Figure 4.4a and 4.4b showed an agreement between the calculated
results from the present method and experimental values for
anharmonic contribution to the second-order cumulant and
anharmonic factor (T), respectively. These experimental values are
extracted from the experimental second-order cumulant results.
Anharmonic factor (T) is a new factor given by Nguyen Van Hung,
et al. in the article [21]. In addition, the above anharmonic
contribution is difficult to measure directly. So when using
anharmonic correlated Einstein model, we can calculate and
represent these anharmonic components that depending temperature
based on theoretical calculations or experimental measurement
values of the second-order cumulant
27 trang |
Chia sẻ: honganh20 | Ngày: 02/03/2022 | Lượt xem: 325 | Lượt tải: 0
Bạn đang xem trước 20 trang tài liệu Tóm tắt Luận án Study of the thermodynamic parameters and cumulants of some materials by anharmonic xafs method, để xem tài liệu hoàn chỉnh bạn click vào nút DOWNLOAD ở trên
eter values. In this model, the
vibrational density of states center at single vibrational
frequency: ( ) ( )jj E .
1.2.2. Equation of motion method [3,38]
2
2 1( ) coth . ( )
2 2 2
ii iij
j i
ij i
R R
T
M
(1.37)
1.2.3. Statistical moment method [39-46]
(1)
0 0(T) (T) (0) y (T)x r r a a (1.58)
2
2 2 2
0 0 0. 2i i iR u u u u u u
(1.59)
CHAPTER 2. ANHARMONIC CORRELATED EINSTEIN
MODEL IN STUDY OF XAFS THERMODYNAMIC
PARAMETERS
2.1. Effective potential in anharmonic correlated Einstein model
The generalized expression of anharmonic effective interaction
potential using in ACEM:
0 ij
, ,
(x) (x)E i
i a b j a b i
x R R
M
(2.3)
5
Basing on the quantum statistical perturbation theory, we can
determine Hamiltonian of the system and can extract anharmonic
effective interaction potential expressions:
2 3
3
1
( )
2
E effa k a k a (2.6)
2 3
3 3(y) (k 3 a ) y yE eff a k k (2.7)
21(x) ( ) ( )
2
E E eff Ea k y y (2.9)
2.2. Morse potential [53]
ij 0 ij 02 (r ) (r )ij( ) 2r rr D e e
Taylor series expansion in approximation up to the third order:
2 2 3 3( ) ( 1 )x D x x (2.13)
Table 2.2. Morse parameters of copper (Cu) and zinc (Zn) from
theoretical calculation.
Materials D (eV) (Å-1) r0(Å) c
Cu [20,60,61] 0.3429 1.3588 2.868 2
Cu [62] 0.3364 1.5550 2.8669 2
Zn [20,15,17,22,23,59,63] 0.1700 1.7054 2.793 1/ 2
2.2.1. Applying Morse potential to calculate the thermodynamic
parameters and effective interaction potential in anharmonic
correlated Einstein model for fcc and hcp structural materials
Figure 2.3. Face centered
cubic Crystal [47]
Figure 2.4. Hexagonal close-
packed Crystal [47]
Derive the expression of effective interaction potential which is
used in the anharmonic correlated Einstein model, we can get:
(x) (x) 2 ( ) 8 ( ) 8 ( )
2 4 4
E
x x x
(2.28)
6
Derive the expressions of effective local force constant, cubic
anharmonic parameter as well as Einstein frequency and temperature
for fcc and hcp crystals:
2 2
3
3
2 3
3
5 1 5
2
5
4
1
( ) 5 (ay y )
4
eff
E
k D a D
k D
y D
2
2
5
5
eff
E
E
E
B B
k D
D
k k
2 2
3
3
2 3
9
5 1 5
10
3
4
3
( ) 5 (ay y ).
