electronic transport properties in solid alloys (amorphous and recrystallized one’s). In other words: does this formalism developped for liquids also describe solids? Gasser Jean-Georges 1 , Abadlia Lakhdar 1 , Khalouk Karim 1 , Gasser Françoise 1 , Kaban Ivan 2 , Aboki Tiburce 3 and Moussa Mayoufi 4 . 1 Laboratoire de Physique des Milieux Denses, Université Paul Verlaine- Metz, France. 2 Institut für Physik Technische Universität Chemnitz Germany. 3 Laboratoire de Métallurgie Structurale ENSCP Paris. 4 Laboratoire de Chimie des Matériaux Inorganiques, Université Badji-Mokhtar Annaba, Algéria
51
Embed
What do we understand under « electronic transport propertie »?
General understanding of electronic transport properties in solid alloys (amorphous and recrystallized one’s). In other words: does this formalism developped for liquids also describe solids?. - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
General understanding of electronic transport properties in solid alloys
(amorphous and recrystallized one’s).In other words:
does this formalism developped for liquids also describe solids?
1 Laboratoire de Physique des Milieux Denses, Université Paul Verlaine- Metz, France.2Institut für Physik Technische Universität Chemnitz Germany.
3 Laboratoire de Métallurgie Structurale ENSCP Paris.4Laboratoire de Chimie des Matériaux Inorganiques, Université Badji-Mokhtar
Annaba, Algéria
What do we understand under « electronic transport propertie »?
By electronic transport we understand:• The electrical resistivity. (Microscopic Ohm’s law)• The thermal conductivity. (Fourier law)• The Absolute Thermoelectric Power (ATP) or
Thermopower or Seebeck coefficient of a couple or of an element. (Device called thermocouple used for measuring temperature)
• The Peltier coefficient of a couple or of an element. (Used for a device called Peltier device for cooling objects like PC’s processors)
Why is the thermopower of chromel (Ni90 Cr10) positive
while that
of alumel (Ni94Al3Si1Mn2 ) is negative ?
FOR WHAT IS THERMOPOWER USED?
THERE ARE WELL KNOWN INDUSTRIAL USE OF ELECTRONIC TRANSPORT
• The Seebeck effect is for measuring temperatures. One expects that the Seebeck coefficient is as constant as possible to be used reproducebly
• In a Peltier device, intense research is done in order to get the best figure of merit by increasing S and and decreasing
2TSZT
THE USE OF THERMOPOWER FOR NON DESTRUCTIVE TESTING IS NOT
WELL KNOWN
A workshop has been organised in 2002 in Lyon by Kleber on this subject
11Massardier presentation colloque TEP INSA-LYON 2002
One has learned that TEP is used to follow microstructural transformations as function of time
Analysis of ageing of nuclear reactors steel after Houzé (INSA Lyon) TEP INSA-LYON 2002
One has learned that TEP is used to follow tempering of nuclear reactors steel
Correlation between segregation and ThermoElectric Power (TEP) after Kleber (INSA Lyon) colloque TEP INSA-LYON 2002
One has learned that TEP is used to scan a surface. The map can be compared to that of the segregation of alloys
These very interesting phenomena led us to examine by ourself the resistivity and
thermopower changes in solids and to try to understand and
to explain them.
OUR NEW MEASUREMENTS• We developped an automatic system to measure
simultaneously the resistivity and thermopower of solid alloys.
It is based • On a « labview » driving program.• On a resistivity measurement method.• On a thermopower small T technique.The experiment is mounted in a furnace with a
programmable temperature regulator. The sample is put under vacuum or inert atmospher.
The sample
17
The labview software (registered) has been written by Françoise Gasser)
The software must be configurated. The data of the thermoelements in contact with the sample and of the thermocouples must be introduced
One can measure automatically either the resistivity alone (right diagram). The temperature gradient is checked (left diagram). All results are recorded in excel files
One can measure automatically the thermopower alone (right diagram). The temperature is oscillating (left diagram) between -2 and +2 Celsius around an increasing or decreasing temperature.
One can also measure simultaneously both the resistivity and the thermopower.
SOME OF OUR EXPERIMENTAL RESULTS
OUR METHODOLOGY
0 100 200 300 400 500 600 700 800-40
-35
-30
-25
-20
-15
-10
-5
0
5
10
25/01/10
Pouvoir thermoélectrique absolu Résistivité électrique
60
80
100
120
140
160
180
Résistivité électrique et pouvoir thermoélectrique absolu du ruban Fe 40Ni38B18Mo4 en fonction de la température
R
ésis
tivi
té é
lect
riq
ue
(µ.
