metals

Article
Estimation of Two Component Activities of Binary Liquid
Alloys by the Pair Potential Energy Containing a Polynomial
of the Partial Radial Distribution Function
Jiulong Hang and Dongping Tao *

Faculty of Metallurgical and Energy Engineering, Kunming University of Science and Technology,
Kunming 650093, China; 17852587095@163.com
* Correspondence: dongpingt@aliyun.com

Abstract: An investigation of partial radial distribution functions and atomic pair potentials within a
system has established that the existing potential functions are rooted in the assumption of a static
arrangement of atoms, overlooking their distribution and vibration. In this study, Hill’s proposed
radial distribution function polynomials are applied for the pure gaseous state to a binary liquid alloy
system to derive the pair potential energy. The partial radial distribution functions of 36 binary liquid
alloy systems from the literature were used to bring in four thermodynamic models for validation.
Results show that the molecular interaction volume model (MIVM) and regular solution model
(RSM) outperform other models when an asymmetric method is used to calculate the partial radial
distribution function. The MIVM exhibits an average SD of 0.095 and an average ARD of 32.2%.
Similarly, the RSM demonstrates an average SD of 0.078 and an ARD of 32.2%. The Wilson model
yields an average SD of 0.124 and an average ARD of 226%. The nonrandom two-liquid (NRTL)
model exhibits an average SD of 0.225 and an average ARD of 911%. On applying the partial radial
distribution function symmetry method, the MIVM and RSM outperform the other models, with an
average SD of 0.143 and an average ARD of 165.9% for the MIVM. The RSM yields an average SD
of 0.117 and an average ARD of 208.3%. The Wilson model exhibits the average values of 0.133 and
305.6% for SD and ARD, respectively. The NRTL model shows an average SD of 0.200 and an average
ARD of 771.8%. Based on this result, the influence of the symmetry degree on the thermodynamic
Citation: Hang, J.; Tao, D. Estimation
model is explored by examining the symmetry degree as defined by the experimental activity curves
of Two Component Activities of
of the two components.
Binary Liquid Alloys by the Pair
Potential Energy Containing a
Polynomial of the Partial Radial
Keywords: pair potential; partial radial distribution function; binary liquid alloy
Distribution Function. Metals 2023,
13, 1773. https://doi.org/10.3390/
met13101773
1. Introduction
Academic Editor: Jürgen Eckert
Pair potentials play a fundamental role in comprehending the static and dynamic
Received: 26 August 2023 properties of gases, liquids, and solids [1]. The main objective of studying pair potentials is
Revised: 9 October 2023 to develop accurate and reliable models that describe interatomic interactions and enable
Accepted: 17 October 2023 simulation, prediction, and explanation of the behavior and properties of atomic systems.
Published: 19 October 2023
Various computational simulation studies, such as for protein folding, drug–receptor in-
teractions, and material property prediction, are conducted using atomic pair potentials
to ascertain the underlying mechanisms of atomic interactions, explore novel material
Copyright: © 2023 by the authors.
properties, and optimize drug design. Materials science and surface science researchers
Licensee MDPI, Basel, Switzerland. have investigated atomic potentials to thoroughly understand phenomena such as material
This article is an open access article structure, stability, crystal defects, and surface adsorption. Understanding interatomic inter-
distributed under the terms and actions is essential in material design, catalyst development, and comprehending material
conditions of the Creative Commons interfaces and surface phenomena. Atomic pair potential models encompass widely used
Attribution (CC BY) license (https:// semiempirical potentials such as the Lennard–Jones potential [2], Morse potential [3], and
creativecommons.org/licenses/by/ Born–Mayer potential [4], as well as atomic force fields, quantum mechanical descriptions,
4.0/). and statistical mechanics methods. Different factors affect the selection and examination of

Metals 2023, 13, 1773. https://doi.org/10.3390/met13101773 https://www.mdpi.com/journal/metals

Metals 2023, 13, 1773 2 of 17

these models, including application needs and accuracy considerations [5–7]. Accurate pair
potentials provide a comprehensive description of energy, geometric structure, and spectral
properties of molecules and serve as the basis for investigating collision and chemical-
reaction dynamics, such as those for atomic collisions. Thus, an in-depth study of potential
pair functions holds significant physical implications and offers broad applications. This
study reveals a direct correlation between the radial distribution function (RDF) and atomic
pair potential. This paper proposes a linkage between the radial distribution function and
the atomic pair potential energy. This is applied to a binary liquid alloy system using a
polynomial expression for the radial distribution function of a pure gas. The group element
activity is estimated based on the symmetries outlined in this paper, utilizing the radial
distribution functions of 36 binary liquid alloys from the literature into the model.

2. Thermodynamic Model and Symmetry

2.1. Miedema Model
In 1973, Miedema et al. introduced a semi-empirical theoretical model that effectively
characterizes the heat of mixing in binary alloys. This model encompasses solid solubil-
ity, the mesostable positioning of metal ions following ion injection, surface aggregation
phenomena, surface energy, the formation energy of single vacancies in metals and metal
compounds, as well as the development of nonphenolic alloys. For any metal binary alloy,
the expression is [8]
2/3 2/3 2 2
2pVmi Vmj × [(q/p)(∆n1/3
ws ) − ( ∆ϕ ) − a (r/p )]
f ij = −1 −1
(1)
(∆n1/3
ws )i + (∆n1/3
ws ) j

xi [1 + µi x j ( ϕi − ϕ j )] x j [1 + µ j xi ( ϕ j − ϕi )]
∆Hm = f ij 2/3 2/3
(2)
xi Vmi [1 + µi x j ( ϕi − ϕ j )] + x j Vmj [1 + µ j xi ( ϕ j − ϕi )]

The activity coefficient is:

  
1 1 µi ( ϕi − ϕ j ) µ j ( ϕ j − ϕi )
− − +

 
 
 
1 − xi 1 + µi (1 − xi )( ϕi − ϕ j ) 1 + µ j xi ( ϕ j − ϕi )
   
  xi  
∆Hm

 
 
 
ln γi = 1 + (1 − x i ) 2/3 2/3 (3)
RT   Vmi [1 + µi (1 − 2xi )( ϕi − ϕ j )] + Vmj [−1 + µ j (1 − 2xi )( ϕ j − ϕi )]
  
−

  
 

   
2/3 2/3
xi Vmi [1 + µi x j ( ϕi − ϕ j )] + x j Vmj [1 + µ j xi ( ϕ j − ϕi )]

 
 
 

In the above equation, xi and x j represent the mole fractions of species i and j respec-
tively. Vmi and Vmj denote the molar volumes of i and j, while (nws )i and (nws ) j stand
for the electron densities, ϕi and ϕ j are electronegative. The variables p,q,r,µi ,µ j and a
are empirical constants. For a binary alloy composed of a transition group metal and a
multivalent non-transition group metal, the formula f ij is transformed to:

2/3 2/3 2 2
2pVmi Vmj × [(q/p)(∆n1/3
ws ) − ( ∆ϕ ) − α (r/p )i (r/p ) j ]
f ij = −1 −1
(4)
(∆n1/3
ws )i + (∆n1/3
ws ) j

2.2. Regular Solution Model (RSM)

Hildebrand proposed the RSM in 1929 [9]. According to this model, the molar excess
volume and mixing entropy are zero, and the molar excess mixing Gibbs free energy is equal
to the molar excess mixing enthalpy [10]. The expression for the molar excess Gibbs free
energy for a binary system is as follows; the activity coefficient of the component i is ln γi .
E
Gm
= ∆Hm = Ωij xi x j (5)
RT

ln γi = Ωij x2j (6)

Metals 2023, 13, 1773 3 of 17

Here, xi and xi are the mole fractions of components i and j, respectively, and Ωij is
the interaction parameter between components i and j. Ωij is only related to temperature
and does not change with the composition of components.

2.3. Wilson Model

In 1964, Wilson [11] introduced a semiempirical and semitheoretical model based on
the local concept. This model assumes that “local concentrations” (represented as volume
fractions) primarily determine molecular interactions. These concentrations are defined in
relation to the Boltzmann distribution probability term for energy. The excess free energy
associated with the concentrations can be expressed as follows.
E
Gm
= − xi ln( xi + A ji x j ) − x j ln( x j + Aij xi ) (7)
RT
!
Aij A ji
ln γi = − ln( xi + Aij x j ) + x j − (8)
xi + Aij x j x j + A ji xi

Here, Aij and A ji are the interaction parameters between components i and j, which
are only related to temperature and do not change with the composition of components [12].