20
eff
E
k D a D
k D
y D
(2.31); (2.32,2.34) (2.33)
2.2.2. Applying Morse potential to calculate thermodynamic
parameters and interatomic effective potential in anharmonic
correlated Einstein model for diamond structural materials
Figure 2.5. Diamond structural crystal [47]
Derive the expression of interaction effective potential which is
used in the anharmonic correlated Einstein model, we can get:
1 1 1 1
(x) (x) 3 3 (x) 3 3
3 3 6 6
E x x x x
M M
(2.36)
Derive the expressions of effective local force constant, cubic
anharmonic parameter as well as Einstein frequency and temperature
for fcc and hcp crystals:
2 3 2 2
3
3
7 35 7 5 7
2 1
6 12 3 2 3
35
36
effk D a D a D
k D
(2.39);
2
2
7
3
7
3
eff
E
E
E
B B
k D
D
k k
(2.40)
Morse parameters for Si[25,64]:
D=1.83 (eV); =1.56 (Å-1) và r0=2.34 (Å)
Morse parameters for Ge[25,64]:
D=1.63 (eV); =1.50 (Å-1) và r0=2.44 (Å)
7
2.3. Stillinger-Weber potential [52,65]
( ) Wij ijkx (2.41)
where the single-pair interaction potential component:
1
ij ij ijexp ,
0, r
p q
ij
ij
ij
r r r
A B a r a
a
(2.42)
The three-body interaction potential component:
2
1 1
ij ik ij
1
W exp ( ) ( ) cos
3
ijk kr a r a
Parameters for Si[52,65]: A=7.049556277; B=0.6022245584; p=4;
q=0; a=1.80; =21.0; =1.20; =2.0951Å; =50 kcal/mol.
Parameters for Ge[52]: A=7.049556277; B=0.6022245584; p=4;
q=0; a=1.80; =31.0; =1.20; =2.181Å; =1.93 kcal/mol.
2.4. Calculating thermodynamic parameters by anharmonic
correlated Einstein model
2.4.1. Calculating cumulants by anharmonic correlated Einstein
model
Atomic vibration is quantized in terms of phonons, anharmonicity
is the result of phonon-phonon interaction. So we can express y in
term of annihilation and creation operators [68]:
0
ˆ ˆy ( )a a with
0
2 E
and ˆ ˆa a n
The above operators have the following properties:
ˆ ˆ ˆ ˆ ˆ ˆ, 1, 1 1 , 1 1 ,a a a n n n a n n n a a n n n
(2.54)
Then the averaging procedure can be calculated by statistical
physics as [69]:
1
( y ),m 1,2,3,...m my Tr
Z
(2.55)
Calculating (2.55) in cases of:
+ m is even value:
0
0 0
1 1 1
( y ) ( y ) Enm m m m
n
y Tr Tr e n y n
Z Z Z
(2.59)
8
We can be received the second-order cumulant:
2 (2) 2
0
1
En
n
y e n y n
Z
(2.60)
+ m is odd value:
'
, '0 '
1
' '
n nE E
m m
E
n n n n
e e
y n n n y n
Z E E
(2.64)
We can receive the first and third-order cumulants.
Finally, we can receive expressions of cumulants for fcc (Cu) and
hcp (Zn) structural materials:
fcc: (1) 2
0
(2) 2
0
4 2
(3) 0
2
3 1
( )
4 1
( 1)
( )
(1 )
( ) (1 10 )
2 (1 )
z
a
z
z
z
z z
z
hcp: (1) 2
0
(2) 2
0
4 2
(3) 0
2
9 1
( )
20 1
( 1)
( )
(1 )
3( ) (1 10 )
10 (1 )
z
a
z
z
z
z z
z
(2.63, 2.73, 2.80)
2.4.2. Derive expressions of cumulants based on the second
cumulant only in anharmonic correlated Einstein model.
From the expression about the relationship between temperature
variable z and mean square relative displacement given by Rabus
[8,9]:
2 2
0
2 2
0
( )
( )
z
, replace into (2.63,2.73,2.83) we can receive
expressions of cumulants based on the Debye-Waller factor or the
XAFS second cumulant only for fcc and hcp structural materials:
(1) 2 (2)
0
(2) 2 2
0
4 2
(3) 2 2 2 20
02
3 1 3
( )
4 1 4
( 1)
( )
(1 )
( ) (1 10 )
[3( ) 2(( ) ) ]
2 (1 ) 2
z
a
z
z
z
z z
z
(1) (2)
(2) 2
(3) 2 2 2 2
0
9
20
3
[3( ) 2(( ) ) ]
10
(2.82)
where 2
0 2
( )
10
E
D
The relationship among cumulants is determined according to:
(1) 2
2(3) 2
0
2
1
4 ( )
2
3
(2.83)
9
Formula (2.83) showed that the ratio among cumulants related only to
the second-order cumulant. This ratio is considered as the standard for the
XAFS theoretical studies method regarding physics [9]. We can see that
this ratio will approach the classical value of ½ when
2
0
2
( )
approaches
to 0 so that the classical limit is applicable.
