Surfaces and Interfaces 44 (2024) 103829

Contents lists available at ScienceDirect

Surfaces and Interfaces

journal homepage: www.sciencedirect.com/journal/surfaces-and-interfaces

Janus 2D B2 P6 : A promising anisotropic thermoelectric material with high

power factor
Vuong Van Thanh a ,∗, Do Van Truong a , Nguyen Tuan Hung b,c ,∗
a School of Mechanical Engineering, Hanoi University of Science and Technology, Hanoi 100000, Viet Nam
b
Frontier Research Institute for Interdisciplinary Sciences, Tohoku University, Sendai 980-8578, Japan
c
Department of Physics, Tohoku University, Sendai 980-8578, Japan

ARTICLE INFO ABSTRACT

Keywords: A two-dimensional (2D) material with high thermoelectric (TE) performance is of great interest for the energy
2D Janus materials conversion device. In this study, by using first-principles calculations, we show that the Janus 2D B2 P6 has
Density functional theory an extremely high TE power factor of 0.10 and 0.07 W/mK2 for p-type and n-type at room temperature,
The Boltzmann transport theory
respectively, which are higher than that of conventional 2D TE materials. We also investigate lattice thermal
Thermoelectricity
conductivity, 𝜅𝑝ℎ , of B2 P6 by using the Boltzmann transport theory with phonon–phonon interaction. With
Ideal strength
𝜅𝑝ℎ of 41.45 and 133.3 W/mK for the armchair and zigzag directions, respectively, the optimized values of
the TE figure of merit are up to 0.7 at 900 K. Moreover, the Janus 2D B2 P6 exhibits anisotropic mechanical
properties, in which the ideal strength of B2 P6 is 10.7 and 18.63 N/m for the armchair and zigzag directions,
respectively. Thus, our results suggest that the Janus 2D B2 P6 is the potentially promising flexible anisotropic
TE material.

1. Introduction materials and high carrier mobility, i.e., they have large 𝑆 while
remaining high 𝜎 for a specific direction. Based on the electric-double-
The global environment demands sustainable and green energy, layer transistor (EDLT) configuration, Saito et al. [20] found that 𝑆 of
in which new energy conversion techniques are needed to reduce the 2D BP material has a high value of up to 510 μV∕K at 𝑇 = 210 K,
dependence on fossil energy [1,2]. Among them, thermoelectric (TE) which is much larger than the reported experimental bulk BP value of
technology, which can directly convert heat into electricity, is promis- 335 ± 10 μV∕K [21]. Liu et al. [22] measured the electrical properties
ing for a sustainable energy solution in the future [3,4]. In a TE device, of the 2D BP with a thickness of about 5 nm and indicated that it is
there are two parameters to evaluate the energy conversion perfor- a p-type semiconductor with a high hole mobility (286 cm2 V−1 s−1 )
mance. The first parameter is a figure of merit 𝑍𝑇 = 𝑆 2 𝜎𝑇 ∕(𝜅𝑒𝑙 + 𝜅𝑝ℎ ), at 300 K. In another experiment, Li et al. [15] indicated that the
where 𝑆 is the Seebeck coefficient, 𝜎 is the electrical conductivity, 𝜅𝑒𝑙 field-effect mobility of the 2D BP with a thickness of about 10 nm
and 𝜅𝑝ℎ are the thermal conductivities by electron and lattice, respec- up to ∼ 1000 cm2 V−1 s−1 at room temperature, which is comparable
tively, and 𝑇 is the temperature [5]. The second parameter is the TE with silicon. In addition, Lv et al. [23] showed that the PF of BP can
power factor, PF = 𝑆 2 𝜎. 𝑍𝑇 is important for TE applications where the
achieve up to 13.89 mW/mK2 . On the other hand, Liao et al. [24] also
heat source is limited, while PF is important for other TE applications
showed that the 2D BP material has significantly anisotropic thermal
when the heat source is unlimited, like solar energy [6,7]. Therefore,
and electrical conductivity, resulting in higher 𝑍𝑇 in the armchair
finding materials with high PF is as important as finding materials with
direction than in the zigzag direction. However, the 2D BP is unstable in
high 𝑍𝑇 , especially for power generation applications, in which the
the air or high-temperature environment [25], and the TE device thus
output power density is more critical than the efficiency [8,9].
can quickly degrade. Furthermore, the actual PF value is not as high as
There are several methods to enhance the PF in the material,
including the quantum confinement effect in low-dimensional semicon- expected and needs further improvement. Therefore, in this study, we
ductors [6,10–12], band convergence [13,14], and using materials with evaluate the TE properties of the B2 P6 monolayer, a new stable material
high intrinsic carrier mobility as such anisotropic 2D black phosphorus in the BP family, to optimize the PF of anisotropic TE materials.
(BP), 2D InSe, etc. [15–19]. Among them, the anisotropic 2D material is The Janus 2D B2 P6 is recently theoretically predicted by Sun et al.
interesting for TE materials with high PF because of the low-dimension [26] using the evolutionary search and first-principles calculations [26].

∗ Corresponding authors.
E-mail addresses: thanh.vuongvan@hust.edu.vn (V.V. Thanh), nguyen.tuan.hung.e4@tohoku.ac.jp (N.T. Hung).

https://doi.org/10.1016/j.surfin.2023.103829
Received 13 November 2023; Received in revised form 22 December 2023; Accepted 26 December 2023
Available online 28 December 2023
2468-0230/© 2023 Elsevier B.V. All rights reserved.
V.V. Thanh et al. Surfaces and Interfaces 44 (2024) 103829

Fig. 1. Atomic structure of the Janus 2D B2 P6 with the top (a) and side (b) views, in which the back box indicates the unit cell. (c) Total energy per unit cell and temperature
as a function of time at 300 K. The inset figure is the atomic structure at four ps.