2.4. Nonrandom Two-Liquid (NRTL) Model

The NRTL model, introduced by Renon and Prausnitz in 1968 [13], has been exten-
sively used in correlating thermodynamic data, computing thermodynamic properties, and
predicting phase equilibrium for diverse fluid systems in chemical processes. This model,
based on the concept of local concentration, permits the determination of molar excess
Gibbs free energy. The activity coefficient of the component i is ln γi .
!
GmE τji Gji τij Gij
= xi x j + (9)
RT xi + x j Gji x j + xi Gij

τji G2ji
" #
τij Gij
ln γi = x2j + (10)
( xi + x j Gji )2 ( x j + xi Gij )2

gij − g jj g ji − gii
τij = τji = (11)
RT RT

Gij = exp(−ατij ) Gji = exp(−ατji ) (12)

Here, Gij and Gji are energy parameters characterizing the interaction between compo-
nents i and j; α is related to nonrandomness in the mixture, independent of temperature and
composition of a solution. Moreover, the characteristic parameter of a solution depends on
the solution type. When the mixture is entirely random, many binary system experimental
data show that α varies from 0.2 to 0.47. In this study, α = 0.3.

2.5. Molecular Interaction Volume Model (MIVM)

In 2000, Tao suggested the volume model of molecular interaction [14], which applied
statistical thermodynamics and fluid phase equilibrium theory to describe the motion of
liquid molecules. This model yields the following expression:
! ! !
E
Gm Vmi Vmj xi x j Zi Bji ln Bji Zj Bij ln Bij
= xi ln + x j ln − + (13)
RT xi Vmi + x j Vmj Bji x j Vmj + xi Vmi Bij 2 xi + x j Bji x j + xi Bij
 Metals 2023, 13, 1773 4 of 17

The activity coefficients of the components i is ln γi .

!
Vmi xi Vmi x j Vmi Bij
ln γi = 1 + ln − −
Vmi xi + Vmj Bji x j Vmi xi + x j Vmj Bji Vmj x j + xi Vmi Bij
" # (14)
x j 2 Zi Bji 2 ln Bji Zj Bij ln Bij
− +
2 ( xi + x j Bji )2 ( x j + Bij xi )2

Here, Zi and Zj are the first coordination numbers of i and j pure substances, respec-
tively, and Vmi and Vmj are the molar volumes of i and j, respectively. Bij and Bji are the
interaction parameters of i − j and j − i, respectively. k is the Boltzmann constant, and T is
the temperature.

2.6. Symmetry
Figure 1 shows the molar fraction xi = x j = 0.5 of the two components as the
symmetry axis of their activity curve. The symmetry degree of the activity curve can be
defined as the average absolute value of the activity difference between the two components
at the same concentration. In this context, S represents the measure of symmetry.
s 2023, 13, x FOR PEER REVIEW m 5
∑ ( ai − a j ) x = x
i j
l =1 l
Sij = (15)
m

1.0 1.0
ai ai
aj aj
Axis of symmetry
0.8 0.8 Axis of symmetry
Activity

Activity

0.6 0.6

0.4 0.4

0.2 0.2

0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
i molar fraction i molar fraction
(a) (b)
FigureFigure
1. (a)1. averages
(a) averagesthat theactivity
that the activity
curvecurve is entirely
is entirely symmetric,symmetric, (b)the
(b) averages that averages that the ac
activity curve
curve is asymmetric.
is asymmetric.

The symmetry increases as the S value decreases. At S = 0, the system exhibits

Al-Zn Cu-Sb
1.0 complete symmetry, where ai and a1.0 j denote the experimental activity of components i and
aZn and m represents the number of experimental activities.aThis
j, respectively, Sb definition also
applies toaAlsimilar geometric figures. The system with low symmetry isaCu shown in Figure 2.
0.8 Table 1 presents the symmetry degrees0.8 of the activity curves for the 36 systems based on
the aforementioned definitions.
Activity

Activity

0.6 0.6

0.4 0.4

0.2 0.2
Metals 2023, 13, 1773 5 of 17

(a) (b)
Figure
Figure 2. (a) is2.
an(a) is an
entirely entirely
symmetric symmetric
system, system,
(b) is the system with(b) is thesymmetry
the worst systeminwith the
this paper. worst
symmetry in this paper.
Table 1. Symmetry of the 36 systems from highest to lowest.

System Co-Ni Al-Zn Cu-Ni Al-Ni Cu-Fe Ge-Sn Ag-Cu Pb-Sb Al-Si Al-Co
Sij 0 0 0.0028 0.0034 0.0048 0.007 0.0088 0.0096 0.0102 0.0114
System Li-Mg Sb-Sn Cu-Zr K-Na Pb-Sn Al-Mg Cs-K Li-Na Au-Cu Al-Li
Sij 0.0114 0.0115 0.0119 0.0136 0.0158 0.0177 0.0228 0.0336 0.0364 0.0452
System Nb-Zr Ni-Pd Cu-Mg Al-Ca Ni-Zr Al-Sn Al-Cu Nb-Ni Cu-Sn Au-Si
Sij 0.0473 0.0494 0.0597 0.0724 0.0807 0.0823 0.1116 0.1334 0.1342 0.139
System Li-Sn Fe-Si Ag-In Ge-Te Al-Au Cu-Sb
Sij 0.14 0.1454 0.1483 0.1548 0.160 0.208

3. Pair Potential Energy Polynomials of the Binary Liquid

In an extremely diluted pure gas, the correlation between the interatomic potential
energy and RDF can be expressed as follows [15]:
 ε 
lim g12 (r ) = exp − 12 (16)
ρ0 →0 kT
However, in practical scenarios, the RDF and interatomic potential energy strongly
depend on the number density of the system. The relation between the RDF and pair
potential energy can be expanded using a polynomial expression in terms of the number
density ρ0 : h i
g12 (r, ρ0 , T ) = e−ε/kT 1 + ρ0 g1 ( R, T ) + ρ20 g2 ( R, T ) + · · · (17)

In this case, we consider a pure gas system comprising three molecules labeled 1, 2, and
3 (Figure 3). These molecules afford six atomic distributions: 1–1, 1–2, 1–3, 2–2, 2–3, and 3–3.
Among these distributions, the RDF of 1–2 represents the spatial distribution of molecule
1 around molecule 2. However, g12 (r, ρ0 , T ) is influenced by molecule 3 and is centered
around this molecule, thereby affecting molecules 1 and 2 in g12 (r, ρ0 , T ). Consequently,
 V 3 V 3  V 3  V 3

This knowledge is applied to binary liquid alloys based on the understanding of

atomic interactions in pure gases. The atomic distribution in the binary liquid system is
characterized by three scenarios: i −i , i − j , and j − j . The RDF i − j describes the distri-
Metals 2023, 13, 1773 6 of 17
bution of moleculei around molecule j in a binary system (Figure 1). However, the
remaining molecule i influences the i molecule in the RDF i − j , while the remaining
based onj these
molecule affects the j molecule
observations, a connection − j . Unlikebetween
RDFbe iestablished
in the can pure gasthecentered
RDF andonpair
the re-
potential energy.
maining molecules 3 affecting the RDF 1–2. In a binary system, the spatial positioning of
(r, ρ0 , T ) =i e−and
the remaining gmolecules ε 12 /kT jexp
influences
ρ0 [e−ε 13the − 1][e−i ε−23 /kT
/kT RDF j for
− 1]molecules i and j .
n R o
12 dr3 V
That is, the i moleculenin the RDF affecting i − j is centered o on the other i molecules,
(18)
= e−ε 12 /kT 1 + ρ0 V [e−ε 13 /kT − 1][e−ε 23 /kT − 1]dr3
R
while the j molecule in the RDF affecting i − j is centered on the other j molecules.

(a) (b)
Figure 3. 3.
Figure (a)(a)
is is
thetheatomic
atomicdistribution diagramofofpure
distribution diagram pure substance,
substance, (b) (b) is atomic
is the the atomic distribution
distribution
diagram of the binary system.
diagram of the binary system.

[e−ε 13 /kT − 1][e−ε 23 /kT − 1]dr3 .