2.4.3. Calculating thermal expansion coefficient in anharmonic
correlated Einstein model
For fcc (Cu) and hcp (Zn) structural material: The expression of
thermal expansion coefficient is derived base on Debye-Waller factor
or the XAFS second cumulant such as:
2 2 4
0 0
2
( ) ( )
T T
T
with
3
0 15
4
T
B
D
rk
and
3
0 9
4
T
B
D
rk
(2.87)
The relationship of cumulants and thermal expansion coefficient is
determined by (2.88). We see that
2
(3)
.T 1
2
T r
when TE means that
from temperature TE, anharmonic effects are significant, we can
apply the classical approximation, and when T<E, anharmonic
effects are insignificant, we must use quantum theory. In particular,
at temperature T=E/2, the ratio from (2.88) approaches the classical
value of 1/2, so when T<E/2 we must consider anharmonic effects.
22
0
2
2 2 2
(3) 22
0
2
( )
1
.T 1 5
. .
2 2 ( )
1
3
T
B
r D
k T
(2.88)
2.4.4. Evaluation of the calculated XAFS second cumulant results
using Morse and Stillinger-Weber potential in the anharmonic
correlated Einstein model for diamond structure semiconductor
materials
Applying the anharmonic correlated Einstein model using Morse
potential:
From (2.39,2.40), replace into (2.7) we can receive:
2 37 5( ) (ay y )
3 12
E y D
10
Replace above expression into (2.59) and (2.64) we can
determine expressions of cumulants for diamond structural materials:
(1) 2 (2)
0
(2) 2 2
0
4 2
(3) 2 2 2 20
02
5 1 5
( )
4 1 4
( 1)
( )
(1 )
5( ) (1 10 ) 5
[3( ) 2(( ) ) ]
6 (1 ) 6
z
a
z
z
z
z z
z
(2.89)
where:
2
0 2
3
( )
14
E
D
+ The expression of thermal expansion coefficient:
2 2 4
0 0
2
( ) ( )
T T
T
với
3
0 35
12
T
B
D
rk
(2.90)
+ The expression of anharmonic factor:
2
2 2 225 5 5( ) (T)[3 (T)(3 (T)]
24 4 4
T
R R
(2.91)
+ The expression of anharmonic contribution:
2 2 2 2 2
0 0(T) (T)[ (T) (T )]= (T)[ (T) ]A H H (2.92)
+ The expression of anharmonic contributions to XAFS spectra
phase shift and amplitude:
2 22 ( )( , ) Ak TAF k T e
(2.93)
2 3 (3)1 1 4( , ) 2 [ (T)( )] k (T)
3
A k T k R
R
(2.94)
Applying the anharmonic correlated Einstein model and the
statistical moment method using Stillinger-Weber potential:
11
Figure 2.6. Temperature
dependence of the second-order
cumulant using Stillinger-Weber
potential in the statistical moment
method for Si.
Figure 2.7. Temperature
dependence of the second-order
cumulant using Stillinger-Weber
potential in the statistical moment
method for Ge.
Figures 2.6 and 2.7 show a good agreement of the statistical
moment method using in the calculation the XAFS second cumulant
values for Si and Ge diamond semiconductors, respectively. For Si,
the results were compared with the obtained values given by M.
Benfatto in the article [70] at 80 K, 300 K, and 500 K. For Ge, the
results have an agreement with experimental values given by A.E.