cm)
Température (°C)
Po
uvo
ir t
her
mo
élec
triq
ue
abso
lu (
µV
/°C
)
In the first experiment we measured resistivity (in blue) and absolute thermoelectric power (in red) for the amorphous Fe40-Ni38-B18-Mo4 . The temperature increases and decreases is at a rate of 0.4K/minute. There is an initial change in resistivity around 380 ° C. A second one on the resistivity and the TEP above 465 ° C, a third change in slope in the resistivity and TEP occurs above 640 ° C. The increasing temperature curve is very different from the decreasing one. At room temperature resistivity was divided by 2.5 and the TEP goes from -3 to -37V/K
Attente à 475°C pendant 8H Attente à 630°C pendant 26H
Attente à 745°C pendant10H
Attente à 380°C pendant1H
25°C-300°C 300°C-100°C 100°C-380°C Attente à 380°C pendant 1H 380°C-400°C Attente à 400°C pendant 8H 400°C-350°C 350°C-475°C Attente à 475°C pendant 10H 475°C -632°C Attente à 632°C pendant 26H 632°C-345°C 350°C -745°C Attente 745°C pendant 10H 745°C-761°C 761°C-30°C 2ème Montée- Résistivité (µohm.cm) 2ème Descente-Résistivité (µohm.cm)
In a second experiment we carry out cycles and periods of waiting several hours at constant temperature. We first stopped at 300 ° C well below the transition temperature located (380 ° C), we decrease to 100 ° C and then go up to 380 ° C. The curves are superimposed this means that no phase transformation took place. At 380 ° C we wait 1H. We observe a small resistivity decreases. To decrease faster we increase from 20 ° C and waited 8H (see figure below). At the end of this waiting period a descent to 350 ° C and a rise to 400 ° C are superimposed. The descent blocks the evolution of the alloy up to 430 °C. Around 470 ° C the resistivity begins to increase. We then wait during 8H at 475 ° C.
ABADLIA THESIS
0 50 100 150 200 250 300 350 400 450 500 550
161,5
162,0
162,5
163,0
163,5
164,0
164,5
165,0
Attente à 400°C pendant 8H
Rés
istiv
ité é
lect
rique
(
.cm
)
Temps(min)
Résistivité électrique d'un ruban Fe40
Ni38
Mo4B
18en fonction du temps
Résistivité (µohm.cm)
We observe, at constant temperature, a curve resembling an exponential decay which seems to reach an asymptotic value.
ABADLIA THESIS
0 100 200 300 400 5000,8
1,0
1,2
1,4
1,6
1,8
2,0
2,2
2,4
2,6
2,8
3,0
3,2
3,4
Attente à 475°C pendant 8H
Pou
voir
ther
moé
lect
rique
abs
olu
(V
/°C
)
Pouvoir thermoélectrique absolu d'un ruban Fe40
Ni38
Mo4B
18en fonction du temps à 475°C
Temps(min)
PTA
At 475 °C the TEP increases exponentially as A(1-exp (-Bt)). The time constant is much shorter than at 400°C. After 3H no further change is observed
Résistivité électrique d'un ruban Fe40Ni38Mo4B18en fonction de la température
Rés
isti
vité
éle
ctri
qu
e (.
cm)
Attente à 400°C pendant 8H
Attente à 475°C pendant 8H Attente à 630°C pendant 26H
Attente à 745°C pendant10H
Attente à 380°C pendant1H
25°C-300°C 300°C-100°C 100°C-380°C Attente à 380°C pendant 1H 380°C-400°C Attente à 400°C pendant 8H 400°C-350°C 350°C-475°C Attente à 475°C pendant 10H 475°C -632°C Attente à 632°C pendant 26H 632°C-345°C 350°C -745°C Attente 745°C pendant 10H 745°C-761°C 761°C-30°C 2ème Montée- Résistivité (µohm.cm) 2ème Descente-Résistivité (µohm.cm)
• A resistivity and thermopower measurement device is a very powerfull tool to study phase transformations.
• Its advantage (compared to DSC) is that we can follow phase transformations at constant temperature.
ELECTRONIC TRANSPORT FOR LIQUID AND AMORPHOUS METALS
(AND ALLOYS)
On what is based the theory of disordered metals?
• Boltzman equation• Nearly free electron theory : density of states ≈E0.5
• Space isotropy• Short range order, high distance disorder• Scattering theory: each electron interacts with scatterers• Factorisation: In Ziman’s integral, the interaction of conduction
electrons with matter is the squared form factor (or t matrix characteristic of the scattering) times the structure factor characteristic of the relative position of the scattering sites
k
dqqEqtqake
mE
2
0
32632
2
),()(4
3
37
If we crudely approximate the product : a(q)t2(q) by a
constant one can then integrate analytically. One obtains:
k
dqqEqtqake
mE
2
0
32632
2
),()(4
3
EE
Ee
TkES KB
ln
ln
3
22
m
kE
2
22
The Ziman free electron approach gives a pretty good description of the resistivity and ATP of liquid metals.
