. Li, D.; Gong, Y.; Chen, Y.; Lin, J.; Khan, Q.; Zhang, Y.; Li, Y.; Zhang, H.; Xie, H. Recent Progress of Two-Dimensional Thermoelectric
Materials. Nano Micro Lett. 2020, 12, 36. [CrossRef] [PubMed]
31. Kanahashi, K.; Pu, J.; Takenobu, T. 2D Materials for Large-Area Flexible Thermoelectric Devices. Adv. Energy Mater. 2020,
10, 1902842. [CrossRef]
32. Rowe, D.M. Thermoelectrics Handbook: Macro to Nano; CRC Press: Boca Raton, FL, USA, 2018.
33. Hochbaum, A.I.; Chen, R.; Delgado, R.D.; Liang, W.; Garnett, E.C.; Najarian, M.; Majumdar, A.; Yang, P. Enhanced thermoelectric
performance of rough silicon nanowires. Nature 2008, 451, 163–167. [CrossRef]
34. Wang, X.; Wang, H.; Liu, B. Carbon nanotube-based organic thermoelectric materials for energy harvesting. Polymers 2018,
10, 1196. [CrossRef]
35. Borca-Tasciuc, D.-A.; Chen, G.; Prieto, A.; Martín-González, M.; Stacy, A.; Sands, T.; Ryan, M.; Fleurial, J. Thermal properties of
electrodeposited bismuth telluride nanowires embedded in amorphous alumina. Appl. Phys. Lett. 2004, 85, 6001–6003. [CrossRef]
36. Blackburn, J.L.; Ferguson, A.J.; Cho, C.; Grunlan, J.C. Carbon-nanotube-based thermoelectric materials and devices. Adv. Mater.
2018, 30, 1704386. [CrossRef]
37. Heremans, J.P.; Dresselhaus, M.S.; Bell, L.E.; Morelli, D.T. When thermoelectrics reached the nanoscale. Nat. Nanotechnol. 2013, 8,
471–473. [CrossRef] [PubMed]
38. Sanad, M.F.; Shalan, A.E.; Abdellatif, S.O.; Serea, E.S.A.; Adly, M.S.; Ahsan, M.A. Thermoelectric Energy Harvesters: A Review of
Recent Developments in Materials and Devices for Different Potential Applications. Top. Curr. Chem. 2020, 378, 1–43. [CrossRef]
39. Assadi, M.H.N.; Katayama-Yoshida, H. Native point defects in Ti3 GeC2 and Ti2 GeC. Comput. Mater. Sci. 2017, 128, 103–108.
[CrossRef]
40. Hicks, L.; Dresselhaus, M.S. Thermoelectric figure of merit of a one-dimensional conductor. Phys. Rev. B 1993, 47, 16631–16634.
[CrossRef] [PubMed]
Nanomaterials 2024, 14, 1272 16 of 18

41. Hicks, L.; Dresselhaus, M.S. Effect of quantum-well structures on the thermoelectric figure of merit. Phys. Rev. B 1993, 47,
12727–12731. [CrossRef]
42. Nomura, M.; Maire, J. Mechanism of the Reduced Thermal Conductivity of Fishbone-Type Si Phononic Crystal Nanostructures. J.
Electron. Mater. 2015, 44, 1426–1431. [CrossRef]
43. Zhang, T.; Wu, S.; Xu, J.; Zheng, R.; Cheng, G. High thermoelectric figure-of-merits from large-area porous silicon nanowire
arrays. Nano Energy 2015, 13, 433–441. [CrossRef]
44. Chen, J.; Zhang, G.; Li, B. Remarkable reduction of thermal conductivity in silicon nanotubes. Nano Lett. 2010, 10, 3978–3983.
[CrossRef]
45. Tang, J.; Wang, H.-T.; Lee, D.H.; Fardy, M.; Huo, Z.; Russell, T.P.; Yang, P. Holey silicon as an efficient thermoelectric material.
Nano Lett. 2010, 10, 4279–4283. [CrossRef] [PubMed]
46. Elyamny, S.; Dimaggio, E.; Magagna, S.; Narducci, D.; Pennelli, G. High power thermoelectric generator based on vertical silicon
nanowires. Nano Lett. 2020, 20, 4748–4753. [CrossRef] [PubMed]
47. Ding, C.; Lu, T.; Wazir, N.; Ma, W.; Guo, S.; Xin, Y.; Li, A.; Liu, R.; Zou, B. New Type of Thermoelectric CdSSe Nanowire Chip.