Stern in [71] at 300 K, G. Dalba in [72] at some temperatures and
with theoretical calculation results given by J.J. Rehr in the article
[4] when using the LDA method at 300 K. Moreover, the obtained
results are consistent with experimental results of A.Yoshiasa in [73]
in some specific temperatures, even the results are calculated from
the GGA and hGGA methods given by J.J.Rehr at 300 K [4]. These
results published in the article [19].
Numerical results of the second-order cumulant using Morse and
Stillinger-Weber potential for Si and Ge crystals by the anharmonic
correlated Einstein model were evaluated and compared in the article
[18,24,25]. The anharmonic correlated Einstein model using two
potentials is consistent with experimental values as well as those
obtained from other methods. Therefore, the anharmonic correlated
12
Einstein model can be applied to diamond semiconductors using Morse
and Stillinger-Weber potential.
2.5. Quantum effects in low temperature limit and classical
approximation in high temperature
The obtained thermodynamic parameter formulas from quantum
theory can be applied at all temperature values. At the high-
temperatures, the formulas include the results of classical
approximation theory. At the low-temperature limit, quantum effects
express through contributions of zero-point energy.
Thermodynamic
Quantity
T0 T
(1) a (1)
0 (1 2 )z 33 / kB effk k T
(2) 20 (1 2 )z / kB effk T
(3) (3)0 (1 12 )z
2 3
36 ( ) / kB effk k T
T
0 2(ln ) (1 2 )T z z z 33 / kBk r
(1) 2
(3)
(1) 2 2 20 0
(3)
0
(1 2 ) 3(1 2 ) 3
(1 12 ) 2(1 12 ) 2
z z
z z
1
2
2
(3)
.TT r
2
(3)
.T 1
3 ln 0T
r
z
z
1
2
CHAPTER 3. EXPERIMENTAL MEASUREMENT AND
APPLICATION OF ANHARMONIC CORRELATED
EINSTEIN MODEL IN STUDY ON XAFS
THERMODYNAMIC PARAMETERS FOR HCP AND FCC
STRUCTURE MATERIALS.
3.1. Synchrotron facility and XAFS spectra experimental station
The preparation for experimental samples depending on
temperature:
13
Figure 3.5. Experimental station
Beamline 8. SLRI
Figure 3.7. Experimental
XAFS measurement
depending on temperature
3.2. Experimental measurements results of the Debye-Waller
factors for hcp structure material.
Experimental values are shown in figure 3.12 and table 3.1.
Figure 3.12. XAFS spectrum and Fourier transform magnitudes of
Zn at 300 K, 400 K, 500 K, and 600 K
Table 3.1. The value of cumulants and thermal expansion
coefficients of Zn: Theoreratical calculation (LT) and experimental
value (TN) at temperatures. Symbol: MHĐH – Harmonic model
T(K)
(1)(Å)
LT
(1)(Å)
TN
2(Å)
LT
2(Å)
MHĐH
2(Å)
TN
(3)(Å)
LT
(3)(Å)
TN
T
(10-5/K)
LT
T
(10-5/K)
TN
300 0.0139 0.0143 0.0110 0.0109 0.0113 0.0003 0.0003 1.555 1.582
400 0.0182 0.0189 0.0146 0.0143 0.0149 0.0005 0.0006 1.582 .618
500 0.022 0.0232 0.0182 0.0177 0.0185 0.0008 0.0009 1.595 1.599
600 0.0270 0.0279 0.0219 0.0211 0.0223 0.0011 0.0012 1.602 1.630
14
3.3. Determining thermodynamic parameters of XAFS from the
experimental values of the Debye-Waller factor or the second-
order cumulant by anharmonic correlated Einstein model for
hcp structure materials.
Figure 3.14. Temperature dependence of the first cumulant, total and
the harmonic second cumulant and experimental values
From the illustration in figure 3.14b, anharmonic correlated
Einstein model, and the harmonic correlated model [82] have certain
deviations for the second-order cumulant or Debye-Waller factor in
the high-temperature range. ACEM is more suitable for experimental
values than the harmonic correlated model. Note that the data of the
first-order cumulant is derived from the experimental value of the
second-order cumulant.
Figure 3.15. Temperature dependence of the third cumulant and
thermal expansion coefficient of Zn calculated from cumulant
experimental values.