R
Consequently, the expansion based on ρ0 is 1 + ρ0 V
Equation (18) is compared with Equation (17).

g1 (r12 , T ) = V [e−ε 13 /kT − 1][e−ε 23 /kT − 1]dr3 = V [e−ε 13 /kT e−ε 23 /kT − e−ε 23 /kT − e−ε 13 /kT + 1]dr3
R R
(19)
= V e−ε 13 /kT e−ε 23 /kT dr3 − V e−ε 23 /kT dr3 − V e−ε 13 /kT dr3 + V dr3
R R R R

This knowledge is applied to binary liquid alloys based on the understanding of

atomic interactions in pure gases. The atomic distribution in the binary liquid system
is characterized by three scenarios: i − i, i − j, and j − j. The RDF i − j describes the
distribution of molecule i around molecule j in a binary system (Figure 1). However,
the remaining molecule i influences the i molecule in the RDF i − j, while the remaining
molecule j affects the j molecule in the RDF i − j. Unlike pure gas centered on the remaining
molecules 3 affecting the RDF 1–2. In a binary system, the spatial positioning of the
remaining molecules i and j influences the RDF i − j for molecules i and j. That is, the i
molecule in the RDF affecting i − j is centered on the other i molecules, while the j molecule
in the RDF affecting i − j is centered on the other j molecules.
If the RDF in a binary liquid system can be expanded as a polynomial based on number
density, then Equation (18) takes the following form:

e−ε ii /kT dr × e−ε jj /kT dr − e−ε ii /kT dr − e−ε jj /kT dr + ( (20)

R R R R R R
g1 (rij , T ) = V V V V V drii + V dr jj )

Suppose that the subradial distribution function g1 (rij , T ) of the principal RDF gij (r, ρ0 , T )
conforms to ρ0 → 0 , i.e., i − j is the primary RDF, then i − i and j − j conform to ρ0 → 0 and
the subradial distribution function is:
 
 ε 
ii ε jj
gii (r ) = exp − g jj (r ) = exp − (21)
kT kT
Metals 2023, 13, 1773 7 of 17

Substituting Equation (21) into Equation (20), we obtain:

Z Z Z Z Z Z 
g1 (rij , T ) = gii (r )dr × g jj (r )dr − gii (r )dr − g jj (r )dr + drii + dr jj (22)
V V V V V V

Then substituting Equation (22) into Equation (18), we obtain:

( " R R R #)
−ε ij /kT V gii (r ) dr × V g jj (r ) dr − V gii (r ) dr
gij (r, ρ0 , T ) = e exp ρ0 R R R
− V g jj (r )dr + ( V drii + V dr jj )
( " R R R #) (23)
−ε ij /kT V gii (r ) dr × V g jj (r ) dr − V gii (r ) dr
=e 1 + ρ0 R R R
− V g jj (r )dr + ( V drii + V dr jj )

The relation between the RDF and potential energy can be obtained using Equation (23):

ε ij gij (r, ρ0 , T )
− = ln ( " R R R (24)
kT
#)
V gii (r ) dr × V g jj (r ) dr − V gii (r ) dr
1 + ρ0 R R R
− V g jj (r )dr + ( V drii + V dr jj )
Pair potential energy between molecules can be accurately calculated using the RDF.
This function represents the ratio of the probability of finding another molecule at a distance
r to the random distribution [16]. In the double distribution function, for a system with
N molecules and volume V, the probability of a molecule appearing in the element dri
is ( N/V )dri , the probability of a molecule appearing at the distance dr j is ( N/V )dr j , and
the probability of atomic pairs appearing at a distance r is ( N/V )2 dri dr j . The double
distribution function is given as follows:

N 2
p(2) (r )dri dr j = ( ) dri dr j (25)
V
In the system, the average potential energy ε between each molecule is:
x
ε= ε ij (r ) p(2) (r )dri dr j (26)
V

However, in the RDF of binary systems, the probability of having molecule i in dri
at ri and molecule j in dr j at r j is p(2) (r )dri dr j . The potential energy is ε, and the average
value of ε is the sum of all possible times of the probabilities:

1 s 1 R 1R
ε ij (r ) gij (r )4πr2 dr
R
ε ij = ε ij (r ) gij (r )dri dr j = 2 dri ε ij (r ) gij (r )dr j =
V2 V V V V
(27)
4π ε ij (r ) gij (r )r2 dr ε ij (r ) gij (r )r2 dr
R R
= R = R
4π r2 dr r2 dr

Substituting Equation (24) into Equation (27) yields the potential energy between
molecules:
 

 


 

g ( r )

 

R 2 ij
gij (r )r ln " R R R # dr
V gii (r ) dr × V g jj (r ) dr − V gii (r ) dr

 



 1 + ρ 0 R R R



ε ij

− V g jj (r )dr + ( V drii + V dr jj ) 
− = R (28)
kT r2 dr

The peak value of the RDF signifies the disparity between the local and bulk molar
fractions. As the mole fraction increases, the contributions of the second and third RDF
peaks diminish while the contribution of the first peak amplifies. The pair potential energy
 peaks diminish while the contribution of the first peak amplifies. The pair potential ene
is then calculated by using Equation (28) and the area of the first peak of the skewed R
The area under the first peak was computed through graphical integration, as depicte
Metals 2023, 13, 1773
Figure 4. Notably, this approach differs from Wangʹs utilization of the L-PPDF 8 of 17
mathem
ical form with Gaussian fitting, which relies on fitting parameters u and v [17,18].
important to emphasize that this study does not incorporate any fitting parameters.
RDFisused in this study
then calculated is defined
by using Equationby three
(28) and thekeyarea
coordinates: r0 (which
of the first peak represents
of the skewed RDF. the st
The area under the first peak was computed through graphical integration, as depicted in
ing point of g(r) before reaching zero), rm (the transverse coordinate of the first pea
Figure 4. Notably, this approach differs from Wang’s utilization of the L-PPDF mathematical
g(r)),form r1 Gaussian
andwith (the transverse coordinate
fitting, which of the
relies on fitting first valley
parameters ofvg(r)).
u and The
[17,18]. It isasymmetric
important met
of calculating the RDF involves integrating the area between r0 and r1 , while the s
to emphasize that this study does not incorporate any fitting parameters. The RDF used in
this study is defined by three key coordinates: r0 (which represents the starting point of
metric method
g(r) before involves
reaching zero), rintegrating twice the area between r0 g(r)),
m (the transverse coordinate of the first peak of
andand rm r (Figure 4).
1 (the
trapezoidal
transversemethod
coordinate [19] is first
of the used to compute
valley of g(r)). Thethese areas method
asymmetric (Equation (29)). Table
of calculating the 2 lists
RDF involves integrating the area between r
references for the partial RDFs of 36 binary liquid alloys.
0 and r 1 , while the symmetric method involves
integrating twice the area between r0 and rm (Figure 4). The trapezoidal method [19] is
r0
used to compute these b − aareas
N (Equation (29)). Table b −2 alists the references for the partial RDFs
g ( r ) liquid
of 36 binary
2N

dr ≈ alloys. [ g ( rn ) + g ( rn +1 ) ] =
n =1
[ g ( r ) + 2 g ( r2 ) + ....... + 2 g ( rN ) + g ( rN +1 ) ]
2N
r1
Rr0 b− a
N
b− a
g(r )dr ≈ 2N ∑ [ g(rn ) + g(rn+1 )] = 2N [ g (r ) + 2g (r2 ) + . . . . . . . +2g (r N ) + g(r N +1 )] (29)
r1 n =1

2.5

2
g(r)
g(r)

0
r0 r1 0.0 r0 rm
0
r 5 0
r 5

(a) (b)
Figure 4. 4.(a)
Figure (a) is illustration
is illustration ofnon-symmetric
of the the non-symmetric method of
method of integration ofradial
integration offunctions,
distribution radial distribu
functions, (b) representation
(b) graphical graphical representation of theofsymmetric
of the symmetric method integration of method of integration
a radial distribution function. of a ra
distribution function.
Table 2. Partial radial distribution function of binary liquid alloys references.