Ecte
1.
Consequence: the ATP of liquid metals is always negative
a(q)
38
But some liquid metals have a positive thermopower. The 1/E resistivity curve is modulated by the maximum of the structure factor. If the Fermi energy is on the increasing side of the resistivity curve the thermopower is positive. This is the case for noble and divalent metals which present a positive thermopower
Modulation of the 1/E curve by the structure factor
39
Mott proposed to correct the mean free path at the Fermi energy:
FEfree
Ziman
En
Eng
gLL
)(
)( with 2
2gZiman Faber demonstrated that :
Consequently: One can correct the calculation of at the Fermi energy
Our proposition: use the Hafner density of states to
correct the resistivity by g2(E) in the whole domain of energy, then derivate it to obtain the thermopower..
Energie
S(q)
ATP>0 ATP<0
n(E)
Free electron DOS
DOS
EF E
Resistivity
)(
)()(
2 Eg
EE Ziman
Modulation of the 1/E curve by the squarred density of states
40
Mott proposed to correct the mean free path at the Fermi energy:
FEfree
Ziman
En
Eng
gLL
)(
)( with 2
2gZiman Faber demonstrated that :
Consequently: One can correct the calculation of at the Fermi energy
Our proposition: use the Hafner density of states to
correct the resistivity by g2(E) in the whole domain of energy, then derivate it to obtain the thermopower..
Energie
S(q)
ATP>0 ATP<0
n(E)
Free electron DOS
DOS
EF E
Resistivity
)(
)()(
2 Eg
EE Ziman
Modulation of the 1/E curve by the squarred density of states
41
0,0 0,2 0,4 0,6 0,8 1,00
50
100
150
200
exp=205.cm
Figure 4
Mn T=1260°C
LDA + Spin GGA-PBE + Spin Indicates the resistivity
at the calculated Fermi energy
Res
istiv
ity (
.cm
)
Energy (Rydberg)
The squared t matrix has a résonance for transition metals.
If the Fermi energy is on the left side of the resonnance (the thermopower is positive (Sc, Ti)
If it is on the right side the TEP is negative (nickel).
Zrouri thesis
Modulation of the 1/E curve by the resonant scattering
To summarize, the resistivity versus energy curve is a 1/E function modulated by :
• the structure factor• the ratio of the squared density of states divided by the free electron density of states • the squared t matrix (resonant scattering for transition metals
The ATP is positive on the left side of each of these modulating functions.
What changes in solids?• For monocristals, the space is no more isotropic.• There is a « large distance » ordering• The electronic « transport properties » are very
sensitive to physical and mechanical properties.• The change of resistivity can be very important• The change of thermopower can be very important• The sign of the thermopower is not understood.
Electronic Transport properties are used for non destructive testing in industry
44
0 5 10 15 20 25 30 35120
130
140
150
160
170
180 Résistivité PTA
Temps (heures)
Rés
istiv
ité (
.cm
)
-10
-8
-6
-4
420°C
PT
A (
V/K
)Fe82Si2B16
KHALOUK THESIS
45
Fe82Si2B16
10 20 30 40 50 60 70 80 902, degrees
amorphous
8 hrs
10 hrs
3 hrs
1 hour
annealing 0.25 hr
13 hrs
35 hrs
24 hrs
26 hrs
0
9000
18000
27000
36000
45000
а)
(330
)
(321
)
(222
)
(310
)
(220
)(211
)
(200
)
(110
)
0
0
0
0
0
0
0
0
0
F e 8 2
S i 2 B
1 6
420°C
Kaban’s structure factor
46
Fe82Si2B16
2kF
0 5 10 15 20 25 30 35120
130
140
150
160
170
180 Résistivité PTA
Temps (heures)
Rés
istiv
ité (
.cm
)
-10
-8
-6
-4
420°C
PT
A (
V/K
)
After 20H, the structure no more changes
KHALOUK THESIS
0.25H
16 and 20H
12H
1H
2 4 6 8 100
1
2
3
4
5
Ge15Te
85 Polycrystalline Amorphous
Y A
xis
Titl
e
X Axis Title
Ge-Te may be interpreted within this scheme
Kaban’s structure factor
Physical interpretation
• The four last slides are the (at least qualitative) experimental proof that the Faber-Ziman formalism can also be used for crystalline solids
• As for liquid and amorphous materials, the resistivity versus energy curve is modulated by the (sharper) structure factor
• The position of the Fermi energy in the resistivity versus energy curve is the crucial point
At the beginning of this presentation I asked :
Why is the thermopower of chromel (Ni90 Cr10) positive
while that
of alumel (Ni94Al3Si1Mn2 ) is negative ?
My answer is that very probably: impurities (Cr or Al…) move the Fermi energy from the left (chromel) to the right (alumel) side of a nickel structure factor peak as on the figure