They demonstrated that the cohesive energy of B2 P6 is 4.36 eV/atom, 𝑧-axis. A 𝐤-points grid of 9 × 15 × 1 is used for electronic structure
which is less than that of monolayer MoS2 (5.02 eV/atom) and graphene calculations. The phonon dispersion relation is computed based on
(7.85 eV/atom), but greater than that of BP (3.48 eV/atom). The higher density-functional perturbation theory (DFPT) [38] with a 𝐪-points grid
cohesive energy indicates more stability and feasibility of experimental of 8 × 8 × 1. The 𝑎𝑏 𝑖𝑛𝑖𝑡𝑖𝑜 molecular dynamics (AIMD) simulations
synthesis. The B2 P6 material is anisotropic like BP and has other are calculated with the NVT ensemble to check the thermal stabil-
favorable TE properties, such as high carrier mobility with the Janus ity, in which the temperature is controlled using the Nose–Hoover
structure. In a previous theoretical study based on the first-principles method [39] with a time step of 1 fs. The BoltzWann code [40] is
calculations, Ren et al. [27] found that B2 P6 has an ultrahigh electron employed for calculating the Seebeck coefficient, electrical conduc-
mobility of ∼5888 cm2 V−1 s−1 , which is much higher than those other tivity, and electronic thermal conductivity, in which the deformation
2D materials, such as Janus Ge2 SSe (786.94 cm2 V−1 s−1 ) [28], BP potential approximation estimates the relaxation time. The ShengBTE
(1100–1140 cm2 V−1 s−1 ) [29], and MoS2 (∼100 cm2 V−1 s−1 ) [30]. package [41] is used to determine the lattice thermal conductivity.
Furthermore, due to the asymmetric Janus structure, the electrical and In this evaluation, the third-order interatomic force constant matrix is
thermal properties of Janus materials could be modulated. For example, computed with a supercell of 4 × 4 × 1, considering interactions up
the lattice thermal conductivity of Janus SnSSe monolayer is lower to the sixth-nearest neighbors as determined by a convergence test.
than those of SnS2 and SnSe2 at the same temperature, originating The phonon Boltzmann transport equation is solved on a grid with
from the breakdown of inversion symmetry, making it a promising TE dimensions of 48 × 48 × 1. The Bader charge analysis is calculated
material [31]. In a recent theoretical study [28], we also discovered using the Bader code [42] to obtain the charge transfer.
that 𝛾− Ge2 SSe with a Janus structure exhibits thermal conductivity as
low as 3.33 W/mK at 300 K. Consequently, this leads to a favorable 𝑍𝑇 3. Results and discussion
value ranging from 0.7 to 0.9 for the temperature range from 300 K to
900 K. Therefore, investigating the TE properties of B2 P6 is essential to 3.1. Structure and stability
open up applications for this new material and, based on that, is the
motivation for experimental research synthesizes it. In Fig. 1, we show the optimized structure of the Janus 2D B2 P6 ,
In this paper, we systematically investigate the TE properties of where the unit cell (black box) contains two B and six P atoms with
Janus anisotropic B2 P6 monolayer by using first-principles calculations sp3 hybridization. At the equilibrium state, the lattice constants are 𝑎
and Boltzmann transport theory. We find that Janus 2D B2 P6 exhibits = 5.456 Å and 𝑏 = 3.254 Å along the 𝑥 (armchair) and 𝑦 (zigzag)
strong anisotropy mechanical, carrier mobility, and TE properties. The directions, respectively, which are very close to the previously reported
p-type Janus 2D B2 P6 has extremely high PF up to 0.05 and 0.12 theoretical values [26,27]. The buckling height 𝑏0 in the 𝑧 direction is
W/mK2 and a 𝑍𝑇 value of 0.55 and 0.7 at 900 K for the armchair 4.77 Å, and the P1 -P2 and B1 -P1 bond lengths are 𝑙1 = 2.26 Å and 𝑙2 =
and zigzag directions, respectively. 1.95 Å, respectively, as illustrated in Fig. 1(b). The P1 -P2 bond length
of the Janus B2 P6 is comparable to those in the 2D BP (2.24 Å) [43]
2. Methodology and the Janus SiP2 (2.27 Å) [44]. The effective layer thickness 𝑑0 of
the Janus 2D B2 P6 is calculated to be 7.878 Å by considering the Van
The calculations in this study are carried out through the Quantum der Waals interaction. Here, 𝑑0 is defined as 𝑐∕2, where 𝑐 represents the
ESPRESSO package [32,33] using density functional theory (DFT). The lattice constant of bulk B2 P6 along the 𝑧 direction (refer to Fig. S1 in
valence electron-ion core interactions use the projector augmented the Supporting Information). Here, we use the vdW-DF2 functional [45]
wave potentials (PAW) [34]. The generalized gradient approximation for the Van der Waals interaction to optimize the structure of the bulk
(GGA) with the Perdew–Burke–Ernzerhof (PBE) [35] is employed to B2 P6 material.
calculate the exchange–correlation interactions. The Heyd–Scuseria– Furthermore, the Bader charge-based ionic charge states of Janus
Ernzerhof (HSE) functional [36] is employed for band gap corrections, B2 P6 are B2 0.444+ , B1 0.457+ , P1 0.396− , P2 0.001− , P3 0.0249− , P4 0.357− , P5 0.053− ,
and the band dispersion based on the HSE calculation is interpolated and P6 0.069− , which indicate that the middle layer with P1 -B1 , B1 -P4 ,
using the Wannier90 package [37]. The cutoff energy was set to 60 and P4 -B2 bonds are the metavalent bonding, while the bottom and
Ry for the plane wave basis. The Broyden-Fretcher-Goldfarb–Shanno top layers with P2 -P3 and P5 -P6 bonds, respectively, are the covalent
algorithm [33] is utilized to optimize the structure, with convergence bonding. The metavalent bonding often leads to soft chemical bonds
criteria set at 1 × 10−5 Ry/a.u. for atomic forces and 5 × 10−2 GPa and introduces a lower phonon group velocity and strong anharmonic.
for all stress components. A significant vacuum space of about 22 Å is Thus, the middle layer might contribute to reducing lattice thermal
inserted to resist the interactions between the periodic layers along the conductivity of B2 P6 . On the other hand, Yu et al. [46] showed that the

2
V.V. Thanh et al. Surfaces and Interfaces 44 (2024) 103829

metavalent bonding in GeSe alloy with 15% AgSbSe2 could increase

carrier mobility, band anisotropy and valley degeneracy, which are
favorable for the high PF. Thus, we expect that the metavalent bonding
will also contribute to obtaining a high PF of B2 P6 .
We next examine the dynamical, thermal, and mechanical stability
of the Janus 2D B2 P6 before investigating its TE properties. First, we
calculate the phonon dispersion to examine the dynamical stability,
as shown in Fig. 6(b). The results show that all phonon frequencies
are positive, suggesting that the Janus B2 P6 is dynamically stable. The
highest phonon frequency is 800 cm−1 , which is higher than that of
the 2D BP (540 cm−1 ) [47]. Second, we perform the ab-initio molecular-
dynamics (AIMD) calculations at 300 K and 900 K to check the thermal
stability using a supercell of 3 × 3 × 1. We perform a simulation for a
duration of 4 ps, employing a time step of 1 fs in the NVT ensemble.
The calculated results indicate that the total energy oscillates around
a constant value, and the atomic configuration of B2 P6 remains nearly
unchanged at 300 K, as illustrated in Fig. 1(c) (calculations at 900 K can
be found in Fig. S2 in the SI). Thus, B2 P6 shows the thermal stability
for a temperature range from 300 K to 900 K, which is suitable for the
TE application at high temperatures. Finally, the Born stability criteria
is used for the mechanical stability. The calculated elastic constants
𝐶11 = 115.43 (N/m), |𝐶12 | = 25.45 (N/m) and 𝐶66 = 61.18 (N/m). These
values meet the Born criterion for mechanical stability (𝐶11 > |𝐶12 | > 0
and 𝐶66 > 0) [48].