ACS Appl. Mater. Interfaces 2021, 13, 30959–30966. [CrossRef] [PubMed]
48. Neophytou, N.; Wagner, M.; Kosina, H.; Selberherr, S. Analysis of thermoelectric properties of scaled silicon nanowires using an
atomistic tight-binding model. J. Electron. Mater. 2010, 39, 1902–1908. [CrossRef]
49. Sansoz, F. Surface faceting dependence of thermal transport in silicon nanowires. Nano Lett. 2011, 11, 5378–5382. [CrossRef]
50. Stillinger, F.H.; Weber, T.A. Computer simulation of local order in condensed phases of silicon. Phys. Rev. B 1985, 31, 5262–5271.
[CrossRef]
51. Jean, V.; Fumeron, S.; Termentzidis, K.; Zianni, X.; Lacroix, D. Monte Carlo simulations of phonon transport in Si nanowires with
constrictions. Int. J. Heat Mass Transf. 2015, 86, 648–655. [CrossRef]
52. Lin, Y.-M.; Sun, X.; Dresselhaus, M. Theoretical investigation of thermoelectric transport properties of cylindrical Bi nanowires.
Phys. Rev. B 2000, 62, 4610–4623. [CrossRef]
53. Zhou, G.; Li, L.; Li, G. Enhancement of thermoelectric figure of merit in bismuth nanotubes. Appl. Phys. Lett. 2010, 97, 023112.
[CrossRef]
54. Kubo, R. Statistical-Mechanical Theory of Irreversible Processes. I. General Theory and Simple Applications to Magnetic and
Conduction Problems. J. Phys. Soc. Jpn. 1957, 12, 570–586. [CrossRef]
55. Green, M.S. Markoff Random Processes and the Statistical Mechanics of Time-Dependent Phenomena. II. Irreversible Processes in
Fluids. J. Chem. Phys. 1954, 22, 398–413. [CrossRef]
56. Demchenko, D.O.; Heinz, P.D.; Lee, B. Determining factors of thermoelectric properties of semiconductor nanowires. Nanoscale
Res. Lett. 2011, 6, 502. [CrossRef] [PubMed]
57. Perdew, J.P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865–3868.
[CrossRef]
58. Shi, L.; Yao, D.; Zhang, G.; Li, B. Large thermoelectric figure of merit in Si1−x Gex nanowires. Appl. Phys. Lett. 2010, 96, 173108.
[CrossRef]
59. Lee, K.-M.; Lee, S.-K.; Choi, T.-Y. Highly enhanced thermoelectric figure of merit of a β-SiC nanowire with a nanoelectromechanical
measurement approach. Appl. Phys. A 2012, 106, 955–960. [CrossRef]
60. Gogoc, S.; Data, P. Organic Thermoelectric Materials as the Waste Heat Remedy. Molecules 2022, 27, 1016. [CrossRef] [PubMed]
61. Taborowska, P.; Janas, D. Seamless design of thermoelectric modules from single-walled carbon nanotubes. J. Mater. Chem. C
2022, 10, 6818–6826. [CrossRef]
62. Rao, A.M.; Ji, X.; Tritt, T.M. Properties of nanostructured one-dimensional and composite thermoelectric materials. MRS Bull.
2006, 31, 218–223. [CrossRef]
63. Asadikia, A.; Mirjalily, S.A.A.; Nasirizadeh, N.; Kargarsharifabad, H. Hybrid nanofluid based on CuO nanoparticles and
single-walled Carbon nanotubes: Optimization, thermal, and electrical properties. Int. J. Nano Dimens. 2020, 11, 277–289.
64. Jiang, P.; Liu, H.; Fan, D.; Cheng, L.; Wei, J.; Zhang, J.; Liang, J.; Shi, J. Enhanced thermoelectric performance of carbon nanotubes
at elevated temperature. Phys. Chem. Chem. Phys. 2015, 17, 27558–27564. [CrossRef]
65. Hung, N.T.; Nugraha, A.R.; Hasdeo, E.H.; Dresselhaus, M.S.; Saito, R. Diameter dependence of thermoelectric power of
semiconducting carbon nanotubes. Phys. Rev. B 2015, 92, 165426. [CrossRef]
66. Taborowska, P.; Wasiak, T.; Sahlman, M.; Lundström, M.; Janas, D. Carbon Nanotube-Based Thermoelectric Modules Enhanced
by ZnO Nanowires. Materials 2022, 15, 1924. [CrossRef] [PubMed]
67. Zhou, W.; Fan, Q.; Zhang, Q.; Cai, L.; Li, K.; Gu, X.; Yang, F.; Zhang, N.; Wang, Y.; Liu, H. High-performance and compact-designed