15
Figure 3.16. Temperature dependence of cumulants ratio, the ratio
between thermal expansion coefficient and cumulants of Zn.
Similar to the first -order cumulant, we are also able to determine
the third-order cumulant and thermal expansion coefficient of zinc
(Zn) at 300 K, 400 K, 500 K, and 600 K. Figure 3.15a and 3.15b
showed the results derived from experimental measurements are very
agreement with the calculations from the theoretical model. To
assess the validity of the theoretical model, we can also check by
establishing the ratio among cumulants according to the expression
(2.83) and ratio among thermal expansion coefficient and cumulants
according to the expression (2.88). Figure 3.16 showed the above
relationships. From figure 3.16, the values are derived from
experimental values that make these ratios reach the value of ½.
These ratios are used as the standard method for assessing cumulant
studies [9,56,81,83], as well as for determining temperature when the
classical limit can be applied [9]. The theoretical results and the
results of these ratios showed that hcp structure materials,
specifically Zn, we can use classical correlation Einstein model when
the temperature is higher than Einstein temperature (E = 206 K).
3.4. Experimental results of XAFS Debye-Waller factors for fcc
structure material
16
Figure 3.17. XAFS spectrum and Fourier transform magnitudes of Cu at
300 K, 400 K, 500 K
Figure 3.18. The process of fitting the XAFS spectrum of Cu at
temperatures
XAFS spectrum at temperature values after merging are fitted to
the theoretical spectra by using Artemis software. The R, k variables
are in R space [1-3 Å] or k space [3.00-14.023 Å-1] run to the optimal
value between theoretical spectra and experimental spectra.
3.5. Determining thermodynamic parameters of XAFS from
experimental values of the Debye-Waller factor or the second-
order cumulant by anharmonic correlated Einstein model for fcc
(Cu) structure material
Figure 3.19. Temperature dependence of the first cumulant, total and
harmonic second cumulant and the experimental values.
Anharmonic correlated Einstein model and harmonic correlated
Einstein model [81] have certain deviations for the second-order
17
cumulant or Debye-Waller factor in high-temperature range (Figure
3.19). The results showed that anharmonic correlated Einstein model
is well suited to experimental values as well as obtained results of S.
a Beccara, et al. [82] for the first-order cumulant and V. Pirog, et al.
[58] for the second-order cumulant. Note that, the data of the first-
order cumulants are derived from the experimental second-order
cumulants values.
Figure 3.20. Temperature dependence of the third-cumulant and
thermal expansion coefficient of Cu calculated from experimental
cumulant values
Similar to the first-order cumulant, we can also identify the third-
order cumulant and thermal expansion coefficient of copper (Cu) at
300 K, 400 K, 500 K. Figure 3.20 showed the results derived from
experiment values were very consistent with the obtained data of V.
Pirog, et al [58] and T. Yokoyama, et al [88] for the third-order
cumulant. Figure 3.20b indicated agreement among calculated results
from the present method and the experimental results and obtained
results from other documents [89] for the thermal expansion
coefficient. To evaluate the validity of the theoretical model, we
verified by establishing the ratio among cumulants according to
expression (2.83) and the ratio among the thermal expansion
coefficient and cumulants according to the expression (2.88). Figure
3.21 showed these relationships.
18
Figure 3.21. Temperature dependence of cumulants ratio, ratio
between thermal expansion coefficient and cumulants of Cu
The values extracted from experiments make these ratios
approach the value of ½ (Figure 3.21). These ratios are used as the
standard method for assessing cumulant studies [9, 81, 90], as well as
for determining temperatures at which classical limits can be applied
[9]. The theoretical results and the results of these ratios showed that
hcp structure materials, specifically Cu, we can use classical
correlation Einstein model [9,81] when the temperature is higher
than Einstein temperature (E = 218 K).
CHAPTER 4. ANHARMONIC CORRELATED EINSTEIN
MODEL IN STUDY OF XAFS PHASE AND AMPLITUDE
CONTRIBUTION OF HCP AND FCC STRUCTURE
MATERIALS
4.1. Overview of anharmonic XAFS spectra
The anharmonic XAFS function is represented by cumulant
expansion approach [21,60,90,91]:
2
(k)
(k) ( )
2
(2 )
( ) ( ) Im exp 2
!
