In our laboratory crystal grown by VDS, we have grown doped InSb and GaSb bulk crystals. Some crystal showed decrease in band gap while other crystal showed increase in band gap. Normally, it is known that the band gap decreases after doping.
The band-gap is a region in energy with no allowed states. The density of states versus energy depends on the chemical composition of the material; if the chemical composition is changed, at least, in principle, the state density distribution should change. Dopants are impurities, so, the chemical composition changes by doping. The change in the energy distribution of the allowed states cannot have a general rule such as: the band-gap will decrease after introduction of impurities. Generally, those impurities are called dopants, which create allowed shallow states in the band-gap. Shallow states have small ionisation energies; and, when the doping density is high, the dopant states generate a band. If this band is very close to the valence or conduction band edge, the band-gap will decrease.
The question was not well defined. Were these observations made for any impurities? Impurities have to satisfy a number of criteria to qualify as dopants.
The increment of band-gap energy in n-type semiconductors is due to Burstein-Moss effect, because there's a charge carriers excess and the Fermi level shifts "inside" the conduction band.
Do you intentionally dope the semiconductor? What is the resistivity/sheet resistance?
How do you measure the bandgap?
The band-gap is a region in energy with no allowed states. The density of states versus energy depends on the chemical composition of the material; if the chemical composition is changed, at least, in principle, the state density distribution should change. Dopants are impurities, so, the chemical composition changes by doping. The change in the energy distribution of the allowed states cannot have a general rule such as: the band-gap will decrease after introduction of impurities. Generally, those impurities are called dopants, which create allowed shallow states in the band-gap. Shallow states have small ionisation energies; and, when the doping density is high, the dopant states generate a band. If this band is very close to the valence or conduction band edge, the band-gap will decrease.
The question was not well defined. Were these observations made for any impurities? Impurities have to satisfy a number of criteria to qualify as dopants.
Samares Kar is exactly right: there is no general rule for the bandgap to go up or down when a material is doped.
The direction and magnitude depends sensitively on the dopant type and concentration.
Doping always brings some changes to the host. Whether or not the bandgap shift depends how accurate you can measure it or how much you care. With sufficient accuracy, one could see bandgap shift with doping level as low as 10^16 cm^-3 in GaAs when doped with N (see Zhang et al., Impurity perturbation to the host band structure and recoil of the impurity state, Phys. Rev. B 68, 75210, 2003).
This question is somewhat related to another question posed on this site: "Any novel way to determine energy band gap from transmission and reflection of film?" You might find some answers posted there useful.
Lior Kornblum
Do you intentionally dope the semiconductor? What is the resistivity/sheet resistance? How do you measure the bandgap?
Dear Lior Kornblum
Thank you for query i
We have grown intentionally lightly doped InSb and Gasb bulk growth by vertical directional solidification (VDS). The resistivity/sheet resistance for these growths is either higher or lower than the pure (undoped) growths. The band gap has been measured by FTIR, Hall effect and Resistivity Vs temperature.
With regards
Dr D B Gadkari
I am not familiar with band gap measurements using FTIR, but did you consider a defect band inside the band gap that may give an excitation at sub-gap wavelengths?
A hint for that may be also hidden in the carrier density; is the density extracted from the Hall measurements in the same range as nominally intended? Do you obtain the nominal band gap when measuring undoped samples?
The effect described by Belen Aleman would be relevant for a relatively large doping, so perhaps you have unintentional doping (defects, stoichiometry...) where the gap is larger, and electronic defects where it is smaller.
The gap widening is due to Burstien- Moss effect. The band gap increases due to shift of Fermi level to the conduction band and hence occupies some of the bottom states in the conduction band, therefore now for a transition from valance band to conduction band it requires large energy.
What is the effect on band gap Eg by doping or by adding impurities in a material ? whether bandgap will decrease or increase?
Hi all, I am working in GaN devices. I have doubts regarding doping in GaN layer. Is any possibilities for boron doping in GaN layer? Whether it will acts a n type or p type dopant?
According to Sze (Semiconductor Devices, chapter 1) the bandgap should be decreasing. If doping is high then the bandgap narrowing effect is significant. For Si at room temperature Delta_Eg=22(N/10e8) meV. So for example if NdNc=2.86e19 the ratio is 11%..
Zainab Raheem,
I can read only incomplete sentence as below--
-- the increase in band gap with reduction in size
Please provide complete answer.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Semiconductor