3.2. Anisotropic mechanical properties

Since the anisotropic behavior of a material can be directly ob- Fig. 2. (a) and (b) are polar diagrams for the angular dependence of Young’s modulus
served by the mechanical properties [49,50], we will investigate the (𝑌2D ) and Poisson’s ratio (𝜐) of the Janus 2D B2 P6 , respectively. (c) Stress is plotted as
mechanical properties of the Janus 2D B2 P6 before the anisotropic TE a function of tensile strain.

properties. For the 2D materials, the angular-dependence of Young’s

modulus 𝑌2D and Poisson’s ratio 𝜐 are defined by [28,51]
2
as 𝜀 = (𝑙𝑖 − 𝑙𝑖0 )∕𝑙𝑖0 (𝑖 = 𝑥, 𝑦), where 𝑙𝑖 and 𝑙𝑖0 are the strained and
𝐶11 𝐶22 − 𝐶12 unstrained lattice constants for the 𝑥- or 𝑦-directions, respectively. In
𝑌2D (𝜃) = ( 2
) , (1)
𝐶11 𝐶22 − 𝐶12 Fig. 2(b), we show the calculated stress–strain curves of B2 P6 . We found
𝐶11 𝛼 4 + 𝐶22 𝛽 4 − 2𝐶12 − 𝛼2 𝛽 2 that the ideal strength in the 𝑦 direction (18.63 N/m) is much higher
𝐶66
than that in the 𝑥 direction (10.7 N/m), indicating the anisotropic
and mechanical properties of B2 P6 . The ideal strength of B2 P6 in the 𝑦-
( 2
)
𝐶11 𝐶22 − 𝐶12 direction is smaller than that of graphene (37.87 N/m), penta-graphene
𝐶12 (𝛼 4 + 𝛽4) − 𝐶11 + 𝐶22 − 𝛼2 𝛽 2 (26.94 N/m), and hydrogenated penta-graphene (33.88 N/m) [52], and
𝐶66
𝜐(𝜃) = ( 2
) , (2) larger than that of striped-borophen in 𝑦 direction (12.11 N/m) [57]
𝐶11 𝐶22 − 𝐶12 and BP (9.99 N/m) [58]. Furthermore, the ideal strain of B2 P6 in
𝐶11 𝛼4 + 𝐶22 𝛽4 − 2𝐶12 − 𝛼2 𝛽 2
𝐶66 the 𝑦 direction has a value as large as 28%, while the ideal strain in
the 𝑥 direction is 15%. This result shows that B2 P6 has potential for
where 𝛼 = sin(𝜃) and 𝛽 = cos(𝜃), where 𝜃 is the angle relative to the
application in flexible TE devices.
𝑥-axis.
The calculated results show that 𝑌2D of B2 P6 varies from 112.15
3.3. Electronic properties and carrier mobility
to 191.94 N/m, as shown in Fig. 2(a). 𝑌2D of B2 P6 in the 𝑦-direction
(191.94 N/m) is smaller than that of graphene (325.05 N/m) [52] and
In Fig. 3(a), we show the energy band structures of the Janus 2D
larger than that of the 2D BP (56.3 N/m) [53] and MoS2 monolayer
B2 P6 based on the HSE method. Our obtained results indicate that B2 P6
(120 N/m) [54]. As shown in Fig. 2(a), 𝜐 of B2 P6 is 0.129 and 0.22 for
is an indirect semiconducting, in which the valence band maximum
𝜃 = 0◦ and 90◦ , respectively. To evaluate the mechanical anisotropy,
𝑌2D (VBM) and the conduction band minimum (CBM) positions located at
we define an anisotropic ratio, 𝑅ani , as the following equation [55]:
the 𝛤 and X points, respectively. The calculated band gap is 1.94 eV,
𝑥 ,𝑌𝑦 )
max(𝑌2D which agrees with the previous theoretical reports (2.09 eV) [26,27].
𝑌
2D 2D
𝑅ani = 𝑥 ,𝑌𝑦 )
, (3) The band gap of the Janus B2 P6 can be compared with that of the 2D
min(𝑌2D 2D
BP (∼ 2 eV) [59]. We note that the calculated band structure of Janus
where, 𝑌2D 𝑥 and 𝑌 𝑦 are the Young’s modulus along the 𝑥- and 𝑦-
2D B2 P6 by the GGA-PBE is shown in Fig. S3 in the SI, in which the band
𝑌2D
directions, respectively. Here, we find that 𝑅ani = 1.7, indicating strong gap is about 1.23 eV. To examine the contribution of each atom to the
mechanical anisotropy. It is noted that for the isotropic materials, formation of band structures, we calculated the partial density of states
𝑌2D 𝑌2D
𝑅ani = 1. Compared with other anisotropic 2D materials, 𝑅ani of the (PDOS), as shown in Fig. 3(b). We can see that the P atoms (red line)
Janus 2D B2 P6 is larger than that of the Janus 2D SiP2 (1.31) [44], and are the main contribution to the total PDOS (black line) compared with
smaller than that of the 2D BP (2.57) [56], and the striped borophene the B atoms (blue line).
(2.43) [57]. We next calculate the anisotropic carrier mobility in the Janus 2D
Furthermore, to calculate the ideal strength and ideal strain of Janus B2 P6 . We noted that Ren et al. [27] determined the carrier mobility
2D B2 P6 , the tensile strain is applied on the rectangular unit cell along of the Janus 2D B2 P6 by using the isotropic deformation potential
the 𝑥- and 𝑦- directions, respectively. Here, the tensile strain is defined approximation (DPA). However, as discussed in Section 3.2, the Janus

3
V.V. Thanh et al. Surfaces and Interfaces 44 (2024) 103829

Fig. 3. (a) Electronic energy dispersion with atomic orbital projections in color and (b) Projected density of states (PDOS) of the Janus 2D B2 P6 .