Co-Ni Al-Zn Cu-Ni Al-Ni Cu-Fe Ge-Sn Ag-Cu Pb-Sb Al-Si Al-Co
System
[20] [21] [22] [23] [24] [25] [23] [26] [27] [28]
Li-Mg Sb-Sn Cu-Zr K-Na Pb-Sn Al-Mg Cs-K Li-Na Au-Cu Al-Li
System
[29] [30] [31] [32] [33] [34] [35] [36] [37] [38]
Nb-Zr Ni-Pd Cu-Mg Al-Ca Ni-Zr Al-Sn Al-Cu Nb-Ni Cu-Sn Au-Si
System
[39] [40] [41] [42] [43] [44] [45] [46] [47] [48]
Li-Sn Fe-Si Ag-In Ge-Te Al-Au Cu-Sb
System
[49] [50] [51] [52] [53] [54]

The RSM has a tunable parameter Ωij . The average coordination number Z is obtained
using the pure coordination number of the two components. Additionally, the temperature
T documented in the literature and the calculated parameter Ωij can be referred to calculate
the parameter Ωij 0 at the desired temperature T 0 .
1
Z= ( Z + Zj ) (30)
2 i
Metals 2023, 13, 1773 9 of 17

  
1
Ωij = kT Z ε ij − (ε ii + ε jj ) (31)
2

T ln Ωij = T 0 ln Ωij 0 (32)

The parameters Aij and A ji of Wilson equation are given. Additionally, the values of
parameters Aij 0 and Aij 0 at the desired temperature T 0 can be obtained using the tempera-
ture T from the literature that was employed to calculate the corresponding parameters Aij
and A ji .
ε ij − ε jj Vj ε ji − ε ii
   
Vi
Aij = exp − A ji = exp − (33)
Vj RT Vi RT

T ln Aij = T 0 ln Aij 0 T ln A ji = T 0 ln Aij 0 (34)

The NRTL has two parameters τij and τji . These parameters obtained using the
temperature T mentioned in the literature are employed to calculate the parameters τij 0
and τij 0 , respectively, at the required temperature T 0 .
ε ij − ε jj ε ji − ε ii
τij = − τji = − (35)
kT kT

T ln τij = T 0 ln τij 0 T ln τji = T 0 ln τij 0 (36)

The MIVM involves parameters Bij and Bji , representing the interaction parameters
for i–j and j–i interactions, respectively. Using Bij and Bji values obtained at temperature T,
the corresponding parameters Bij 0 and Bij 0 can be calculated at the desired temperature T 0 .

ε ij − ε jj ε ji − ε ii
   
Bij = exp − Bji = exp − (37)
kT kT

T ln Bij = T 0 ln Bij 0 T ln Bji = T 0 ln Bij 0 (38)

4. Result Analysis
4.1. Asymmetric Method for Calculating the RDF
The asymmetric method uses the area between r0 and r1 presented in Figure 4 and
Equation (28) to determine the parameters for each model, see Table 3. Table 4 demonstrates
that among the first 12 systems, the RSM performs better than other models for six systems.
The average standard deviation (SD) is 0.027, and the average relative deviation (ARD) is
7.7%. In the case of the 12 systems with moderate symmetry, the RSM outperforms the
other three models in six systems, resulting in an average SD of 0.059 and an average ARD
of 13.83%. Notably, the MIVM also performs well, with an average SD of 0.091 and an
average ARD of 25.7%. For the 12 systems with low symmetry, the RSM outperforms the
other three models in three systems, yielding an average SD of 0.101 and an average ARD
of 39.5%. Additionally, the MIVM outperforms the other three models in five systems
when the systems have even lower symmetry, with an average SD of 0.131 and an average
ARD of 45.3%. Considering the average performance across all 36 binary liquid alloy
systems, the Wilson and NRTL models exhibit the poorest performance, displaying larger
SD and ARD values than the other models. As shown in Figure 5, the SD and ARD values
generally increase with decreasing symmetry, albeit not significantly. Further analysis of
high, medium, and low-symmetry systems reveals a strong correlation between the RSM
and symmetry.
Metals 2023, 13, x FOR PEER REVIEW 12 of 17
Metals 2023, 13, 1773 10 of 17

Miedema MIVM
MIVM RSM
RSM
0.8
Wilson 4
NRTL
0.6

ARD
SD

0.4

0.2

0.0

system 0
system
(a) (b)

80
Wilson
NRTL
Miedema

60
ARD

40

20

system

(c)
Figure 5. 5.
Figure (a)(a)
is isthe
theSD
SDof
of36
36 systems, (b)isisthe
systems, (b) theARD
ARDofof MIVM
MIVM andand
RSM RSM models
models of 36of 36 systems,
systems, (c) is (c) is
thethe
ARD of the Wilson equation, NRTL equation, and Miedema of
ARD of the Wilson equation, NRTL equation, and Miedema of 36 systems. 36 systems.

4.2.Table
Symmetric
3. ModelMethod for of
parameters Calculating the RDF
the non-symmetric method.

MIVM
The methodology RSM
based on symmetry involves deriving parameters
Wilson NRTL
for each model
System using the region from0 r0 to0 r1 (Figure 4) and Equation (28), see0 Table 5. Analysis of data
Bij 0 Bij 0 Ωij = Ωij Aij 0 Aij 0 τij τij 0
presented in Table 6 reveals that among the 12 systems characterized by high symmetry,
Co-Ni 0.964 0.977 0.343 0.964 0.977 −0.036 −0.023
Al-Zn 1.246 the MIVM0.743outperforms the other three
0.423 1.166 models with
0.795an average SD of 0.127 −
0.220 and
0.297 an average
Cu-Ni 1.086 ARD of0.863
30.5%. The RSM0.374exhibits an average SD 0.938
0.999 of 0.111 and an average ARD
0.082 of 44.6%. In
−0.147
Al-Ni 1.493 1.649
the 12 systems −5.203
with medium 0.991 the Miedema
symmetry, 2.485 outperforms
0.401 the other0.500
three models,
Cu-Fe 0.943 0.805 1.512 0.943 0.805 −0.059 −0.217
Ge-Sn 0.740
yielding1.273
a average SD of 0.067 and
0.266
an ARD of 0.776
1.214
22.8%. By contrast,
−0.302 the MIVM 0.241
exhibits a
Ag-Cu 0.675 average1.194
SD of 0.118 and
1.225an ARD of 36.7%,
0.471 and the RSM
1.709 exhibits a
−0.394 average SD
0.177 0.088 and
of
Pb-Sb 1.558 0.517 1.046 1.457 0.553 0.443
an ARD of 32.8%. For the 12 systems with low symmetry, the Miedema surpasses − 0.660 the other
Al-Si 1.405 1.077 −1.856 1.226 1.235 0.340 0.074
three models. The RSM exhibits an average SD of 0.116 and an average ARD of 70.2%.
Al-Co 1.112 1.884 −4.213 0.802 2.612 0.106 0.633
Li-Mg 1.357 Each model
0.914 exhibits−distinct
1.095 performance
1.469 characteristics
0.844 depending
0.305 on the−system
0.090 repre-
Sb-Sn 0.769 sentation. As shown−in0.847
1.545 Figure 6, the0.735
data comparison
1.618 indicates an increasing0.435
−0.262 trend in ARD
Cu-Zr 0.908 −2.277
1.669 decreasing
values with symmetry.1.782 0.851 −0.096 0.512
K-Na 0.700 1.176 1.018 0.366 2.252 −0.357 0.162
Pb-Sn 1.044 0.912 0.267 0.932 1.021 0.043 −0.092
Al-Mg 0.820 Table 5. 1.123
Parameters of each
0.459 model of the weighting method.
1.162 0.793 −0.198 0.116
MIVM RSM Wilson NRTL
System
Bij ' Bji ' Ωij ' = Ω ji ' Aij ' Aji ' τ ij ' τ ji '
Co-Ni 1.089 0.900 0.118 1.089 0.900 0.085 −0.106
Al-Zn 1.317 0.938 −1.162 1.232 1.003 0.275 −0.064
Metals 2023, 13, 1773 11 of 17

Table 3. Cont.

MIVM RSM Wilson NRTL

System
Bij 0 Bij 0 0
Ωij = Ωij 0 Aij 0 Aij 0 τij 0 τij 0
Cs-K 1.204 0.635 1.310 0.801 0.955 0.185 −0.454
Li-Na 0.764 0.999 1.347 1.416 0.539 −0.270 −0.001
Au-Cu 1.163 1.444 −2.877 0.812 2.068 0.151 0.368
Al-Li 1.136 1.379 −2.356 1.485 1.055 0.128 0.321
Nb-Zr 0.814 1.255 −0.110 1.048 0.975 −0.206 0.227
Ni-Pd 1.168 1.087 −1.307 1.590 0.798 0.153 0.080
Cu-Mg 1.312 1.258 −2.781 2.575 0.641 0.272 0.230
Al-Ca 2.202 1.259 −5.761 5.826 0.476 0.789 0.230
Ni-Zr 1.522 1.730 −5.373 3.247 0.811 0.420 0.548
Al-Sn 0.624 1.440 0.598 1.024 0.877 −0.471 0.365
Al-Cu 1.568 1.530 −5.209 2.192 1.152 0.476 0.450
Nb-Ni 1.772 1.071 −3.558 1.070 1.774 0.572 0.069
Cu-Sn 1.987 0.849 −2.904 0.874 1.932 0.687 −0.163
Au-Si 1.224 1.684 −3.129 1.034 1.993 0.202 0.521
Li-Sn 3.396 2.186 −10.225 4.263 1.742 1.223 0.782
Fe-Si 5.678 1.776 −9.822 4.695 2.148 1.737 0.574
Ag-In 1.595 0.925 −2.227 2.476 0.596 0.467 −0.078
Ge-Te 0.409 1.891 1.285 0.912 0.848 −0.894 0.637
Al-Au 1.680 1.660 −5.740 2.287 1.219 0.519 0.507
Cu-Sb 1.804 0.881 −2.315 0.757 2.098 0.590 −0.127

Table 4. Deviation and relative error of each model in the non-symmetric method.

