Hi,

I think its the very first peak. But I wanted to know that what was the excitation wavelength used for the PL experiment? Because, the band gap of ZnO should appear near ~380nm. Higher order peaks may be originated from the defects levels.

@avishek

The excitation wavelength used was 365 nm (3.39 eV)..

In this analysis, I didnt obtained any peak in UV range..(there has no peak appeared at 380nm++).The peaks only appeared in visible range as shown in figure. 

Is it okay for me to denote the 1st peak in the visible range (as shown in the figure) as the energy band gap,Eg? Because I thought..from what I read, the Eg value is related to near band edge emission which appeared in UV range. 

Band gap of ZnO is 3.3 eV which corresponds to 375nm.  Your first peak is at around 3.04 eV. I highly doubt that it is band gap. In PL you get an excitation which is very close to your conduction band min. So the value should be very close to 3.3eVEven then electrical and optical bandgap shouldn't be confused. The UV emission peak at about 408 nm (3.03 eV) corresponds to the near-band-edge emission of the ZnO crystal.

I believe in the following paper you will find the answers you are looking for

http://www.beilstein-journals.org/bjnano/single/articleFullText.htm?publicId=2190-4286-5-140

In textbooks band gap appears to be very precise, but experimentally depending on the measurements you are making you can get slightly different answers depending on the method of measurement, or on the details of the material. Also for PL measurements it is good practice to note the temperature and wavelength of the measurement.

I think the first problem with your PL curves is that you need to scan from about 300 nm to 600 nm to get a better idea of what is going on. You may have been limited by the laser or light source you were using, but since the scan's are starting at 400 nm it is hard to know if you were exciting the sample above the band gap.

Your first impulse to say the highest energy peak was connected to the band gap was correct, you do not sum up or add up the values of the peaks. 

I am not sure without knowing more about your samples or experimental set-up what the origin of the different peaks are. You might want to normalize the intensity of the red and green curves and see how similar they are. 

Using the Tauc method for band gap is an ok way to compare samples of a similar thickness. However, it is not the best way to determine the band gap from absorption measurements since it doesn't take the exciton into account. Also for your sample you are probably averaging a lot of different size nanowires in your measurements and they may have slightly different band gaps, or doping levels. 

Ideally you should have a direct connection between the absorption spectrum and the PL spectrum through the principle of detailed balance. With a single crystal film you can see that in the attached papers. For nanowires it is more complicated, but for very pure nanowires you should see something similar, a pretty sharp absorption band edge, and a peak corresponding to the sharp absorption. If you have an impurity like sulfer you can see very strong impurity emission, or often with oxygen vacancies you may see a broad sub bandgap emission. There is a lot of literature on ZnO nanowires that you can read about.

Article Absorption coefficient, energy gap, exciton binding energy, ...

Article Excitonic structure and absorption coefficient measurements ...

Thank you for your comments.

Firstly, several methods have been developed for determining bandgap of ZnO, thus leading to the difference in their values

Example, the basic absorption spectrum can be used to determine the value of optical band gap.

The band gap (Eg) of the samples was calculated by plotting the square of the photon energy absorption coefficients as a function of energy.

Or the value of optical band gap (Eg) was calculated from the average radii (R) of the nanocrystallites, band shift and bulk value of ZnO.

Secondly, compared to the optical band gap, The UV emission peak is shifted to the red, due to the Stokes shift effect.

Dzetty soraya abdul aziz When a semiconductor or a dielectric material is excited above the bandgap energy and as a result of relaxation, luminescence is observed is called photoluminescence. As a result of relaxation, if only photons are emitted, the bandgap will be a "direct" bandgap, while, if in addition to photons, phonons are also emitted, the bandgap will be an "indirect" bandgap. Since, the phonons emission is due to the defects, impurities, and dopants between CB and VB. When only photons are emitted, the relaxation is called radiative, while, when only phonons are emitted, the relaxation is called nonradiative. We can only find the bandgap energy from PL data in the case of radiative relaxation (direct bandgap). In nonradiative relaxation, when an electron passes its energy to the phonons, is called Shockley–Read–Hall recombination and when a relaxing electron passes its energy to another electron or hole is called Auger recombination. From PL data, intensity, line edge, and FWHM are the characteristics parameters i.e. higher intensity means low defect density and greater FWHM gives poor structure and vice versa. In addition to all the above, the energy bandgap calculation from PL data is subject to many limitations. For example, PL emissions do not give an exact bandgap like UV-Vis absorbance (Tauc plot). The bandgap calculated by the PL study will always be less than the original bandgap. Emission spectra are usually solvent dependent and are shifted with the change in solvent polarity due to solvent relaxation, the excitation spectra are usually preferred for bandgap calculations.

In the following video, I have explained all the above discussions in detail. Links to the files used in the Origin tutorial video have been provided in the video description. Thanks

https://youtu.be/WCPSil8uryg

first of all in PL we give an excitation frequency which we are taken from Uv-visible data and the emission spectra we obtain also gives us the the information about band gap but to notice the temperature at which we are doing PL