Table 1 (𝑇 = 300 K) are listed in Table 1. In particular, we find that the

Calculated effective masses 𝑚∗ ∕𝑚0 , where 𝑚0 is the mass of free electron, elastic mod-
electron mobility along the 𝑥 direction is 489 cm2 V−1 s−1 , which is
ulus 𝐶11(22) (N/m), deformation potential 𝐸𝑥(𝑦) (eV), carrier mobilities 𝜇2D (cm2 V−1 s−1 ),
and relaxation time 𝜏 𝑒𝑙 (s) of Janus 2D B2 P6 at 300 K.
smaller than that using the isotropic DPA (5888 cm2 V−1 s−1 ) [27].
Our calculated results demonstrated that the Janus 2D B2 P6 exhibits
Carrier type 𝑚∗ /𝑚0 𝐶11(22) 𝐸𝑥(𝑦) 𝜇2D 𝜏 𝑒𝑙
a significant electron mobility (1386 cm2 V−1 s−1 along the 𝑦 direction)
Electron (x) 1.230 115.43 0.977 489 3.423×10−13
Hole (x) −0.844 115.43 −2.948 740 3.550×10−13
due to its low electron effective mass. The electron mobility of the
Electron (y) 0.198 197.55 −7.996 1386 1.560×10−13 Janus 2D B2 P6 along the 𝑦 direction is higher than those of the 2D
Hole (y) −3.677 197.55 1.457 461 9.629×10−13 BP (∼1000 cm2 V−1 s−1 ) [15], MoS2 (∼100 cm2 V−1 s−1 ) [30], 𝛾−GeSSe
(940.45 cm2 V−1 s−1 ) [28], and WSSe (276.64 cm2 V−1 s−1 ) [61]. The
high value of carrier mobility implies that the Janus 2D B2 P6 could
be a promising material for TE applications.
B2 P6 is an anisotropic 2D material. Thus, the isotropic DPA might not
To evaluate the carrier mobility’s anisotropy, we calculate an
be suitable. A modified DPA for the anisotropic 2D materials has been
anisotropic ratio, 𝑅𝜇ani , as [55]
introduced by Lang et al. [55], and they showed that the carrier mo-
bility is significantly changed. For example, the 2D BP has the carrier max(𝜇𝑥 , 𝜇𝑦 )
𝑅𝜇ani = . (6)
mobilizes (𝜇𝑥 = 1120; 𝜇𝑦 = 80 cm2 V−1 s−1 for the electron and 𝜇𝑥 = 670; min(𝜇𝑥 , 𝜇𝑦 )
𝜇𝑦 = 16000 cm2 V−1 s−1 for the hole) and (𝜇𝑥 = 800; 𝜇𝑦 = 400 cm2 V−1 s−1
The calculated results show that 𝑅𝜇ani for the electron and hole mo-
for the electron and 𝜇𝑥 = 2770; 𝜇𝑦 = 180 cm2 V−1 s−1 for the hole)
based on the isotropic and anisotropic DPA, respectively. On the other bilities are 2.83 and 1.60, respectively, which is larger than 1.0 of
hand, the field-effect transistors have observed the carrier mobility of an isotropic system. The anisotropy electron mobility of the Janus 2D
the black-phosphorene (FETs) about 1350-2700 cm2 V−1 s−1 [60]. Thus, B2 P6 is smaller than that of the 2D BP (𝑅𝜇ani = 15.1) [55], and higher
the isotropic DPA shows the overestimated values for carrier mobility. than that of single-layered Hittorf’s phosphorus (𝑅𝜇ani = 1.01) [55]. The
In order to investigate the anisotropic behavior of the Janus 2D B2 P6 , electronic and hole relaxation time 𝜏 𝑒𝑙 are also determined based on the
we apply the anisotropic DPA [55] as carrier mobility 𝜇2D as 𝜏 𝑒𝑙 = (𝑚∗ 𝜇2D )∕𝑒. The calculated 𝜏 𝑒𝑙 are listed
in Table 1 at 𝑇 = 300 K. The relationship between temperature and
5𝑒ℏ3 (5𝐶11 + 3𝐶22 ) relaxation time 𝜏 𝑒𝑙 is shown in Figure S6 in the SI.
𝜇𝑥 = , (4)
2𝑘𝐵 𝑇 (𝑚𝑥 ) (𝑚∗𝑦 )1∕2 (9𝐸𝑥2 + 7𝐸𝑥 𝐸𝑦
∗ 3∕2 + 4𝐸𝑦2 )
and 3.4. Thermoelectric properties

5𝑒ℏ3 (3𝐶 11 + 5𝐶22 )

𝜇𝑦 = , (5) 3.4.1. Power factor
2𝑘𝐵 𝑇 (𝑚∗𝑥 )1∕2 (𝑚∗𝑦 )3∕2 (4𝐸𝑥2 + 7𝐸𝑥 𝐸𝑦 + 9𝐸𝑦2 ) In Figs. 4 (a), we show the absolute value of the Seebeck coefficients
where ℏ is the Planck constant, 𝑘𝐵 is the Boltzmann constant, 𝑇 is |𝑆| of the Janus 2D B2 P6 as functions of carrier concentration 𝑛 for the
the temperature, 𝑚∗𝑥 and 𝑚∗𝑦 are the effective mass along the 𝑥 and p-type and n-type B2 P6 at different temperatures. We can see that |𝑆|
𝑦 directions, respectively. The elastic modulus 𝐶11(22) is defined as decreases with increasing 𝑛 and decreasing 𝑇 because |𝑆| ∝ 1∕𝑛 and
𝐶11(22) = [𝜕 2 𝐸total ∕𝜕(𝑙𝑥(𝑦) ∕𝑙𝑥0(𝑦0) )2 ]∕𝑆0 where 𝐸total is the total energy |𝑆| ∝ 𝑇 [62,63], respectively. 𝑆 in the 𝑦 direction is larger than that in
after applying uniaxial strain, where 𝑆0 is the area of the unit cell at the 𝑥 direction for p-type doping due to the higher hole effective mass
equilibrium. 𝑙𝑥(𝑦) and 𝑙𝑥0(𝑦0) are the lattice constant after and before in the 𝑦 direction (see Table 1). In Figs. 4 (b), we show the electrical
deformation along the 𝑥- and 𝑦- directions, respectively (see Fig. S4 conductivity 𝜎 as functions of 𝑛 for both p-type and n-type. We noted
in the SI). 𝐸𝑥(𝑦) = 𝛥𝐸𝑥(𝑦) ∕𝜀𝑥(𝑦) is the deformation potential constant, that the 𝜎 values are rescaled by ℎ∕𝑑0 , where ℎ = 30 Å represents
in which 𝛥𝐸𝑥(𝑦) is the energy shift of the band edge of the CBM (or the length of the cell along the 𝑧 axis, and 𝑑0 = 7.878 Å denotes the
the VBM) with respect to the vacuum level under strain along the effective layer thickness, as detailed in Section 3.1. In Fig. 4(c), we
𝑥(𝑦) direction (see Fig. S5 in the SI). From Eqs. (4) and (5), we can show the electronic thermal conductivity, 𝜅𝑒𝑙 as a function of carrier
see that the carrier mobility in one direction is contributed by the concentration.
elastic modulus (𝐶11 and 𝐶22 ), the effective mass (𝑚𝑥 and 𝑚𝑦 ), and By combining |𝑆| and 𝜎, we obtain the PF = 𝑆 2 𝜎 of the p- and
the deformation potential constants (𝐸𝑥 and 𝐸𝑦 ) along both directions. n-type as a function of carrier concentration in the 𝑥- and 𝑦- directions,
The parameters calculated for the Janus 2D B2 P6 at room temperature as shown in Fig. 4(c). At room temperature (𝑇 = 300 K), the PF values