R
n
i n
n
e ik
k F k e ikR
kR n
(4.1)
XAFS amplitude expression [9,90-92]:
2
(1) 2 2 3 3 4 44 (T)k 4 2W( , ) 2 (T) 2 ( ) 1 ik (T) k (T) ...
(k) 3 3
i R
k T ki k T
R
(4.2)
2 3 (3)
0
1 1 4
( , ) ( , ) ( , ) 2 [ (T)( )] k (T)
3
A k T k T k T k R
R
(4.3)
19
Với 2 2 2
0(T) (T) (T ) (4.4)
4.2. XAFS Debye-Waller factor with contribution anharmonic.
In the high-temperature range, the Debye-Waller factor includes 2
components: a harmonic and anharmonic contribution component.
2 2 2(T) (T) (T)H A (4.5)
where
2 2 2 2 2
0 0(T) (T)[ (T) (T )]= (T)[ (T) ]A H H (4.6)
Replace (4.5) into (4.4) we receive:
2 2 2 2 2 2 2
0 0 0(T) (T) (T)[ (T) ] (1 (T)[ (T) ]H H H
With (T) is called the anharmonic factor of the XAFS second
cumulant which depending on temperature and Grüneisen parameter.
(T) 2 G
V
V
với
ln
ln
E
G
V
4.2.1. Determination of Grüneisen parameter G
From (2.32, 2.34) we can determine lnE/T
(4.9) and lnV/T (4.10). Therefore, we can determine:
2
2 2
3
( )
ln 4
9ln 4(1 )
8
E
G
R
V
(4.11)
4.2.2. Determination of anharmonic factor (T)
Determine the change in volume due to thermal expansion V/V
and from (4.12) we can determine:
2
2 2 29 3( ) (T)[1 (T)(1 (T)]
8 4 4
T
R R
(4.14)
4.5. XAFS spectra with contribution anharmonic components
The Debye-Waller factor includes two components as expression
(4.5). To accurately describe the actual spectra so that the XAFS
phase and amplitude in (1.14) need to be added to the anharmonic
factors. In detail, the phase component is added to the anharmonic
factor:
2 22 ( )( , ) Ak TAF k T e
20
The amplitude component is added to the anharmonic factor:
2 3 (3)1 1 4( , ) 2 [ (T)( )] k (T)
3
A k T k R
R
(4.16)
Then, the generalized expression of XAFS become to:
2 2
2
2 2 / ( )0
2
(k) ( ) ( , ) sin 2 (k) (k,T)j j
k R kj
j A j j A
j j
S N
f k F k T e e kR
kR
(1.17)
4.6. XAFS anharmonic phase and amplitude components for
HCP (Zn) structure material
The anharmonic XAFS components increased with increasing
temperature and k-wavenumber values (Figure 4.1).
Figure 4.1. Temperature dependence of anharmonic amplitude and
phase components with the k-wave number of XAFS spectra for hcp
(Zn) structural material.
These components (phase and amplitude) contribute to
anharmonic XAFS spectra show in figure 4.2 in both of the
theoretical calculations by the anharmonic correlated Einstein model
and experimental values.
Figure 4.2. Theoretical and experimental XAFS spectra with hcp
structure material (Zn) at temperatures.
21
Figure 4.3. Comparison of
Fourier transform magnitudes
of theoretical spectra with
experimental XAFS results
for hcp (Zn) structural
material at temperatures.
Figure 4.3 showed an agreement between the theoretical results of
the model with obtained Fourier transform magnitudes from
experimental measurements. In addition, we can see that the
magnitude of spectra decreases with the increasing temperature
gradually. Note that the anharmonic contribution components to the
XAFS phase and amplitude are calculated base on second-order
cumulant only. Moreover, by using anharmonic correlated Einstein
model, we can reconstruct XAFS spectra and the Fourier transform
magnitudes from the obtained experimental second-order cumulant
values. This study has shown that the obtained experimental results
are consistent with theoretical c
Các file đính kèm theo tài liệu này:
- tom_tat_luan_an_study_of_the_thermodynamic_parameters_and_cu.pdf