Ionic conductivity

We obtained the Na ionic conductivities of Na2 X 2 TeO 6 ¿ Mg , Zn ) by

conducting the ab-initio molecular dynamics (AIMD) simulations of the
2 ×2 ×1 Na2 X 2 TeO6 supercell in the NVT ensemble with a Nosé-Hoover
thermostat. 38 The simulation temperatures ranged from 600 K to 900 K . To
reduce the calculation expense, the Brillouin zone was sampled only at the r
-point. At each temperature, the time step was set as 2 fs , and the total
simulation time was 100 ps. The Na self-diffusion coefficient ( D ¿ ) could be
obtained by the slope of the mean square displacements (MSD) over the
time scale by the equation 39

N
1 1
D¿ =
2dt
MSD (Na)= ∑
2 dNt i=1
2
❑ ( r i ( t+t 0 )−r i ( t 0 ) ) , (eq 3 )

where r i ( t+ t 0 ) is the position of current Na atom with respect to the starting

position r i ( t 0 ) ,t is the simulation time, N is the number of diffusing Na atoms,
and d is the number of diffusion dimensions. By calculating the self-diffusion
coefficients at elevated temperatures, the activation energy ( E a ) and
diffusion coefficient at room temperature could be obtained by the Arrhenius
equation 40

¿
D =D0 ⋅exp ⁡
( ) −E α
ks T
.

Here D 0 represents the pre-factor of the diffusion and k B is the Boltzmann

constant. Then we calculated the Na ionic conductivity using the Nernst-
Einstein relation 41

2 2
ρz F ¿
σ= D ,
k yT

where the density ρ is the molar density of Na ions in the unit cell, z is the
integer charge of Na ion ( +1 for Na ion), F is the Faraday constant, k B is the
Boltzmann constant, and T , is the temperature. To get detailed jump Deypes 10
âdo hopping possibilities, the transitions between stable Na sites inside the
structures were monitored during the whole MD simulations. By counting
the number of jumps J n between stable Na sites, the mean jump rate Γ n
could be obtained by 42

Jn
Γn=
Nt

where the subscript n denotes the type of jumps. Using the Einstein-
Smulochowski relation the jump rates are related to the jump diffusivity
( D J ) 4 42
 2
Γ n ⋅ an
D J =∑ ❑
n 2d

Here a n is the jump distance of the jump type n . The sum is over all jump
types contributing to the whole macroscopic diffusion.

In the materials with the diffusion ions at high occupancy concentration,

collective ion transport plays an important role to form a smooth diffusion
energy landscape, through collective migrations of multiple ions with low
energy barriers. 43 To study the collective Na ion transport in Na2 X 2 TeO 6, the
collective climbing nudged elastic band (CI-NEB) method 44 was adopted.
The diffusion paths of two adjacent Na ions in the interlayer to their next
equilibrium positions were chosen. 5 transitional images were used, and the
calculation parameter settings were similar to the calculations in the
structural optimizations. Especially, the energy and force change
convergencies were set to 10−5 eV and 0.02 eV /ÅÅ .

Therefore, the cohesive energy per Li-ion does not increase when the
number of Li ions increases. The adsorption ( Eadsorption ), binding ( E binding ), and
cohesive energies ( Ecohcsive ) are the fundamental parameters that govern the
stability of a structure as they quantify the favorable intermolecular
interactions in a system. Thus, we can describe the energies as

E adsorption = Ecomplex −E substrate −E Li

E + n× E Li−E complex
E binding = substratc
n
Ecomplex −Esubstrate −n× E Li
Ecohcsivc =
n

where Ecomplex , Esubstrate , and E Li represent the total energies of the complex
system (Li + graphene), isolated adsorbate (free adsorbate atoms such as
Li ), and isolated substrate with a free surface, respectively. These equations
indicate that a negative cohesive energy results in greater structural
cohesion. The formation energy (unit: meV per 6C) of a solid or compound is
the difference between the energy of the elements in their normal state and
that of the state wherein the solid or compound is formed as shown in Fig.
1h. Formation energy can be described using eqn (4) (ref. 44) as follows:

Eformation =E ( Li x C 6 )−(1−x )E ( C 6 ) −xE ( LiC 6 )

where E ( Li x C6 ) , E ( C 6 ), and E ( LiC 6 ) are energies of the sampled configuration,

AB-stacked graphite (x=0) and AA -stacked LiC 6 ⁡¿ 1), respectively. The
formation energy must be negative for a stable structure. As compounds are
rarely made from isolated atoms, formation energy, rather than cohesive
energy, is a more practical measure of structural stability. The second
section elucidates the details of using formation energy in experimental
studies.
2.1.1.1 Field of application. DFT calculations of adsorption, binding,
cohesive, and formation energies are pivotal for Li-ion battery component
development. These computations illuminate interactions between Li-ions
and materials, aiding material screening and optimization. They offer
insights into degradation mechanisms, guide material design, and predict
battery performance.

2.1.2 Gibbs free energy. In theory, ab initio molecular dynamics (AIMD)

simulations provide a parameter-free approach to evaluate ionic diffusivities,
but in practice, they require significant computer resources. Since each
AIMD step is essentially an entirely new DFT calculation, the convergence of
diffusivity may require nearly tens of thousands of computational steps.

The following eqn (5) and (6) define the Gibbs free energy of a system:

Δ G=Δ H −T Δ S
Δ H =Δ E−P Δ V

where Δ G , Δ H , Δ S , T , Δ E , P , and Δ V represent the Gibbs free energy,

enthalpy, entropy, temperature, internal energy, pressure, and volume of a
system, respectively.