4
V.V. Thanh et al. Surfaces and Interfaces 44 (2024) 103829

Fig. 4. (a) Seebeck coefficient 𝑆, (b) electrical conductivity 𝜎, (c) electronic thermal conductivity 𝜅𝑒𝑙 , and (d) power factor PF are plotted as a function of the carrier concentration
along the 𝑥- (dashed lines) and 𝑦- (solid lines) directions for the p-type and n-type Janus 2D B2 P6 .

reach 0.08 and 0.10 W/mK2 for the n-type and p-type doping along
the 𝑦 direction, respectively. As shown in Fig. 4 (a) and (b), for the
p-type B2 P6 , 𝑆 and 𝜎 along the 𝑦-direction are larger than that along
the 𝑥-direction, leading to the higher PF along the 𝑦-direction. We
note that p-type B2 P6 has high 𝜎 along the 𝑦-direction, even though
it has a smaller carrier mobility. This is because the significantly larger
𝜏 el = 9.629 × 10−13 s in the 𝑦 direction. On the other hand, the high
PF for n-type might originate from the band convergence, in which the
difference energy between the bottom of the conduction bands at the X
point and along Y→ 𝛤 (see Fig. 3(a)) is 𝛥𝐸 = 0.03 eV. When the band
convergence happens, it will enhance the density of the state without
modifying the effective mass [14]. Thus, it increases the charge carrier
concentration for a given Fermi level, which typically enhances the PF.
By using the two-band model, it was proved that the PF is enhanced
with 𝛥𝐸 < 4𝑘𝐵 𝑇 = 0.1 eV at 𝑇 = 300 K [14]. Therefore, the high PF of
Fig. 5. Lattice thermal conductivity 𝜅ph is plotted as a function of temperature along
the Janus 2D B2 P6 comes not only from the anisotropic band structure 𝑥- and 𝑦- directions of the Janus 2D B2 P6 .
but also from the band convergence effect. As the power factor PF
is directly proportional to the output power density, a higher PF will
result in an increased output power density for the TE device [6].

5
V.V. Thanh et al. Surfaces and Interfaces 44 (2024) 103829

Fig. 6. (a) Phonon dispersion of the Janus 2D B2 P6 . (b) Phonon group velocity along 𝑥- and 𝑦-direction are plotted as a function of frequency. (c) Phonon-phonon scattering rate
and (d) the Grüneisen parameter are plotted as a function of frequency at 𝑇 = 300 K. Vertical dashed lines indicate the acoustic phonon modes region.

3.4.2. Figure of merit ZT Table 2

In Fig. 5, we show the lattice thermal conductivity 𝜅𝑝ℎ as a function The optimized power factor PFmax (W/mK2 ) of the Janus 2D B2 P6 and other typical
2D TE materials at room temperature.
of the temperature along 𝑥- and 𝑦- directions of the Janus 2D B2 P6 . It
Material Type n (cm−2 ) PFmax Direction
is noted that 𝜅𝑝ℎ is also scaled by a factor of ℎ∕𝑑0 . The calculated 𝜅𝑝ℎ
shows an anisotropic behavior, in which 𝜅𝑝ℎ in the 𝑦-direction is larger B2 P 6 n-type 1.05 × 1016 0.08 zigzag
(In this study) p-type 2.05 × 1016 0.10 zigzag
than that in the 𝑥-directions. This behavior can be explained by the
anisotropic phonon-group velocity of the acoustic phonon around the BP [23] p-type 6.04 × 1012 0.01 armchair

𝛤 point. As shown in Fig. 6(a), we can see that the longitudinal acoustic 𝛾−Ge2 SSe [28] p-type 2.30 × 1012 0.08 zigzag
along the 𝑦 direction (𝛤 → Y) is higher than that along the 𝑥 direction OsTe2 [63] n-type – 0.027 –
(𝛤 → X), resulting in phonon group velocity along the 𝑥 direction is InSe [18] p-type 2.40 × 1021 0.05 armchair
smaller than that along the 𝑦 direction, as shown in Fig. 6(b). MoS2 [73] n-type 1.00 × 1012 0.03 –
At 𝑇 = 300 K, the calculated 𝜅𝑝ℎ are 41.45 and 133.3 Wm−1 K−1
PdSeTe [74] p-type – 0.04 𝑥
along the 𝑥- and 𝑦- directions, respectively, which are comparable
Al2 SSe [75] n-type 1.00 × 1013 0.04 –
with the 2D BP (∼ 40 and 110 Wm−1 K−1 ) [64]. Since 𝜅𝑝ℎ of B2 P6
is relatively high compared with the common TE materials, such as
Bi2 Te3 [65] or Mg3 Bi2 [66](𝜅𝑝ℎ ∼ 1 − 2 Wm−1 K−1 ), one should
applied the strain engineering or constructing heterojunction to reduce Yang et al. [71] showed that the four-phonon processes could play a
𝜅𝑝ℎ [67,68], which should be the future work. dominant role in materials that have a large frequency gap between
In Figs. 6 (c) and (d), we show the phonon scattering rates and the acoustic and optical phonon modes or the bunching of acoustic
Grüneisen parameters at 𝑇 = 300 K, respectively. Our calculations and low-frequency optical phonon modes. On the other hand, the
clearly show that the lattice thermal conductivity (𝜅𝑝ℎ ) is primar-
phonon dispersion of Janus B2 P6 has a frequency gap between the high-
ily affected by three acoustic phonon modes with frequencies below
frequency optical modes, as shown in Fig. 6(a). Thus, we expect that
200 cm−1 . For the phonon–phonon scattering, there is the normal pro-
the four-phonon processes might not significantly contribute to 𝜅𝑝ℎ .
cess (that conserves energy and momentum) and the Umklapp process
Finally, by combining the PF, 𝜅𝑝ℎ , and 𝜅𝑒𝑙 , we obtain the 𝑍𝑇 values
(that conserves only energy and momentum is altered). The normal
for both the p- and n-type at different temperatures, as shown in
process varies linearly with 𝜔, while the Umklapp process varies with
Figs. 7 (a) and (b), respectively. We can see that the 𝑍𝑇 values exhibit
𝜔2 [65]. As shown in Fig. 6(c), the scattering rate is proportional to
the fitting line 𝜔2 . Therefore, the phonon-phonon scattering is mainly anisotropy along the 𝑥- and 𝑦- directions for both the p- and n-type,
dominated by the Umklapp process [65,69]. We also can find a high which originate from the anisotropy behavior of the electronic and
anharmonicity (or high scattering rate up to 0.2 ps−1 ) around 50 cm−1 phonon transports. The maximum 𝑍𝑇 reaches to 0.7 for the n-type at
at the X point (see Fig. 6(b)), which origin from the vibration of the 𝑇 = 900 K, which is higher than that of InSe monolayer (0.21 at 300
P atoms around the B atoms in the 𝑧 direction, as shown in Fig. S9. K) [14] and MoS2 monolayer (0.1 at 500 K) [72].
We also note that the peak of the scattering rate at around 400 cm−1 We note that the DPA gives reasonable estimates of 𝜏 𝑒𝑙 to calcu-
comes from the overlapping of many optical phonon modes with the late the PF and 𝑍𝑇 for only electron-acoustic-phonon interactions.
frequency of around 400 cm−1 [70], as shown in Fig. 6(b). Therefore, we should still expect the contributions of the electron-
Since a large frequency gap of about 100 cm−1 is found at 500 cm−1 , optical-phonon interactions and other scattering mechanisms to the
as shown in Fig. 6(a), the four-phonon processes might occur. However, total 𝜏 𝑒𝑙 [68]. Because of possible overestimation in the 𝜏 𝑒𝑙 calculated

6
V.V. Thanh et al. Surfaces and Interfaces 44 (2024) 103829

Declaration of competing interest

The authors declare the following financial interests/personal rela-

tionships which may be considered as potential competing interests:
Nguyen Tuan Hung reports financial support was provided by Tohoku
University. If there are other authors, they declare that they have
no known competing financial interests or personal relationships that
could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.