Miedema MIVM RSM Wilson NRTL

System
SD ARD/% SD ARD/% SD ARD/% SD ARD/% SD ARD/%
Co-Ni 0.013 3.852 0.036 10.8 0.029 8.8 0.003 0.8 0.015 4.4
Al-Zn 0.075 15.536 0.108 22 0.058 12.1 0.100 20.5 0.116 23.7
Cu-Ni 0.247 49.697 0.085 16.5 0.085 16.5 0.120 23.1 0.134 25.7
Al-Ni 0.210 1467.533 0.044 18.3 0.037 125 0.208 1472 0.398 4520
Cu-Fe 0.889 93.800 0.106 12.5 0.128 15.1 0.321 37.5 0.375 44
Ge-Sn 0.019 5.384 0.071 19.4 0.011 3.1 0.016 4.1 0.027 7.5
Ag-Cu 0.108 19.924 0.062 10.8 0.025 4.6 0.143 26.1 0.167 30.5
Pb-Sb 0.438 79.951 0.044 9.9 0.047 10.5 0.070 15.9 0.080 18.1
Al-Si 0.023 10.017 0.110 40.5 0.058 23.2 0.041 18.5 0.125 63.3
Al-Co 0.034 10.163 0.080 39.2 0.027 10.5 0.142 341 0.300 1085
Li-Mg 0.043 17.882 0.056 20.7 0.032 12.3 0.035 14.2 0.082 34.9
Sb-Sn 0.025 6.367 0.078 29.3 0.012 3.6 0.054 24.5 0.093 44.2
Cu-Zr 0.124 53.790 0.064 31.6 0.015 8.4 0.098 74.4 0.185 162
K-Na 0.019 3.619 0.081 15.3 0.012 2 0.144 27.8 0.145 28
Pb-Sn 0.015 1.847 0.016 3.2 0.009 1.7 0.063 14.3 0.087 19.6
Al-Mg 0.042 14.076 0.087 32.9 0.092 34.9 0.048 17 0.033 11.2
Cs-K 0.020 4.475 0.129 36 0.148 41.7 0.016 3.3 0.066 16.9
Li-Na 0.255 24.451 0.210 22.9 0.192 21.3 0.364 40.5 0.405 45.2
Au-Cu 0.028 7.697 0.076 34.5 0.038 12.1 0.105 101 0.219 255
Al-Li 0.034 17.993 0.048 17 0.037 12.2 0.118 115 0.209 236
Nb-Zr 0.036 8.347 0.089 21.1 0.066 15.1 0.058 13 0.056 12.4
Ni-Pd 0.115 72.225 0.044 15 0.039 14.1 0.090 53.7 0.140 91
Cu-Mg 0.022 8.778 0.054 30.2 0.038 12.4 0.104 103 0.226 286
Al-Ca 0.100 56.098 0.129 64.3 0.087 37.4 0.094 144 0.338 1081
Ni-Zr 0.080 20.755 0.076 80.2 0.085 266 0.225 1661 0.433 4415
Al-Sn 0.122 18.737 0.226 34.4 0.142 21.3 0.199 30.4 0.226 34.5
Al-Cu 0.099 228.621 0.076 43 0.075 30.1 0.158 516 0.343 2031
Nb-Ni 0.159 517.211 0.128 47.2 0.089 64 0.161 540 0.288 1478
Cu-Sn 0.073 38.233 0.131 51.2 0.103 33.3 0.135 48 0.524 544
Metals 2023, 13, 1773 12 of 17

Table 4. Cont.

Miedema MIVM RSM Wilson NRTL

System
SD ARD/% SD ARD/% SD ARD/% SD ARD/% SD ARD/%
Au-Si 0.106 43.309 0.116 44 0.090 45.6 0.147 279 0.290 847
Li-Sn 0.081 159.055 0.098 59.1 0.117 130 0.216 1347 0.596 7906
Fe-Si 0.095 67.563 0.259 82.9 0.137 58.3 0.088 186 0.516 5003
Ag-In 0.114 97.817 0.077 32.8 0.095 30.3 0.095 76 0.185 191
Ge-Te 0.240 319.161 0.099 27.3 0.333 496 0.195 239 0.150 148
Al-Au 0.159 53.722 0.149 55.9 0.139 42.7 0.168 405 0.333 1822
Cu-Sb 0.095 80.468 0.072 28 0.096 30.4 0.122 104 0.196 229
Ave 0.121 102.7 0.095 32.2 0.078 47.4 0.124 226 0.225 911
r
∑ (a
pre −a
exp )2 1 a pre − aexp
SD = N ; ARD = N∑ aexp × 100%. a pre -calculated value of the activity; aexp [55–59]-
experimental activity value.

4.2. Symmetric Method for Calculating the RDF

The methodology based on symmetry involves deriving parameters for each model
using the region from r0 to r1 (Figure 4) and Equation (28), see Table 5. Analysis of data
presented in Table 6 reveals that among the 12 systems characterized by high symmetry,
the MIVM outperforms the other three models with an average SD of 0.127 and an average
ARD of 30.5%. The RSM exhibits an average SD of 0.111 and an average ARD of 44.6%. In
the 12 systems with medium symmetry, the Miedema outperforms the other three models,
yielding a average SD of 0.067 and an ARD of 22.8%. By contrast, the MIVM exhibits a
average SD of 0.118 and an ARD of 36.7%, and the RSM exhibits a average SD of 0.088
and an ARD of 32.8%. For the 12 systems with low symmetry, the Miedema surpasses
the other three models. The RSM exhibits an average SD of 0.116 and an average ARD of
70.2%. Each model exhibits distinct performance characteristics depending on the system
representation. As shown in Figure 6, the data comparison indicates an increasing trend in
ARD values with decreasing symmetry.

Table 5. Parameters of each model of the weighting method.

MIVM RSM Wilson NRTL

System
Bij 0 Bji 0 Ωij 0 =Ωji 0 Aij 0 Aji 0 τij 0 τji 0
Co-Ni 1.089 0.900 0.118 1.089 0.900 0.085 −0.106
Al-Zn 1.317 0.938 −1.162 1.232 1.003 0.275 −0.064
Cu-Ni 1.391 0.877 −1.138 1.280 0.953 0.330 −0.132
Al-Ni 1.155 2.099 −5.114 0.766 3.163 0.144 0.742
Cu-Fe 0.890 1.145 −0.103 0.890 1.145 −0.117 0.135
Ge-Sn 0.562 1.573 0.545 0.923 0.958 −0.576 0.453
Ag-Cu 1.309 0.582 1.538 0.914 0.833 0.269 −0.542
Pb-Sb 1.231 0.810 0.017 1.151 0.866 0.207 −0.211
Al-Si 1.534 1.140 −2.499 1.338 1.307 0.428 0.131
Al-Co 0.830 1.774 −2.218 0.551 2.673 −0.186 0.573
Li-Mg 1.087 0.988 −0.363 1.177 0.912 0.083 −0.012
Sb-Sn 0.825 1.576 −1.288 0.788 1.650 −0.192 0.455
Cu-Zr 1.206 1.206 1.206 1.206 1.206 1.206 1.206
K-Na 0.562 1.364 1.390 0.293 2.613 −0.577 0.311
Pb-Sn 1.601 0.830 −1.549 1.430 0.929 0.471 −0.186
Al-Mg 0.808 1.126 0.528 1.144 0.796 −0.213 0.119
Cs-K 1.642 0.660 −0.391 1.092 0.992 0.496 −0.416
Li-Na 0.728 1.069 1.245 1.350 0.577 −0.317 0.067
Au-Cu 2.423 0.515 −1.229 1.692 0.737 0.885 −0.664
Al-Li 0.813 1.374 −0.580 1.062 1.051 −0.208 0.318
Metals 2023, 13, 1773 13 of 17

Table 5. Cont.