Gibbs free energy (Δ G) of a metal (represented by X: Li, Na, K, Zn , Al, etc.)-

incorporated system can be deduced from the difference between the DFT-
calculated internal energy of the structure before X insertion and that after
X insertion (Subsection 2.7 provides a more in-depth explanation of this
topic). Δ G can be used to evaluate the relative stability of isomers and
polymorphs under different conditions, such as temperature and pressure. A
fully charged battery has a highly reactive environment; hence, the
separator material must be chemically stable against the electrolyte and
electrode components. The PHONOPY package45 can be used to calculate
various thermal properties, including DG at a constant pressure, using the
quasi-harmonic approximation. PHONOPY provides a robust interface for
various simulation packages such as VASP, QE, SIESTA, WIEN2K, ABINIT,
CRYSTAL, TURBOMOLE, CP2K, CASTEP, etc. to calculate the above thermal
properties. The VaspGibbs46 tool is also available for calculating the Gibbs
free energy for a given temperature. It will automatically compute the
moment of inertia and symmetry of your molecule and compute rotational
and translational contributions. 2.1.2.1 Field of application. Gibbs free
energy calculations using DFT are highly relevant to experimental Li-ion
battery components. These calculations provide critical insights into the
thermodynamic stability, phase transitions, and reaction kinetics of
materials within Li-ion batteries. They guide material selection, design of
electrolytes, and the formation of the solid electrolyte interphase (SEI) layer.

2.3 Thermoelectrical properties (thermal, electrical, and ionic conductivity)

LIBs are particularly sensitive to temperature increases, which can have
serious consequences on their performance, lifetime, and safety. Therefore,
it is critical to have a thorough understanding of the battery's heat sources
and sinks. The Seebeck coefficient, electrical and thermal conductivity, and
relaxation time are all determined using the BoltzTrap 70 code based on the
smoothed Fourier interpolation of the bands, which is important for studying
the thermoelectric behavior of the interconnects. It is possible to
characterize the performance of a thermoelectric material for a given
temperature variation using a dimensionless quantity called the figure of
merit ZT , given in eqn (13):
2
σS
ZT= T
k

where Seebeck coefficient, electrical conductivity, thermal conductivity, and

absolute temperature are denoted as S , σ , k , and T , respectively.

To obtain the thermal conductivity of the materials, both electronic

component of k e and lattice component of κ 1 are required. The standard
BoltzTraP code considers only the electronic component of the thermal
conductivity; thus, κ 1 is neglected. It can be determined from eqn (14) as
follows:

1
κε= 2
e T
∫¿
The phono3py code 71 is a code that can be used to determine the lattice
thermal conductivities ( κ 1 ). In contrast to PHONOPY, which only allows the
displacement of one atom at a time, Phono3py allows two atoms to be
displaced at once, enabling
and 2.2 W mK −1, respectively, as shown in Fig. 9b and c.

the study of third-order force constants and phonon anharmonicity. When

comparing third- and second-order force constants, the computational cost
of the latter is considerably higher. Both relaxation time approximation
(RTA) and the full solution of the linearized Boltzmann transport equation
(LBTE) were used to determine kl. In the previously published work,14 we
described how different anode materials have different electrical
conductivities, as described in Fig. 9a. It is also necessary for the electrolyte
and separator to maintain their temperature stability despite working at
typical temperatures.72,73 Improving the internal thermal management of
LIBs can be achieved in part by increasing the thermal conductivity of solid
electrolyte (SE) materials. For ages, thermal transport characteristics of SE
materials have received little attention from researchers, and a few papers
have reported these features.74,75 Using experimental measurement, rst-
principles calculation, and the minimum temperature model, Cheng et al.75
investigated the thermal conductivity of NASICON-type
Li1.5Al0.5Ge1.5(PO4)3 (LAGP), garnet-type Li6.4La3Zr1.4Ta0.6O12
(LLZTO), and some amorphous sulde-type Li2S–P2S5 (LPS), Li6PS5Cl
(LPSC), and Na3. According to the results, the thermal conductivity of
amorphous sulde-type materials lies between 0.45 and 0.70 W Mk−1 at
room temperature (RT) (298 K), whereas the conductivities of LAGP and
LLZTO are around 1.4
It is very challenging to measure the ionic conductivity ( σ 0 ) of the materials
experimentally. 76 ,77 Thus, theoretical calculations have proven extremely
useful in directly comparing σ 0 between different systems. The Arrhenius
equation, as shown in eqn (15), relates the ionic conductivity of a material to
the activation energy based on ion migration. It provides a quantitative
expression of the ionic conductivity that varies with temperature.

( )
σ =σ 0 exp ⁡
−E α
KB T

where σ is the ionic conductivity at temperature (Kelvin), σ 0 is the pre-

exponential factor or pre-factor, representing the conductivity at infinite
temperature, E a is the diffusion activation energy, k B is Boltzmann's
constant, and T is the temperature. To obtain σ o, the Nernst-Einstein
equation is employed. E a and σ 0 are extracted from the slope and intercept
of the plot at different temperatures, respectively.

The ionic conductivity is an intrinsic property of a material and can be

directly correlated with the ionic mobility (μ) through the equation σ =nqμ ,
where σ represents the conductivity, n denotes the density of ions
responsible for carrying charge, and q represents the charge of the ions.
Recently, a theoretical framework has been introduced to elucidate the
phenomenon of ion drags in solids, establishing a correlation between a
higher value of the parameter μ and fundamental characteristics such as
reduced density and increased stiffness. 7 ∞ , 70
2.3.1 Field of application. It ensures safety through effective heat
management, promotes efficient charge transfer, enables rapid charging,
guides material selection and optimization, enhances interface
understanding, and supports high-rate capabilities.