Seebeck effect - Thermopower - Thermoelectric power - Thermoelectric
effect

The general defnition as follows

When we keep any electrically conducting material in a nonuniform
temperature distribution then it will develope an emf. It is called Seebeck
emf and the emf is called as Seebeck emf.

Then what do the other terms refers? Which one is the correct one? Why does
this confusion for simple phenomenon?

This confusion was started by the inventor ( Seebeck) himself
and it continues for 180 years ! Simple and straight forward reason for that
was due to the lack of clear understanding of the nature and origin of electricity
during his period. Let us see what might have led to this confusion.

The starting point of thermoelectric effects was around
1791. In that year Luigi Galvani noticed that the nerves and muscles of
a bisected frog’s leg contracted when they were in contact with two dissimilar
conducting materials. In 1793 Alessantro Volta gave explanatoin as the current
caused the contraction. Nothing moved for another 40 years. In 1821 Thomas
Johann Seebeck found that a magnetic needle held near a circuit made up
of two dissimilar conducting materials got deflected when part of the circuit
was heated( Notice it. Not junction was heated). It happened just one year
after the famous discovery of Oersted (A current carrying conductor deflects
a magnetic needle). To add further, Oersted was good friend of Seebeck. Immediately
Oersted gave explanation as the electricity caused the the deflection of
needle when the part of the circuit heated. But unfortunately Seebeck himself
was not ready to to accept this explanation. He was trying to relate his
efeect with influence of temperature difference on his circuit and earth’s
terrestial magnetic field. It made serious consequences. By that he midirected
others from right approach and probably the whole field slept for about 112
years! Inbetween a bit disturbance came in the form of Peltier effect. In
1834 a french watch maker Charles Peltier had found out that the passage
of current through Seebecks circuit caused liberation or absorbtion of heat
acroos the junctions depending on the direction current.

Again luck was not in favour. Peltier too failed to understand
the true nature and origin of his invention. Four years later Lenz demonstrated
this effect in an effective way. A drop of water placed at the junction frozed
when current was passed and it melted up on the reversal of current direction.
Eventhough it was a quiet remarkable discovery rediculously this too slept
for 100 years for unknown reasons. Because in application wise 3% efficency
was far better than any other means available during those period.

Not only interms of applications, due to its complicated nature
of theoretical aspect also it could not meet success. Free elctron theory
failed to give explanation regarding the sign and magnitude of the emf. For
example copper has positive emf eventhough the charge carriers are electrons.

For years the explanation might be given on the basis of combination
of two dissimilar conducting materials. The second thing the effect was
first observed in a circuit. It seems we were( or are!) ( Atleast myself
)coupled stronly with coupled dissimilar conducting materials interms of
explanation and nomenculature. If it is the case then is it possible to
observe directly the absolute emf of a single conducting material (with out
Thompson effect)? No( except superconductors). The reason as follows. Suppose
a conducting material (A)is in a nonuniform temperature distribution. Across
the conductor let us take two points(p,q) which where at temperatures T1
and T2 ( T1>T2). To measure the emf one has to connect probes of a voltmeter
at the two points(p,q). The probe is a conducting material and its ends are
at temperatures T1 and T2. So it will develope its own emf with respect
to the temperature difference. What we observe now is the net or relative
Seebeck emf of material (A) + probe!

Total emf (DE)= emf of material(DE1) + emf of probe(DE2)

The Seebeck coefficient is given as ,

S=DE/DT

DT=T1~T2

So in a thermocouple which is made up of two ( Let us call dissimilar conducting
materials as thermoelements) thermoelements A & B,

S(Relative)=S(A)+S(B).

It is this relative Seebeck coefficient that has been called by the ANACHRONISTIC
and INEPT term as “ THERMOPOWER” etc( “The Measurement Instrumentation and
Sensor”- Handbook, by John.C.Webster. CRC Press in cooperation with IEE
press-1995. Section 32. Page No.44).