MIVM RSM Wilson NRTL

System
Bij 0 Bji 0 Ωij 0 =Ωji 0 Aij 0 Aji 0 τij 0 τji 0
Nb-Zr 0.671 1.414 0.279 0.864 1.098 −0.399 0.346
Ni-Pd 1.174 1.105 11.250 −1.445 1.599 0.811 0.159
Cu-Mg 0.882 1.538 −1.689 1.731 0.783 −0.126 0.430
Al-Ca 2.423 0.515 −1.229 1.692 0.737 0.885 −0.664
Ni-Zr 2.150 4.152 −12.151 4.587 1.947 0.766 1.424
Al-Sn 0.649 1.393 0.566 1.065 0.849 −0.433 0.332
Al-Cu 1.963 1.534 −6.647 1.472 2.180 0.714 0.453
Nb-Ni 3.602 1.064 −7.456 2.174 1.763 1.281 0.062
Cu-Sn 4.758 0.617 −5.972 2.091 1.403 1.560 −0.484
Au-Si 1.438 0.875 −0.992 1.214 1.036 0.363 −0.134
Li-Sn 2.077 2.672 −8.741 2.608 2.129 0.731 0.983
Fe-Si 1.534 1.843 −4.414 1.268 2.228 0.428 0.611
Ag-In 1.066 1.467 −2.560 1.654 0.946 0.064 0.384
Ge-Te 0.636 3.025 −3.272 1.419 1.356 −0.452 1.107
Al-Au 1.903 0.844 −2.370 0.799 2.011 0.643 −0.169
Cu-Sb 1.743 1.017 −3.207 1.743 1.017 0.556 0.017

Table 6. Model deviations and relative errors of weighting methods.

Miedema MIVM RSM Wilson NRTL

System
SD ARD/% SD ARD/% SD ARD/% SD ARD/% SD ARD/%
Co-Ni 0.013 3.852 0.002 0.4 0.004 1.1 0.007 2.1 0.011 3.3
Al-Zn 0.075 15.536 0.233 45.8 0.201 40 0.128 26 0.085 17.5
Cu-Ni 0.247 49.697 0.252 46.9 0.220 41.3 0.147 28.3 0.107 20.8
Al-Ni 0.210 1467.533 0.081 31.7 0.039 133 0.257 2130 0.335 3370
Cu-Fe 0.889 93.800 0.365 42.8 0.359 42 0.351 41.2 0.347 40.7
Ge-Sn 0.019 5.384 0.135 35.4 0.045 13.2 0.007 1.6 0.039 10.8
Ag-Cu 0.108 19.924 0.063 10.2 0.087 16.3 0.115 21.2 0.175 31.8
Pb-Sb 0.438 79.951 0.054 17.9 0.164 62.6 0.038 13.3 0.004 0.9
Al-Si 0.023 10.017 0.147 50.8 0.089 34.2 0.027 12.1 0.142 72.6
Al-Co 0.034 10.163 0.074 39.6 0.070 130 0.161 411 0.258 852
Li-Mg 0.043 17.882 0.020 7.9 0.026 10.2 0.052 21.7 0.067 28.5
Sb-Sn 0.025 6.367 0.100 36.5 0.029 11.3 0.046 20.7 0.103 49.6
Cu-Zr 0.124 53.790 0.065 33 0.026 14.4 0.084 62.1 0.193 171
K-Na 0.019 3.619 0.130 24.3 0.076 15.2 0.154 29.6 0.156 29.9
Pb-Sn 0.015 1.847 0.244 51.4 0.194 42 0.105 23.7 0.049 11
Al-Mg 0.042 14.076 0.095 36.2 0.101 38.4 0.049 17.5 0.031 10.7
Cs-K 0.020 4.475 0.132 32.3 0.074 18.8 0.046 11.5 0.036 8.9
Li-Na 0.255 24.451 0.244 26.7 0.211 23.4 0.363 40.4 0.404 45.1
Au-Cu 0.028 7.697 0.117 40.7 0.067 58.8 0.135 137 0.168 181
Al-Li 0.034 17.993 0.093 86.2 0.110 106 0.151 156 0.171 182
Nb-Zr 0.036 8.347 0.092 21.6 0.032 6.3 0.052 11.5 0.065 14.7
Ni-Pd 0.115 72.225 0.046 15.2 0.036 11.3 0.088 52.4 0.143 93.2
Cu-Mg 0.022 8.778 0.065 22.3 0.051 40 0.136 144 0.202 245
Al-Ca 0.100 56.098 0.089 50.3 0.075 20 0.107 175 0.301 901
Ni-Zr 0.080 20.755 0.248 75.6 0.098 35.8 0.140 86.2 0.573 6633.0
Al-Sn 0.122 18.737 0.216 32.8 0.147 21.9 0.200 30.6 0.224 34.3
Al-Cu 0.099 228.621 0.113 57.6 0.078 39.9 0.114 289 0.417 2930
Nb-Ni 0.159 517.211 0.216 67.7 0.104 49.9 0.121 302 0.363 2270
Cu-Sn 0.073 38.233 0.255 75.4 0.163 56.7 0.076 28 0.200 167
Au-Si 0.106 43.309 0.070 79 0.136 240 0.192 425 0.234 591
Li-Sn 0.081 159.055 0.150 86.9 0.116 187 0.244 1730 0.564 7420
Fe-Si 0.095 67.563 0.138 45.8 0.105 79.9 0.165 693 0.358 2900
 Al-Li 0.034 17.993 0.093 86.2 0.110 106 0.151 156 0.171 182
Nb-Zr 0.036 8.347 0.092 21.6 0.032 6.3 0.052 11.5 0.065 14.7
Ni-Pd 0.115 72.225 0.046 15.2 0.036 11.3 0.088 52.4 0.143 93.2
Cu-Mg 0.022 8.778 0.065 22.3 0.051 40 0.136 144 0.202 245
Al-Ca 0.100 56.098 0.089 50.3 0.075 20 0.107 175 0.301 901
Metals 2023, 13, 1773 Ni-Zr 0.080 20.755 0.248 75.6 0.098 35.8 0.140 86.2 0.573 6633.0
14 of 17
Al-Sn 0.122 18.737 0.216 32.8 0.147 21.9 0.200 30.6 0.224 34.3
Al-Cu 0.099 228.621 0.113 57.6 0.078 39.9 0.114 289 0.417 2930
Nb-Ni 0.159 517.211 0.216 67.7 0.104 49.9 0.121 302 0.363 2270
Table 6. Cont.
Cu-Sn 0.073 38.233 0.255 75.4 0.163 56.7 0.076 28 0.200 167
Miedema Au-Si 0.106
MIVM 43.309 0.070 RSM 79 0.136 240 Wilson
0.192 425 0.234
NRTL 591
System Li-Sn 0.081 159.055 0.150 86.9 0.116 187 0.244 1730 0.564 7420
SD ARD/% SD ARD/% SD ARD/% SD ARD/% SD ARD/%
Fe-Si 0.095 67.563 0.138 45.8 0.105 79.9 0.165 693 0.358 2900
Ag-In 0.114 97.817Ag-In 0.1220.114 37.9
97.817 0.097 37.9 30.10.097
0.122 0.109
30.1 90.8 90.8
0.109 0.193
0.193 204
204
Ge-Te 0.240 319.161Ge-Te 0.2450.240 73.9
319.161 0.114 73.9 29.10.114
0.245 0.135
29.1 118 118
0.135 0.211
0.211 268
268
Al-Au 0.159 53.722Al-Au 0.1490.159 55.9
53.722 0.139
0.149 55.9 42.70.139 0.168
42.7 405
0.168 405 0.333
0.333 1820
1820
Cu-Sb 0.095 80.468 0.072 29.9 0.097 30 0.122 105 0.196 228
Cu-Sb 0.095 80.468 0.072 29.9 0.097 30 0.122 105 0.196 228
Ave 0.121 102.7 Ave 0.137 0.121 42.35
102.7 0.105 42.3549.24
0.137 0.105 0.128
49.24 219.2 219.2
0.128 0.207
0.207 884.9
884.9

Miedema
MIVM MIVM
0.8 RSM RSM
Wilson
NRTL
2
0.6

ARD
SD

0.4

0.2
0

0.0

system
system
(a) (b)

Wilson
NRTL
Miedema
60
ARD

40

20

system

(c)
Figure 6. 6.
Figure (a)(a)
is is
thethe
SDSD
of of
3636
systems, (b)(b)
systems, is is
thethe
ARD of of
ARD MIVM and
MIVM RSM
and models
RSM of of
models 3636
systems, (c)(c)
systems, is is
the ARD of the Wilson equation, NRTL equation, and Miedema of 36 systems.
the ARD of the Wilson equation, NRTL equation, and Miedema of 36 systems.

