Why does an electric current flow in metal plates when there is a
temperature differential between two of the surfaces?
A simple device placed on a wood stove ran a fan. It showed that
thermal energy could be turned into electrical energy. Internet
information described the process but not in terms that I could
explain to a twelve year old. How efficient can it be and how do I
explain it simply?

" EVERYONE knows that heat can produce motion. That it possesses vast
motive-power no one can doubt, in these days when the steam-engines is
everywhere so well known..."
 http://www.history.rochester.edu/steam/carnot/1943/Section2.htm

Question is " How good can a steam engine be?"
 
  The question 

"How efficient can a  thermal machine be?"
was answered by Sadi Carnot in his famous treaty:

  Réflexions sur la puissance motrice du feu, published 1824
 which is concerned with the relation between heat and mechanical energy. 
 
 Formula is remarkably simple, elegant - and one of the cornerstones of science
 called Thermodynamics 
(called incorrectly : It is really about thermostatics only)
 Formula applies to all conversion, heat to mechanical, electrical,
indeed any form of energy.
 
 
 The efficiency ? of a heat engine is defined as the ratio of the work
done on the surroundings to the heat input.
 It is always less ten 1. It depends on the ratio of absolute temperatures of the 
                   Heat Sink  (cooler)  Tc
              and Heat Source  (furnace) Th
	      
	efficiency =   eta = 1 - Tc/Th  
  
  as written in a nice way by eq (3) here: 
 http://www.answers.com/topic/carnot-heat-engine
 

 Why he Studies the question:
 The problem occupying Carnot was how to design good steam engines.
Steam power already had many uses - draining water from mines,
excavating ports and rivers, forging iron, grinding grain, and
spinning and weaving cloth - but it was inefficient. The import into
France of advanced engines after the war with Britain showed Carnot
how far French design had fallen behind. It irked him particularly
that the British had progressed so far through the genius of a few
engineers who lacked formal scientific education..
 http://www-groups.dcs.st-and.ac.uk/~history/Mathematicians/Carnot_Sadi.html
 

Personal and political biography 
http://www.corrosion-doctors.org/Biographies/CarnotBio.htm


Carnot'd famous  paper and historical context is now on line, here:
http://www.thermohistory.com/


SEARCH TERM: Sadi Carnot

For a 12 year old (or son and dad team) I would recommend suplementsupplementary

SEARCH TERM : Stirling Engine   or
://www.google.com/search?hl=en&ie=ISO-8859-1&q=Stirling+Engine&btnG=Google+Search

It's something you can build at home
http://users.bigpond.net.au/mechtoys/stirling.html

Hedgie

P.S. Commentators are begged to refrain from pointing out that 'heat pump'
can have a different ration defined, which can be larger then 1. Heat pump
follows the laws of thermodynamics, the second law in particular.
The  COP (Coefficient of Performance) - the ratio of heat output, to
electrical energy input can be larger then 1. It is a different
quantity then eta.

---

Request for Answer Clarification by boob4u-ga on 22 Feb 2005 16:15 PST

Hedgie-ga thank you very much for your efforts. However I am looking
for information on thermal to electric. Your research while very
interesting does not help me with my question. Xarqi-ga is right, the
nanothermel site is the direction that I was inquiring about. It had
forgotten that the Voyagers used this generator and I apologize for
not mentioning that it was on the Cassini probe. I did enjoy the
information in the Science and American Scientist articles, but I am
still left with the problem of what to tell my twelve year old. She is
bright but not ready for the math required to understand n and p-type
materials. Is there a source that will help me simplify the
explanation?

---

Clarification of Answer by hedgie-ga on 22 Feb 2005 21:39 PST

boob4u-ga 
            I am aware of the fact that I did only covered  some  of
your questions.
	    
	    You have asked:

 Why does an electric current flow .. ?
 How efficient can it be and 
 how do I explain it simply?
 
 I answered the middle one - the efficiency of energy  conversion devices
 is subjected to Carnot's constraint. It covers the fan, the Peltier effect
 Seebeck effect-- or rather the Peltier-Seebeck effect - for they are
one and the same, and are one  example of the at least 200 such cross
effects, each having a name (or two names), history and structural
theories.
 
 Your last question is particularly hard, but I hope you appreciate
the effort. I will not answer it directly - but I do keep the age in
mind.
 
 They are different approaches to physics, to study and learning of
science in general. One is to collect  zillion of examples, to
memorize them, in a manner
 of a man assembling a butterflies collection, and the other is to ask "why".
 
 "Why" is a question for 'deeper theory' - which is of course physics,
and it comes  in two forms, one is phenomenological, the other
structural theory.
 
 The structural explanation is reduction - the statistical mechanics -
which explains e.g. Gas Law in terms in terms of molecules, atoms and
properties of solids in terms of p,n junctions ..
 
