vinhphysics said:
The peaks means we can find some maximum points on the spectrum (See absorption spectrum in link: http://www.viewsfromscience.com/documents/webpages/nanocrystals.html ). The Band means no maximum points can be found on the spectrum (See absorption spectrum in link:
http://www.springerimages.com/Images/RSS/1-10.1007_s13204-011-0005-4-3)
I don't think you are using the word "band" correctly. A "band" can have a maximum.
You probably mean "low energy threshold". For example, I was working with spectra from an impurity in a semiconductor. The emission intensity showed a relative peak at a photon energy of 105 meV. However, the absorptivity associated with this impurity showed a low energy threshold at 105 meV.
The "threshold" means that the impurity did not absorb radiation with a photon energy less than or equal to 105 meV. However, the impurity did absorb radiation with a photon energy greater than 105 meV. The absorptivity of the impurity increased monotonically with photon energy. Therefore, there was no relative maximum in absorptivity. The absorption spectrum showed a step at the threshold, not a pointy hat.
You are asking why some transitions are associated with peaks and some with thresholds. Both spectra are associated with electrons that are being boosted in energy.
In spectra associated with peaks, one energy level is "catching" all the electrons. The electrons lose energy, but there is a bottleneck between the initial energy and the ground state. The narrow peak is caused by transitions from this bottleneck to the ground state (or some lesser excited state).
That is why my emission spectrum showed a peak at 105 meV. I boosted the electrons very high in energy. However, the bottom of the conduction-band caught the electrons before they got to ground state. The final transition was from the conduction band to the impurity state.
In spectra associated with thresholds, there is no bottleneck. There is a continuum of energy states that can receive the electron. The density of this continuum generally increases with photon energy. The threshold is associated with the bottom of this continuum. Therefore, the absorption goes up.
That is why my absorption spectrum showed a threshold at 105 meV. The conduction band is a continuum. However, there is no bottleneck to absorption. So the bigger the photon energy, the bigger the absorption.
Note it was the same impurity in both cases. Even the same energy (105 meV). However, the emission process allows some time for the electron to be caught by a bottleneck. The absorption process did not allow any time for the electron to be caught.
This is an old article. However, I posted it on ResearchGate. If you go to the ResearchGate website, maybe you can get a free or inexpensive copy. I don't know if they charge, so I won't promise anything. The article does refer to both spectra of the 105 meV impurity.The reference is:
"Native defects in undoped semi-insulaing CdSe studied by photoluminescence and absorption," by D. L. Rosen, Q. X. Li and R. R. Alfano, Physical Review B31, 2396-2403 (15 February 1985).