5. Conclusions
This study uses polynomials to describe the partial RDF in pure gases and extends
this approach to binary liquid systems. The primary aim of this study is to characterize the
atomic distribution and unravel interatomic interactions, which are essential for accurate
thermodynamic calculations. The RDF exhibits irregularities when the symmetric method
is used instead of the asymmetric one for RDF calculation. Notably, the estimation of binary
liquid alloy activity favors using the asymmetric method, especially when considering
the average results obtained from both methods for the 36 binary liquid alloy systems
investigated in this study. We selected and compared five thermodynamic models based on
their symmetry degree. Data analysis reveals that the RSM exhibits the highest dependency
Metals 2023, 13, 1773 15 of 17

on the symmetry degree. Conversely, the MIVM demonstrates superior adaptability to

symmetric and asymmetric systems.

Author Contributions: D.T.: Theoretical guidance, review; J.H.: Conceptualization, Writing—original

draft, Writing—review & editing. All authors have read and agreed to the published version of
the manuscript.
Funding: This work was financially supported by the National Natural Science Foundation of China
under Grant No. 51464022.
Data Availability Statement: Data available on request from the authors.
Conflicts of Interest: The authors declare that they have no known competing financial interest or
personal relationship that could have appeared to influence the work reported in this paper.

References
1. Xie, J.C.; Kar, T.; Xie, R.-H. An Accurate Pair Potential Function for Diatomic Systems. Chem. Phys. Lett. 2014, 591, 69–77.
[CrossRef]
2. Lennard-Jones, J.E. Cohesion. Proc. Phys. Soc. 1931, 43, 461. [CrossRef]
3. Morse, P.M. Diatomic Molecules According to the Wave Mechanics. II. Vibrational Levels. Phys. Rev. 1929, 34, 57–64. [CrossRef]
4. Born, M.; Mayer, J.E. Zur Gittertheorie der Ionenkristalle. Z. Phys. 1932, 75, 1–18. [CrossRef]
5. Ter Horst, M.A.; Schatz, G.C.; Harding, L.B. Potential Energy Surface and Quasiclassical Trajectory Studies of the CN+H2 Reaction.
J. Chem. Phys. 1996, 105, 558–571. [CrossRef]
6. Hirst, D.M. Ab Initio Potential Energy Surfaces for Excited States of the NO2+ Molecular Ion and for the Reaction of N+ with O2 .
J. Chem. Phys. 2001, 115, 9320–9330. [CrossRef]
7. Grandinetti, F.; Vinciguerra, V. Adducts of NF2+ with Diatomic and Simple Polyatomic Ligands: A Computational Investigation
on the Structure, Stability, and Thermochemistry. Int. J. Mass Spectrom. 2002, 216, 285–299. [CrossRef]
8. Fan, T.X.; Yang, G.J.; Chen, J.Q.; Zhang, D. Model Prediction of Thermodynamics Activity in Multicomponent Liquid Alloy.
Key Eng. Mater. 2006, 313, 19–24. [CrossRef]
9. Hildebrand, J.H. Solubility. XII. Regular Solutions. J. Am. Chem. Soc. 1929, 51, 66–80. [CrossRef]
10. Hildebrand, J.H.; Prausnitz, J.M.; Scott, R.L. (Eds.) Regular and Related Solutions: The Solubility of Gases Liquids, and Solids;
Van Nostrand Reinhold Company: New York, NY, USA, 1970.
11. Wilson, G.M. Vapor-Liquid Equilibrium. XI. A New Expression for the Excess Free Energy of Mixing. J. Am. Chem. Soc. 1964, 86, 127–130.
[CrossRef]
12. Nagata, I. Correlation of ternary liquid-liquid equilibria using a modification of the wilson equation. Fluid Phase Equilibria
1992, 72, 1–14. [CrossRef]
13. Renon, H.; Prausnitz, J.M. Local Compositions in Thermodynamic Excess Functions for Liquid Mixtures. AIChE J. 1968, 14, 135–144.
[CrossRef]
14. Tao, D.P. A New Model of Thermodynamics of Liquid Mixtures and Its Application to Liquid Alloys. Thermochim. Acta 2000,
363, 105–113. [CrossRef]
15. Hill, T.L. Statistical Mechanics: Principles and Selected Applications; Courier Corporation: Chelmsford, MA, USA, 1957.
16. Hu, Y. Molecular Thermodynamics of Fluids; Higher Education Press: Beijing, China, 1982.
17. Wang, C.; Chen, X.; Tao, D. Estimation of Component Activities and Molar Excess Gibbs Energy of 19 Binary Liquid Alloys from
Partial Pair Distribution Functions in Literature. Metals 2023, 13, 996. [CrossRef]
18. Wang, C.; Chen, X.; Tao, D. Development of a Non-Integral Form of Coordination Number Equation Based on Pair Distribution
Function and Gaussian Function. Metals 2023, 13, 384. [CrossRef]
19. Department of Mathematics, Tongji University. Advanced Mathematics, 6th ed.; Higher Education Press: Beijing, China, 2007.
20. Chanda, S.; Ahmed, A.Z.Z.; Bhuiyan, G.M.; Barman, S.K.; Sarker, S. A Test of Distribution Function Method in the Case of Liquid
Transition Metals Alloys. J. Non-Cryst. Solids 2011, 357, 3774–3780. [CrossRef]
21. Trybula, M.; Jakse, N.; Gasior, W.; Pasturel, A. Structural and Physicochemical Properties of Liquid Al–Zn Alloys: A Combined
Study Based on Molecular Dynamics Simulations and the Quasi-Lattice Theory. J. Chem. Phys. 2014, 141, 224504. [CrossRef]
22. Hongri, C.; Xiufang, B.; Hui, L.; Li, W. Molecular Dynamics Simulation on Structures of Cu–Ni Alloy. China J. Chem. Phys.
2002, 15, 288–294. [CrossRef]
23. Belova, I.V.; Ahmed, T.; Sarder, U.; Yi Wang, W.; Kozubski, R.; Liu, Z.-K.; Holland-Moritz, D.; Meyer, A.; Murch, G.E. Computer
Simulation of Thermodynamic Factors in Ni–Al and Cu–Ag Liquid Alloys. Comput. Mater. Sci. 2019, 166, 124–135. [CrossRef]
24. Wang, H.; Wei, B. Understanding Atomic-Scale Phase Separation of Liquid Fe–Cu Alloy. Chin. Sci. Bull. 2011, 56, 3416–3419.
[CrossRef]
25. Goto, R.; Shimojo, F.; Munejiri, S.; Hoshino, K. Structural and Electronic Properties of Liquid Ge–Sn Alloys: Ab Initio Molecular-
Dynamics Simulations. J. Phys. Soc. Jpn. 2004, 73, 2746–2752. [CrossRef]
Metals 2023, 13, 1773 16 of 17