( Surprisingly in all leading journals most of the time any one of this
inaccurate term appears.)

Let us try to have a final conclusions.

1. Thermoelectric effect means it includes all three types of effects( Seebeck,
Peltier and
Thompson)
2. We could not find out any reason or justification to use the term “ Thermopower”
at any
specific place.
3. In a thermocouple the Seebeck emf occurs across each thermoelement seperately.
4. Junctions are just to provide electrical contact and nothing to do with
the Seebeck emf.
5. By the principle of Seebeck effect it is impossible to measure the absolute
emf of a
single thermoelement as long as there is a corresponding emf from other
thermoelement.
6. Absolute emf measurement is possible only when the other thermoelement
is in a
superconducting state(!)
7. One has to use proper term at proper place while dealing with this thermoelectric

effects.

Some of my pionts and views may be incorrect. It is purely based on my
openion. Comments and suggestions are welcome.

------ Sivakumar ----

References

1.Manual on The Use of Thermocouples in Temperature Measurement. Fourth Edition.

American Society for Testing and Materials(ASTM) Publications, ASTM Manual
Series
(1993)
2. "The Measurement Instrumentation and Sensor”- Handbook, by John.C.Webster.
CRC Press in cooperation with IEE press-1995. Section 32).
3. The Review of Thermoelectricity by Abram F.Joffe. Scientific American
-1958
4. Thermoelectricity in metals and Alloys . R.D.Barnard . Taylor and Francis
Ltd.
U.K.1972
5. Methods of Thermoelectric Characterisation . V.Damodara Das. Aparna Publications

India 1993
6. Thermoelectricity an Introduction to the Principles - D.K.C.MacdonaldJohn
Wiley &
Sons 1961
7. Thermoelectric Power of Metals - Frank J.Blatt, Peter A.Schroeder, Carl
L.Foiles and
Denis Greig - Plenum Press 1976.

What is the physical meaning of electrical power factor defined as electrical conductivity multiplied by the square of the thermoelectric power?

I will be grateful to anyone who could be of help. Thanks.

Department of Physics
University of Ghana
Legon Accra

Legon,

Great question.  To see the physical meaning, look to the definition of the thermal conductivity (k). The ordinary k is measured under conditions of zero electrical current and defined as the ratio of the heat flow (Q) to the temperature difference (dT).

        k_I=0 = [Q/dT]_I=0

But instead let's imagine doing the measurement with zero electric field (E) in the sample instead.  To achieve zero electric field we can imagine 'shorting out' the sample by placing it in parallel with a wire with negigable electrical resistance (r) and negligable Seebeck coefficient (S).  A superconductor would fit the bill.  Or, for most ordinary thermoelectric materials,  a copper wire will do.

What I've described is a thermocouple shorted at both the hot and cold side.

Now, we can define a new thermal conductivity measured at zero electric field:

        k_E=0 = [Q/dT]_E=0

Naturally,  k_E=0 > k_I=0 because the electrical current circulating throught the thermocouple carriers heat, due to the Peltier effect, from the hot to cold side.

A short derivation using only the definitions of these coefficients will show:

       k_E=0 = k_I=0  +  S²T/r

where the second term on the RHS is your electrical power factor, multiplied by the absolute temperture (T).

So, the electrical power factor represents the additional heat flowing through material with zero electric field, compared to the ordinary thermal conductivity.

There is a close parallel here to the specific heat under constant pressure which is always larger than the specific heat under constant volume because under constant pressure conductions some of the heat input is used to expand the material.

 I have a badly written paper that expands on this, which I've attached below:

C. B. Vining, "The thermoelectric process," in Materials Research Society Symposium Proceedings: Thermoelectric Materials - New Directions and Approaches, vol. 278, T. M. Tritt, M. G. Kanatzidis, H. B. Lyon, Jr., and G. D. Mahan, Eds. Pittsburgh, PA, USA: Mater. Res. Soc., 1997, pp. 3-13.