Acknowledgments

N.T.H. acknowledges the financial support from Frontier Research

Institute for Interdisciplinary Sciences, Tohoku University, Japan.

Appendix A. Supplementary data

Supplementary material related to this article can be found online

at https://doi.org/10.1016/j.surfin.2023.103829.

References

[1] L. Yang, Z.-G. Chen, M.S. Dargusch, J. Zou, High performance thermoelectric
materials: Progress and their applications, Adv. Energy Mater. 8 (6) (2018)
1701797.
[2] K. Biswas, Z. Ren, Y. Grin, K.H. Lee, T. Mori, L. Chen, Thermoelectric materials
science and technology toward applications, Appl. Phys. Lett. 121 (7) (2022).
Fig. 7. Figure of merit 𝑍𝑇 is plotted as a function of carrier concentration along the [3] J.P. Heremans, M.S. Dresselhaus, L.E. Bell, D.T. Morelli, When thermoelectrics
𝑥- (dashed lines) and 𝑦- (solid lines) directions for (a) p-type (b) and n-type Janus 2D reached the nanoscale, Nat. Nanotechnol. 8 (7) (2013) 471–473.
B2 P6 . [4] C.B. Vining, An inconvenient truth about thermoelectrics, Nature Mater. 8 (2)
(2009) 83–85.
[5] H.J. Goldsmid, et al., Introduction To Thermoelectricity, Vol. 121, Springer,
2010.
by the DPA model, in Figs. S7 and S8 in the SI, we plot the PF and [6] N.T. Hung, E.H. Hasdeo, A.R. Nugraha, M.S. Dresselhaus, R. Saito, Quantum
𝑍𝑇 values as a function of carrier concentration with several values effects in the thermoelectric power factor of low-dimensional semiconductors,
of 𝜏 𝑒𝑙 . The PF and 𝑍𝑇 are reduced by the presence of other scattering Phys. Rev. Lett. 117 (3) (2016) 036602.
[7] H.S. Kim, W. Liu, G. Chen, C.-W. Chu, Z. Ren, Relationship between thermoelec-
mechanisms. Nevertheless, considering the same 𝜏 𝑒𝑙 , our calculated PF
tric figure of merit and energy conversion efficiency, Proc. Natl. Acad. Sci. USA
values are larger than the previously reported 2D materials, ranging 112 (27) (2015) 8205–8210.
from 0.01 to 0.08 W/mK2 at 300 K [18,23,73,74], as shown in Table 2. [8] W. Liu, H.S. Kim, S. Chen, Q. Jie, B. Lv, M. Yao, Z. Ren, C.P. Opeil, S. Wilson,
Thus, our obtained results suggest that the anisotropic Janus B2 P6 could C.-W. Chu, et al., N-type thermoelectric material Mg2 Sn0.75 Ge0.25 for high power
serve as suitable TE materials for the design of future devices. generation, Proc. Natl. Acad. Sci. 112 (11) (2015) 3269–3274.
[9] W. Liu, H.S. Kim, Q. Jie, Z. Ren, Importance of high power factor in thermoelec-
tric materials for power generation application: A perspective, Scr. Mater. 111
4. Conclusions (2016) 3–9.
[10] L. Hicks, M. Dresselhaus, Use of quantum-well superlattices to obtain a high
In summary, using DFT calculations, we investigated the TE proper- figure of merit from nonconventional thermoelectric materials, in: MRS Online
ties of the anisotropic Janus 2D B2 P6 . Our results demonstrate that the Proceedings Library (OPL), Vol. 326, 1993.
Janus 2D B2 P6 exhibits significant anisotropic TE properties. Specifi- [11] M.S. Dresselhaus, G. Chen, M.Y. Tang, R. Yang, H. Lee, D. Wang, Z. Ren, J.-P.
Fleurial, P. Gogna, New directions for low-dimensional thermoelectric materials,
cally, it demonstrates a remarkably high electron carrier mobility of
Adv. Mater. 19 (8) (2007) 1043–1053.
1386 cm2 V−1 s−1 along the 𝑦 (or zigzag) direction. Additionally, the [12] N.T. Hung, R. Saito, The origin of quantum effects in low-dimensional
PF for p-type reaches a high value of up to 0.1 W/mK2 at 900 K. The thermoelectric materials, Adv. Quantum Technol. 4 (1) (2021) 2000115.
optimized values of 𝑍𝑇 for p-type and n-type reach 0.6 and 0.7 at 900 [13] H. Lv, W. Lu, D. Shao, Y. Sun, Enhanced thermoelectric performance of
K, respectively. Moreover, the Janus 2D B2 P6 exhibits high mechanical phosphorene by strain-induced band convergence, Phys. Rev. B 90 (8) (2014)
085433.
anisotropy, with an ideal strength of 18.62 N/m in the 𝑦 direction,
[14] N.T. Hung, A.R. Nugraha, T. Yang, Z. Zhang, R. Saito, Thermoelectric perfor-
which is 1.74 times higher than that in the 𝑥 direction (10.7 N/m). mance of monolayer InSe improved by convergence of multivalley bands, J.
Based on these findings, we propose that the anisotropy Janus 2D B2 P6 Appl. Phys. 125 (8) (2019) 082502.
can be utilized for good potential high-performance and flexible TE [15] L. Li, Y. Yu, G.J. Ye, Q. Ge, X. Ou, H. Wu, D. Feng, X.H. Chen, Y. Zhang, Black
material. phosphorus field-effect transistors, Nat. Nanotechnol. 9 (5) (2014) 372.
[16] F. Xia, H. Wang, Y. Jia, Rediscovering black phosphorus as an anisotropic layered
material for optoelectronics and electronics, Nature Commun. 5 (1) (2014) 4458.
CRediT authorship contribution statement [17] J. Zhang, H. Liu, L. Cheng, J. Wei, J. Liang, D. Fan, P. Jiang, L. Sun, J. Shi,
High thermoelectric performance can be achieved in black phosphorus, J. Mater.
Vuong Van Thanh: Conceptualization, Data curation, Formal anal- Chem. C 4 (5) (2016) 991–998.
ysis, Investigation, Methodology, Writing – original draft, Writing – re- [18] N.T. Hung, A.R. Nugraha, R. Saito, Two-dimensional InSe as a potential
thermoelectric material, Appl. Phys. Lett. 111 (9) (2017) 092107.
view & editing. Do Van Truong: Conceptualization, Writing – review &
[19] D.A. Bandurin, A.V. Tyurnina, L.Y. Geliang, A. Mishchenko, V. Zólyomi, S.V.
editing. Nguyen Tuan Hung: Conceptualization, Data curation, Formal Morozov, R.K. Kumar, R.V. Gorbachev, Z.R. Kudrynskyi, S. Pezzini, et al.,
analysis, Funding acquisition, Investigation, Methodology, Supervision, High electron mobility, quantum hall effect and anomalous optical response in
Writing – original draft, Writing – review & editing. atomically thin inse, Nat. Nanotechnol. 12 (3) (2017) 223–227.