26. Song, B.; Xu, N.; Jiang, W.; Yang, B.; Chen, X.; Xu, B.; Kong, L.; Liu, D.; Dai, Y. Study on Azeotropic Point of Pb–Sb Alloys by
Ab-Initio Molecular Dynamic Simulation and Vacuum Distillation. Vacuum 2016, 125, 209–214. [CrossRef]
27. Qin, J.; Pan, S.; Qi, Y.; Gu, T. The Structure and Thermodynamic Properties of Liquid Al–Si Alloys by Ab Initio Molecular
Dynamics Simulation. J. Non-Cryst. Solids 2016, 433, 31–37. [CrossRef]
28. Kbirou, M.; Mazroui, M.; Hasnaoui, A. Atomic Packing and Fractal Behavior of Al-Co Metallic Glasses. J. Alloys Compd. 2018,
735, 464–472. [CrossRef]
29. Canales, M.; González, D.J.; González, L.E.; Padró, J.A. Static Structure and Dynamics of the Liquid Li–Na and Li–Mg Alloys.
Phys. Rev. E 1998, 58, 4747–4757. [CrossRef]
30. Pu, Z.; Zhang, H.; Li, Y.; Yang, B. Study on Sn-Sb Alloy by Ab-Initio Molecular Dynamic Simulation. IOP Conf. Ser. Mater. Sci. Eng.
2018, 394, 032098. [CrossRef]
31. Jakse, N.; Pasturel, A. Local Order and Dynamic Properties of Liquid and Undercooled Cux Zr1−x Alloys by Ab Initio Molecular
Dynamics. Phys. Rev. B 2008, 78, 214204. [CrossRef]
32. Korkmaz, Ş.; Korkmaz, S.D. Structure and Inter-Diffusion Coefficients of Liquid Nax K1−x Alloys. J. Phase Equilib. Diffus. 2010, 31,
15–21. [CrossRef]
33. Ishii, Y.; Takanaga, T. Atomic and Electronic Structures and Dynamics in Liquid Alloys near Eutectic Point. J. Phys. Soc. Jpn.
2000, 69, 3334–3341. [CrossRef]
34. Wang, C.C.; Wong, C.H. Short-to-Medium Range Order of Al–Mg Metallic Glasses Studied by Molecular Dynamics Simulations.
J. Alloys Compd. 2011, 509, 10222–10229. [CrossRef]
35. Korkmaz, S.D.; Korkmaz, Ş. A Study for Structure and Inter-Diffusion Coefficient of Liquid K1−x Csx Metal Alloys. Phys. Chem. Liq.
2011, 49, 801–810. [CrossRef]
36. Costa Cabral, B. First Principles Molecular Dynamics of a Liquid Li–Na Alloy. J. Mol. Struct. Theochem 1999, 463, 145–149.
[CrossRef]
37. Bai, Y.W.; Zhao, X.L.; Bian, X.F.; Song, K.K.; Zhao, Y. Structure Evolution of Au50 Cu50 Alloy from Melt to the Disordered Solid
Solution. Mater. Sci. Forum 2020, 993, 273–280. [CrossRef]
38. Shi, L.; Jia, L.; Ning, P.; Sun, X.; Wang, C.; Ma, Y.; Wang, F.; Qu, T.; Li, K. Vacuum Distillation and Ab Initio Molecular Dynamic
Simulation of Al–Li Alloys. Vacuum 2023, 210, 111877. [CrossRef]
39. Yang, S.J.; Hu, L.; Wang, L.; Wei, B. Molecular Dynamics Simulation of Liquid Structure for Undercooled Zr-Nb Alloys Assisted
with Electrostatic Levitation Experiments. Chem. Phys. Lett. 2018, 701, 109–114. [CrossRef]
40. Özdemir Kart, S.; Tomak, M.; Uludoğan, M.; Çağın, T. Liquid Properties of Pd–Ni Alloys. J. Non-Cryst. Solids 2004, 337, 101–108.
[CrossRef]
41. Liu, D.; Qin, J.Y.; Gu, T.K. The Structure of Liquid Mg–Cu Binary Alloys. J. Non-Cryst. Solids 2010, 356, 1587–1592. [CrossRef]
42. Rao, R.V.G.; Venkatesh, R. Investigations of the Dynamic Properties of the Liquid Phase of Glassy Ca–Al Alloys. Phys. Stat. Sol. B
1992, 170, 39–46. [CrossRef]
43. Mendelev, M.I.; Kramer, M.J.; Hao, S.G.; Ho, K.M.; Wang, C.Z. Development of Interatomic Potentials Appropriate for Simulation
of Liquid and Glass Properties of NiZr2 Alloy. Philos. Mag. 2012, 92, 4454–4469. [CrossRef]
44. Roik, O.S.; Yakovenko, O.M.; Kazimirov, V.P.; Sokol’skii, V.E.; Golovataya, N.V.; Kashirina, Y.O. Structure of Liquid Al Sn Alloys.
J. Mol. Liq. 2021, 330, 115570. [CrossRef]
45. Roik, O.S.; Samsonnikov, O.V.; Kazimirov, V.P.; Sokolskii, V.E.; Galushko, S.M. Medium-Range Order in Al-Based Liquid Binary
Alloys. J. Mol. Liq. 2010, 151, 42–49. [CrossRef]
46. Chen, F.; Cao, C.; Zhong, Q.; Liu, J.; Yang, L.; Chen, Z. Ab Initio Molecular Dynamics Study on Local Structure and Dynamic
Properties of Liquid Ni62 Nb38 Alloy. Mater. Today Commun. 2021, 27, 102207. [CrossRef]
47. Gruner, S.; Kaban, I.; Kleinhempel, R.; Hoyer, W.; Jóvári, P.; Delaplane, R.G. Short-Range Order and Atomic Clusters in Liquid
Cu–Sn Alloys. J. Non-Cryst. Solids 2005, 351, 3490–3496. [CrossRef]
48. Jakse, N.; Nguyen, T.L.T.; Pasturel, A. Local Order and Dynamic Properties of Liquid Au x Si1−x Alloys by Molecular Dynamics
Simulations. J. Chem. Phys. 2012, 137, 204504. [CrossRef] [PubMed]
49. Chou, C.-Y.; Kim, H.; Hwang, G.S. A Comparative First-Principles Study of the Structure, Energetics, and Properties of Li–M
(M = Si, Ge, Sn) Alloys. J. Phys. Chem. C 2011, 115, 20018–20026. [CrossRef]
50. Li, X.; Wang, J.; Qin, J.; Dong, B.; Pan, S. The Reassessment of the Structural Transition Regions along the Liquidus of Fe–Si Alloys
and a Possible Liquid–Liquid Structural Transition in FeSi2 Alloy. Phys. Lett. A 2018, 382, 2655–2661. [CrossRef]
51. Bhuiyan, G.M.; Ali, I.; Rahman, S.M.M. Atomic Transport Properties of AgIn Liquid Binary Alloys. Phys. B Condens. Matter
2003, 334, 147–159. [CrossRef]
52. Weber, H.; Schumacher, M.; Jóvári, P.; Tsuchiya, Y.; Skrotzki, W.; Mazzarello, R.; Kaban, I. Experimental and Ab Initio Molecular
Dynamics Study of the Structure and Physical Properties of Liquid GeTe. Phys. Rev. B 2017, 96, 054204. [CrossRef]
53. Peng, H.L.; Voigtmann, T.; Kolland, G.; Kobatake, H.; Brillo, J. Structural and Dynamical Properties of Liquid Al–Au Alloys.
Phys. Rev. B 2015, 92, 184201. [CrossRef]
54. Guo, F.; Tian, Y.; Qin, J.; Xu, R.; Zhang, Y.; Zheng, H.; Lv, T.; Qin, X.; Tian, X.; Sun, Y. Structure of Liquid Cu–Sb Alloys by Ab Initio
Molecular Dynamics Simulations, High Temperature X-ray Diffraction, and Resistivity. J Mater Sci 2013, 48, 4438–4445. [CrossRef]
Metals 2023, 13, 1773 17 of 17

55. Binary Systems. Part 1_Elements and Binary Systems from Ag-Al to Au-Tl; Franke, P.; Neuschütz, D. (Eds.) Lehrstuhl für Theoretis-
che Hüttenkunde, Rheinisch-Westfälische Technische Hochschule Aachen; Landolt-Börnstein—Group IV Physical Chemistry;
Springer: Berlin/Heidelberg, Germany, 2002; Volume 19B1, ISBN 978-3-540-65327-1.
56. Franke, P.; Neuschütz, D. (Eds.) Binary Systems. Part 2: Elements and Binary Systems from B–C to Cr–Zr; Landolt-Börnstein—Group
IV Physical Chemistry; Springer: Berlin/Heidelberg, Germany, 2004; Volume 19B2, ISBN 978-3-540-20205-9.
57. Franke, P.; Neuschütz, D. (Eds.) Binary Systems. Part 3: Binary Systems from Cs-K to Mg-Zr; Landolt-Börnstein—Group IV Physical
Chemistry; Springer: Berlin/Heidelberg, Germany, 2005; Volume 19B3, ISBN 978-3-540-23119-6.
58. Franke, P.; Neuschütz, D. (Eds.) Binary Systems. Part 4: Binary Systems from Mn-Mo to Y-Zr; Springer: Berlin/Heidelberg, Germany,
2006; Volume 19B4.
59. Franke, P.; Neuschütz, D. (Eds.) Binary Systems. Part 5: Binary Systems Supplement 1: Phase Diagrams, Phase Transition Data, Integral
and Partial Quantities of Alloys; Landolt-Börnstein—Group IV Physical Chemistry; Springer: Berlin/Heidelberg, Germany, 2007;
Volume 19B5, ISBN 978-3-540-45279-9.

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.