The Naked Scientists

The Naked Scientists Forum

Author Topic: Why does the electrical resistance of metals and semiconductors change with temperature?  (Read 31505 times)


  • Guest
navaneethan asked the Naked Scientists:


If we increase the temperature of a metal such as copper and also a semiconductor, why does the resistance of the metal increase, whilst the semiconductor's resistance decreases?

What do you think?



  • Guest
Hi navaneethan
Simple explanation for metals: at higher temperatures, the atoms are vibrating more and are more likely to be hit as the electrons go past. Hence higher resistance.
This is flawed really because vibrating doesn't actually change the collision cross section of an object.
A better explanation is on the level of quantum theory. For an electron to interact with an atom and lose some of its energy, there needs to be an appropriate energy level change. When you increase the temperature, the distribution of energy levels in the atoms widens and there are more possibilities of interaction between an electron and an atom. Hence the energy is lost quicker - which is the same thing as increasing the resistance.
The first explanation seems to be enough for most people tho'.

For semiconductors, you can get the reverse effect (negative temperature coefficient) because, unlike in a metal, there are few conduction electrons available. Heating the material up can make more electrons available (as can shining a light on the surface) which will decrease the resistance. But, of course, the resistance of semiconductors is vastly higher than the resistance of metals because there are relatively so few electrons available to conduct.

Offline Pumblechook

  • Hero Member
  • *****
  • Posts: 569
    • View Profile
It is a chalk and cheese comparison.  Different mechanisms are involved.  If I remember correctly pure semiconductors (e.g pure silicon) is essentially an insultor and relies on impurities (boron, arsenic) to allow conduction. The behaviour of lumps of silicon with metal terminations will be different from P-N junctions.  You need to read up on minority and majority carriers, holes and electronics, P type, N type, diodes, bipolar transistors, junction FETs, MOSFETS and thermal runaway.   Keep you busy for a while. 


  • Guest
Yes, indeed.
The movement of current through N and P type semiconductors is still by electron movement, of course. But the physics involved doesn't have the equivalent to the simple picture that we use for metallic conduction. The idea of 'holes' propagating through a material and carrying a current is bizarre but it works very well. Holes even behave as if they have mass.

Offline Quantumorigin7

  • Full Member
  • ***
  • Posts: 80
    • View Profile
I'm reading a book called The Path To No Resistance: The Story of The Revolution In Superconductivity" by Bruce Schecter. It's a book to read on free time, just to enjoy. But basically, an early superconductor I believe was composed of Lanthanum, barium, carbon and oxygen. Excuse me if I'm wrong. As far as I have read, there are 135-140K. superconductors. There could be room temperature superconductors now for all I know.

The Naked Scientists Forum


SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums