March 28, 2024, 07:20:40 AM
Forum Rules: Read This Before Posting


Topic: How many donor impurity atoms would be required for silicon to be n-type?  (Read 1028 times)

0 Members and 1 Guest are viewing this topic.

Offline Lettuce

  • Very New Member
  • *
  • Posts: 1
  • Mole Snacks: +0/-0
I'm doing some studying right now and there is a question saying "A silicon sample at T=300k contains an acceptor impurity concentration of NA = 10^16 / cm^3. Determine the concentration of donor impurity atoms that must be added so that the silicon is n-type."  Would I be correct to say that you just need to add more than 10^16 /cm^3 of donor impurity atoms?
« Last Edit: June 16, 2020, 11:06:53 PM by Lettuce »

Offline Enthalpy

  • Chemist
  • Sr. Member
  • *
  • Posts: 4041
  • Mole Snacks: +304/-59
That's correct with all the accuracy needed for technological purposes and with reasonable assumptions.

One could argue that some impurities are not fully activated at room temperature. That would be deep level dopants. Normal people don't expect that in such a question.

In some semiconductors, dopants can give or catch two electrons. However, I don't remember any for silicon. Where such dopants exist, the second level is not shallow, and is not activated at room temperature, as far as I know and remember.

Also, some dopants can be added to "compensate" a semiconductor. This is common for GaAs, but I haven't seen it done for Si. It works like a "buffer solution" for chemists: significant doping with an element that introduces a level deep in the forbidden band stabilized the Fermi level at this deep level. You would need the donor doping to exceed the acceptor doping and to saturate the deep level before an N semiconductor appears. However, compensation isn't done at 1016cm-3, normally not in silicon, and this must not be expected in this question.

Sponsored Links