flexible thermoelectric modules enabled by a reticulate carbon nanotube architecture. Nat. Commun. 2017, 8, 14886. [CrossRef]
[PubMed]
68. Li, Y.; Qiao, J.; Zhao, Y.; Lan, Q.; Mao, P.; Qiu, J.; Tai, K.; Liu, C.; Cheng, H. A flexible thermoelectric device based on a
Bi2 Te3 -carbon nanotube hybrid. J. Mater. Sci. Technol. 2020, 58, 80–85. [CrossRef]
69. Kumanek, B.; Janas, D. Thermal conductivity of carbon nanotube networks: A review. J. Mater. Sci. 2019, 54, 7397–7427. [CrossRef]
70. Marconnet, A.M.; Panzer, M.A.; Goodson, K.E. Thermal conduction phenomena in carbon nanotubes and related nanostructured
materials. Rev. Mod. Phys. 2013, 85, 1295–1326. [CrossRef]
Nanomaterials 2024, 14, 1272 17 of 18

71. Pop, E.; Mann, D.; Wang, Q.; Goodson, K.; Dai, H. Thermal conductance of an individual single-wall carbon nanotube above
room temperature. Nano Lett. 2006, 6, 96–100. [CrossRef] [PubMed]
72. Balandin, A.A. Thermal properties of graphene and nanostructured carbon materials. Nat. Mater. 2011, 10, 569–581. [CrossRef]
[PubMed]
73. Hone, J.; Llaguno, M.; Biercuk, M.; Johnson, A.; Batlogg, B.; Benes, Z.; Fischer, J. Thermal properties of carbon nanotubes and
nanotube-based materials. Appl. Phys. A 2002, 74, 339–343. [CrossRef]
74. Han, Z.; Fina, A. Thermal conductivity of carbon nanotubes and their polymer nanocomposites: A review. Prog. Polym. Sci. 2011,
36, 914–944. [CrossRef]
75. Cao, J.X.; Yan, X.H.; Xiao, Y.; Ding, J.W. Thermal conductivity of zigzag single-walled carbon nanotubes: Role of the umklapp
process. Phys. Rev. B 2004, 69, 073407. [CrossRef]
76. Evans, W.J.; Shen, M.; Keblinski, P. Inter-tube thermal conductance in carbon nanotubes arrays and bundles: Effects of contact
area and pressure. Appl. Phys. Lett. 2012, 100, 261908. [CrossRef]
77. Stuart, S.J.; Tutein, A.B.; Harrison, J.A. A reactive potential for hydrocarbons with intermolecular interactions. J. Chem. Phys. 2000,
112, 6472–6486. [CrossRef]
78. Boroushak, S.; Ansari, R.; Ajori, S. Molecular dynamics simulations of the thermal conductivity of cross-linked functionalized
single-and double-walled carbon nanotubes with polyethylene chains. Diam. Relat. Mater. 2018, 86, 173–178. [CrossRef]
79. Chalopin, Y.; Volz, S.; Mingo, N. Upper bound to the thermal conductivity of carbon nanotube pellets. J. Appl. Phys. 2009,
105, 084301. [CrossRef]
80. Aliev, A.E.; Guthy, C.; Zhang, M.; Fang, S.; Zakhidov, A.A.; Fischer, J.E.; Baughman, R.H. Thermal transport in MWCNT sheets
and yarns. Carbon 2007, 45, 2880–2888. [CrossRef]
81. Berber, S.; Kwon, Y.-K.; Tománek, D. Unusually high thermal conductivity of carbon nanotubes. Phys. Rev. Lett. 2000, 84,
4613–4616. [CrossRef]
82. Speranza, G. Carbon nanomaterials: Synthesis, functionalization and sensing applications. Nanomaterials 2021, 11, 967. [CrossRef]
[PubMed]
83. Voznyakovskii, A.; Neverovskaya, A.; Vozniakovskii, A.; Kidalov, S. A Quantitative chemical method for determining the surface
concentration of stone–Wales defects for 1D and 2D carbon nanomaterials. Nanomaterials 2022, 12, 883. [CrossRef]
84. Batra, R.; Song, L.; Ramprasad, R. Emerging materials intelligence ecosystems propelled by machine learning. Nat. Rev. Mater.
2021, 6, 655–678. [CrossRef]
85. Hart, G.L.W.; Mueller, T.; Toher, C.; Curtarolo, S. Machine learning for alloys. Nat. Rev. Mater. 2021, 6, 730–755. [CrossRef]
86. Choudhary, K.; DeCost, B.; Chen, C.; Jain, A.; Tavazza, F.; Cohn, R.; Park, C.W.; Choudhary, A.; Agrawal, A.; Billinge, S.J.L.; et al.
Recent advances and applications of deep learning methods in materials science. NPJ Comput. Mater. 2022, 8, 59. [CrossRef]
87. Atwal, G.S.; Ashcroft, N.W. Relaxation of an electron system: Conserving approximation. Phys. Rev. B 2002, 65, 115109. [CrossRef]
88. Hirosawa, T.; Schäfer, F.; Maebashi, H.; Matsuura, H.; Ogata, M. Data-Driven Reconstruction of Spectral Conductivity and
Chemical Potential Using Thermoelectric Transport Properties. J. Phys. Soc. Jpn. 2022, 91, 114603. [CrossRef]
89. Kimura, K.; Iwasaki, Y. Solving Long-standing Problems in Thermoelectric Properties using Machine Learning. JPSJ News
Comments 2023, 20, 01. [CrossRef]
90. Unke, O.T.; Chmiela, S.; Sauceda, H.E.; Gastegger, M.; Poltavsky, I.; Schütt, K.T.; Tkatchenko, A.; Müller, K.-R. Machine Learning
Force Fields. Chem. Rev. 2021, 121, 10142–10186. [CrossRef] [PubMed]
91. Poltavsky, I.; Tkatchenko, A. Machine Learning Force Fields: Recent Advances and Remaining Challenges. J. Phys. Chem. Lett.
2021, 12, 6551–6564. [CrossRef] [PubMed]
92. Sauceda, H.E.; Gálvez-González, L.E.; Chmiela, S.; Paz-Borbón, L.O.; Müller, K.-R.; Tkatchenko, A. BIGDML—Towards accurate
quantum machine learning force fields for materials. Nat. Commun. 2022, 13, 3733. [CrossRef] [PubMed]
93. Qin, G.; Wei, Y.; Yu, L.; Xu, J.; Ojih, J.; Rodriguez, A.D.; Wang, H.; Qin, Z.; Hu, M. Predicting lattice thermal conductivity from
fundamental material properties using machine learning techniques. J. Mater. Chem. A 2023, 11, 5801–5810. [CrossRef]
94. Cui, C.; Zhang, Y.; Ouyang, T.; Chen, M.; Tang, C.; Chen, Q.; He, C.; Li, J.; Zhong, J. On-the-fly machine learning potential
accelerated accurate prediction of lattice thermal conductivity of metastable silicon crystals. Phys. Rev. Mater. 2023, 7, 033803.
[CrossRef]
95. Jain, A.; Ong, S.P.; Hautier, G.; Chen, W.; Richards, W.D.; Dacek, S.; Cholia, S.; Gunter, D.; Skinner, D.; Ceder, G.; et al. Commentary:
The Materials Project: A materials genome approach to accelerating materials innovation. APL Mater. 2013, 1, 011002. [CrossRef]
96. Xu, Y.; Yamazaki, M.; Villars, P. Inorganic Materials Database for Exploring the Nature of Material. Jpn. J. Appl. Phys. 2011,
50, 11RH02. [CrossRef]
97. Zhou, J.; Shen, L.; Costa, M.D.; Persson, K.A.; Ong, S.P.; Huck, P.; Lu, Y.; Ma, X.; Chen, Y.; Tang, H.; et al. 2DMatPedia, an open
computational database of two-dimensional materials from top-down and bottom-up approaches. Sci. Data 2019, 6, 86. [CrossRef]
[PubMed]
98. Wei, Q.; Xiong, F.; Tan, S.; Huang, L.; Lan, E.H.; Dunn, B.; Mai, L. Porous One-Dimensional Nanomaterials: Design, Fabrication
and Applications in Electrochemical Energy Storage. Adv. Mater. 2017, 29, 1602300. [CrossRef] [PubMed]
99. Jin, T.; Han, Q.; Wang, Y.; Jiao, L. 1D Nanomaterials: Design, Synthesis, and Applications in Sodium–Ion Batteries. Small 2018,
14, 1703086. [CrossRef] [PubMed]
Nanomaterials 2024, 14, 1272 18 of 18

100. Duong, H.M.; Einarsson, E.; Okawa, J.; Xiang, R.; Maruyama, S. Thermal degradation of single-walled carbon nanotubes. Jpn. J.
Appl. Phys. 2008, 47, 1994–1999. [CrossRef]
101. Mahajan, A.; Kingon, A.; Kukovecz, Á.; Konya, Z.; Vilarinho, P.M. Studies on the thermal decomposition of multiwall carbon
nanotubes under different atmospheres. Mater. Lett. 2013, 90, 165–168. [CrossRef]
102. Yee, S.K.; LeBlanc, S.; Goodson, K.E.; Dames, C. $ per W metrics for thermoelectric power generation: Beyond ZT. Energy Environ.
Sci. 2013, 6, 2561–2571. [CrossRef]
103. Kim, S.I.; Lee, K.H.; Mun, H.A.; Kim, H.S.; Hwang, S.W.; Roh, J.W.; Yang, D.J.; Shin, W.H.; Li, X.S.; Lee, Y.H.; et al. Dense dislocation
arrays embedded in grain boundaries for high-performance bulk thermoelectrics. Science 2015, 348, 109–114. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.