7
V.V. Thanh et al. Surfaces and Interfaces 44 (2024) 103829

[20] Y. Saito, T. Iizuka, T. Koretsune, R. Arita, S. Shimizu, Y. Iwasa, Gate-tuned [48] F. Mouhat, F. Coudert, Necessary and sufficient elastic stability conditions in
thermoelectric power in black phosphorus, Nano Lett. 16 (8) (2016) 4819–4824. various crystal systems, Phys. Rev. B 90 (2014) 224104.
[21] E. Flores, J.R. Ares, A. Castellanos-Gomez, M. Barawi, I.J. Ferrer, C. Sánchez, [49] J. Tao, W. Shen, S. Wu, L. Liu, Z. Feng, C. Wang, C. Hu, P. Yao, H. Zhang, W.
Thermoelectric power of bulk black-phosphorus, Appl. Phys. Lett. 106 (2) (2015). Pang, et al., Mechanical and electrical anisotropy of few-layer black phosphorus,
[22] H. Liu, A.T. Neal, Z. Zhu, Z. Luo, X. Xu, D. Tománek, P.D. Ye, Phosphorene: An ACS Nano 9 (11) (2015) 11362–11370.
unexplored 2D semiconductor with a high hole mobility, ACS Nano 8 (4) (2014) [50] M. Siskins, M. Lee, F. Alijani, M.R. Van Blankenstein, D. Davidovikj, H.S. Van
4033–4041. Der Zant, P.G. Steeneken, Highly anisotropic mechanical and optical properties
[23] H. Lv, W. Lu, D. Shao, Y. Sun, Large thermoelectric power factors in black of 2D layered As2 S3 membranes, ACS Nano 13 (9) (2019) 10845–10851.
phosphorus and phosphorene, 2014, arXiv preprint arXiv:1404.5171. [51] N.T. Hung, A.R. Nugraha, R. Saito, Two-dimensional MoS2 electromechanical
[24] B. Liao, J. Zhou, B. Qiu, M.S. Dresselhaus, G. Chen, Ab initio study of actuators, J. Phys. D: Appl. Phys. 51 (7) (2018) 075306.
electron-phonon interaction in phosphorene, Phys. Rev. B 91 (23) (2015) [52] V. Van Thanh, N.T. Hung, et al., Charge-induced electromechanical actuation of
235419. two-dimensional hexagonal and pentagonal materials, Phys. Chem. Chem. Phys.
[25] J. Pei, X. Gai, J. Yang, X. Wang, Z. Yu, D.-Y. Choi, B. Luther-Davies, Y. Lu, 21 (40) (2019) 22377–22384.
Producing air-stable monolayers of phosphorene and their defect engineering, [53] J.-W. Jiang, H.S. Park, Mechanical properties of single-layer black phosphorus,
Nature Commun. 7 (1) (2016) 10450. J. Phys. D: Appl. Phys. 47 (38) (2014) 385304.
[26] M. Sun, U. Schwingenschlogl, B2 p6 : A two-dimensional anisotropic janus material [54] S. Bertolazzi, J. Brivio, A. Kis, Stretching and breaking of ultrathin MoS2 , ACS
with potential in photocatalytic water splitting and metal-ion batteries, Chem. Nano 5 (12) (2011) 9703–9709.
Mater. 32 (2020) 4795–4800. [55] H. Lang, S. Zhang, Z. Liu, Mobility anisotropy of two-dimensional semiconduc-
[27] K. Ren, H. Shu, W. Huo, Z. Cui, J. Yu, Y. Xu, Mechanical, electronic and optical tors, Phys. Rev. B. 94 (23) (2016) 235306.
properties of a novel B2 P6 monolayer: Ultrahigh carrier mobility and strong [56] J.-W. Jiang, H.S. Park, Mechanical properties of single-layer black phosphorus,
optical absorption, Phys. Chem. Chem. Phys. 23 (43) (2021) 24915–24921. J. Phys. D: Appl. Phys. 47 (38) (2014) 385304.
[28] V.V. Thanh, D.V. Truong, N. Tuan Hung, Janus 𝛾-gesse monolayer as a high- [57] V.V. Thanh, N.D. Van, D.V. Truong, N.T. Hung, Charge-induced high-
performance material for photocatalysis and thermoelectricity, ACS Appl. Energy performance actuation of borophene, J. Phys. D: Appl. Phys. 54 (10) (2020)
Mater. 6 (2) (2023) 910–919. 105504.
[29] J. Qiao, X. Kong, Z.-X. Hu, F. Yang, W. Ji, High-mobility transport anisotropy and [58] Q. Wei, X. Peng, Superior mechanical flexibility of phosphorene and few-layer
linear dichroism in few-layer black phosphorus, Nature Commun. 5 (1) (2014) black phosphorus, Appl. Phys. Lett. 104 (25) (2014) 251915.
4475. [59] V. Tran, R. Soklaski, Y. Liang, L. Yang, Layer-controlled band gap and anisotropic
[30] Y. Cai, G. Zhang, Y.-W. Zhang, Polarity-reversed robust carrier mobility in excitons in few-layer black phosphorus, Phys. Rev. B 89 (23) (2014) 235319.
monolayer MoS2 nanoribbons, J. Electrochem. Soc. 136 (17) (2014) 6269–6275. [60] X. Chen, Y. Wu, Z. Wu, Y. Han, S. Xu, L. Wang, W. Ye, T. Han, Y. He, Y. Cai, et
[31] R. Gupta, B. Dongre, C. Bera, J. Carrete, The effect of janus asymmetry on al., High-quality sandwiched black phosphorus heterostructure and its quantum
thermal transport in SnSSe, J. Phys. Chem. C 124 (32) (2020) 17476–17484. oscillations, Nature Commun. 6 (1) (2015) 7315.
[32] P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, [61] A. Patel, D. Singh, Y. Sonvane, P. Thakor, R. Ahuja, ACS Appl. Mater. Interfaces
G.L. Chiarotti, M. Cococcioni, I. Dabo, et al., QUANTUM ESPRESSO: A modular 12 (41) (2020) 46212–46219.
and open-source software project for quantum simulations of materials, J. Phys. [62] H. Minhas, S. Das, B. Pathak, Ultralow thermal conductivity and high thermoelec-
Condens. Matter 21 (39) (2009) 395502. tric performance of 𝛾-GeSe: Effects of dimensionality and thickness, ACS Appl.
[33] N.T. Hung, A.R. Nugraha, R. Saito, Quantum ESPRESSO Course for Solid-State Energy Mater. 5 (8) (2022) 9914–9928.