 The phenomenological explanation - in this case Linear Irreversible
Thermodynamics - offers a different answer to 'why'. It explains that
there are 'fluxes' = and 'forces' or thermodynamical potentials - and
these two are connected by (usually) linear coefficients  L(i,j). (see
page 4 here:)
 http://www.mpikg-golm.mpg.de/gf/Trnsp Procs & Irrev Thrmd, Lazar P, 14 May 04.pdf
 
 There are some general laws, such as Osanger Symmetry, which say
L(i,j)=L(j,i) , which in our case means, if there is a Seeback effect,
there must be Peltier Effect, (and vice versa)  and the coefficient is
same.
 See page  17, and 18 (etc) of 
 http://ocw.mit.edu/NR/rdonlyres/Materials-Science-and-Engineering/3-205Fall2003/018C70CD-7CE5-43F6-91EB-B853173D9A72/0/lec01_03.pdf
 
 Page 18 lists some direct effects (e.g. Ohm law connects ELECTRIC
current to ELECTRIC potentials)
 
 Then there  are  cross=effects
 http://www.mpikg-golm.mpg.de/gf/Trnsp Procs & Irrev Thrmd, Lazar P, 14 May 04.pdf
 
  The cross effects are common -- actually, there is a law, Curie's or
restriction, (same Curie) which explains why some of them are missing.
  
  So, I am afraid, the answer to your question 1 is 'why not?' = 
  from phenomenological point of view.
  
  A general structural explanation of cross-effects, 
  such as electro-thermal  effect we are dealing with here, is
  that some molecules (and so on) have multiple properties 
 (magnetic,electric, mechanical ..) and so transport of molecules
caused by one gradient induces fluxes of several properties.

  The p-n junctions are not necessary, just convenient. The effect
exits even in gases
  http://www.iop.org/EJ/abstract/0022-3727/28/7/009
  but is weak.
  
  Here is an article, which explains why some materials - alloys of
bismuth telluride (Bi2Te3),  are more active (Fig 7) then others.
  http://jchemed.chem.wisc.edu/Journal/Issues/1996/Oct/abs940.html
  and it also explains  Relation to Thermodynamics and Carnot efficiency.
  
   Ultimately,(and if your daughter has a mind of a physicist - that
what she is curious about) answer is  similar an answer to question :
why some crystals are (more) conductive, piezo-electric,
electro-strictive, ferromagnetic .. (then others?)
   { That answers is "atoms did it" - different atoms just hold on
their electrons differently - and when they (atoms) join forces - in
molecules and crystals - you get a variety of effects, which all have
names, and all are describes by thermodynamics}
   
   So - what do you want for $4 - exactly?
    
  The p and n junctions are just a way how some structures holds on
their electrons, (and lacks of electrons   (holes) ) and how these
travel in solids.
    I would not discuss that with young people before they had at
least conceptual course of classical physics. Even a bright 12 year
old has no good reason to believe in atoms. Do not spoil the journey
of discovery by providing spoilers. Concept of entropy did come before
the concept of atom and electron.
    
    Hedgie

---

Request for Answer Clarification by boob4u-ga on 23 Feb 2005 16:39 PST

Thank you very much for your efforts Hedgie. I was hopping that I
could find something that would explain the Peltier and Seebeck
coefficients in the simplest terms. I wanted very much to make my
explanation as accessible as possible. Thank you again for your
efforts.

---

Clarification of Answer by hedgie-ga on 23 Feb 2005 23:21 PST

As I said, there is some 200  cross-effects, which convert energy between fluxes

this site lists some of them

http://www.control.hut.fi/Kurssit/AS-74.136/materials/effects_t.pdf

Here are few more references on Seeback-Peltier effect (at the bottom)

http://www.searchspaniel.com/index.php/Peltier-Seebeck_effect

 Here is a simple description
 
http://arstechnica.com/guide/cooling/peltier-1.html

 Perhaps, if you need more info, you may elaborate the question a bit more.
 Sometime, art of asking a question is half of an answer.
 Perhaps I am not sure what word explanation mean, perhaps this RFC was
 just a 'thank you note'?  However, we have to respond to a RFC.
 RFC= request for clarification

---

Request for Answer Clarification by boob4u-ga on 24 Feb 2005 13:39 PST

You were right. The key word was Anders Killander. I do appreciate
your help and Thank you for the work.

---

Clarification of Answer by hedgie-ga on 25 Feb 2005 10:47 PST

You are welcome.
Thanks for the rating.

Thank you again for your help.

Stirling engines are all very, well, sterling I guess, but is your
question not about the thermoelectric effect as described here:
http://www.nanothermel.org/public_main.htm ?