Physics, Jenny Stanford Publishing, New York, 2022. [63] S.-Z. Huang, J.-X. Guo, B.-Y. Wang, H.-D. Yang, Q.-Y. Feng, B. Li, X. Xiang, X.-T.
[34] G. Kresse, D. Joubert, From ultrasoft pseudopotentials to the projector Zu, H.-X. Deng, Optical-acoustic phonon hybridization enhanced thermoelectric
′
augmented-wave method, Phys. Rev. B 59 (3) (1999) 1758. performance in a 1T Phase OsTe2 monolayer, ACS Appl. Energy Mater. 5 (11)
[35] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made (2022) 14513–14521.
simple, Phys. Rev. Lett. 77 (18) (1996) 3865. [64] A. Jain, A.J. McGaughey, Strongly anisotropic in-plane thermal transport in
[36] J. Heyd, G.E. Scuseria, M. Ernzerhof, Hybrid functionals based on a screened single-layer black phosphorene, Sci. Rep. 5 (1) (2015) 8501.
Coulomb potential, Chem. Phys. 118 (18) (2003) 8207–8215. [65] N.T. Hung, A.R. Nugraha, R. Saito, Designing high-performance thermoelectrics
[37] A.A. Mostofi, J.R. Yates, G. Pizzi, Y.-S. Lee, I. Souza, D. Vanderbilt, N. Marzari, in two-dimensional tetradymites, Nano Energy 58 (2019) 743–749.
An updated version of wannier90: A tool for obtaining maximally-localised [66] N.T. Hung, The role of spin–orbit interaction in low thermal conductivity of
wannier functions, Comput. Phys. Comm. 185 (8) (2014) 2309–2310. Mg3 Bi2 , Appl. Phys. Lett. 123 (2023) 252109.
[38] S. Baroni, S. De Gironcoli, A. Dal Corso, P. Giannozzi, Phonons and related crystal [67] X.-K. Chen, X.-Y. Hu, P. Jia, Z.-X. Xie, J. Liu, Tunable anisotropic thermal
properties from density-functional perturbation theory, Rev. Modern Phys. 73 transport in porous carbon foams: The role of phonon coupling, Int. J. Mech.
(2001) 515. Sci. 206 (2021) 106576.
[39] G.J. Martyna, M.L. Klein, M. Tuckerman, Nosé–Hoover chains: The canonical [68] X.-K. Chen, E.-M. Zhang, D. Wu, K.-Q. Chen, Strain-induced medium-temperature
ensemble via continuous dynamics, J. Chem. Phys 97 (4) (1992) 2635–2643. thermoelectric performance of Cu4 Ti Se4 : The role of four-phonon scattering,
[40] G. Pizzi, D. Volja, B. Kozinsky, M. Fornari, N. Marzari, BoltzWann: A code Phys. Rev. A 19 (4) (2023) 044052.
for the evaluation of thermoelectric and electronic transport properties with a [69] T. Ghosh, M. Dutta, D. Sarkar, K. Biswas, Insights into low thermal conductivity
maximally-localized wannier functions basis, Comput. Phys. Comm. 185 (2014) in inorganic materials for thermoelectrics, J. Am. Chem. Soc. 144 (23) (2022)
422–429. 10099–10118.
[41] W. Li, J. Carrete, N.A. Katcho, N. Mingo, ShengBTE: A solver of the Boltz- [70] E. Di Lucente, M. Simoncelli, N. Marzari, Crossover from Boltzmann to wigner
mann transport equation for phonons, Comput. Phys. Comm. 185 (6) (2014) thermal transport in thermoelectric skutterudites, Phys. Rev. Res. 5 (2023)
1747–1758. 033125.
[42] W. Tang, E. Sanville, G. Henkelman, A grid-based bader analysis algorithm [71] X. Yang, T. Feng, J. Li, X. Ruan, Stronger role of four-phonon scattering than
without lattice bias, J. Phys.: Condens. Matter 21 (8) (2009) 084204. three-phonon scattering in thermal conductivity of III-V semiconductors at room
[43] X.-B. Li, P. Guo, T.-F. Cao, H. Liu, W.-M. Lau, L.-M. Liu, Structures, stabilities temperature, Phys. Rev. B 100 (24) (2019) 245203.
and electronic properties of defects in monolayer black phosphorus, Sci. Rep. 5 [72] Z. Jin, Q. Liao, H. Fang, Z. Liu, W. Liu, Z. Ding, T. Luo, N. Yang, A revisit
(1) (2015) 10848. to high thermoelectric performance of single-layer MoS2 , Sci. Rep. 5 (1) (2015)
[44] T. Yu, C. Wang, X. Yan, G. Yang, U. Schwingenschlogl, Anisotropic janus SiP2 18342.
monolayer as a photocatalyst for water splitting, J. Phys. Chem. Lett. 12 (9) [73] H. Babaei, J. Khodadadi, S. Sinha, Large theoretical thermoelectric power factor
(2021) 2464–2470. of suspended single-layer MoS2 , Appl. Phys. Lett. 105 (19) (2014).
[45] K. Lee, É.D. Murray, L. Kong, B.I. Lundqvist, D.C. Langreth, Higher-accuracy van [74] H. Huang, Z. Sun, C. Hu, X. Fan, Janus penta-PdSeTe: A two-dimensional
der Waals density functional, Phys. Rev. B 82 (8) (2010) 081101. candidate with high thermoelectric performance, J. Alloys Compd. 924 (2022)
[46] Y. Yu, C. Zhou, T. Ghosh, C.-F. Schön, Y. Zhou, S. Wahl, M. Raghuwanshi, P. 166581.
Kerres, C. Bellin, A. Shukla, et al., Doping by design: Enhanced thermoelectric [75] G.S. Khosa, S. Gupta, R. Kumar, First-principles investigations of electronic and
performance of GeSe alloys through metavalent bonding, Adv. Mater. 35 (19) thermoelectric properties of janus Al2 SSe monolayer, Phys. B: Condens. Matter
(2023) 2300893. 615 (2021) 413057.
[47] S. Zhang, M. Xie, F. Li, Z. Yan, Y. Li, E. Kan, W. Liu, Z. Chen, H. Zeng,
Semiconducting group 15 monolayers: A broad range of band gaps and high
carrier mobilities, Angew. Chem. 128 (5) (2016) 1698–1701.