The simplest way is to take your film, record the UV-vis-NIR absorption spectrum, take the wavelength when the the absorption intensity start to take off (If the intensity does not take off sharply, you can plot the intensity in logarithm). Then you can derive the optical bandgap using equation: Eg (eV) = 1240/(wavelength in nm). For example, if the intensity arises from the baseline at 620 nm, then you have a material with bandgap of 2 eV. Cheers!

Hello Savita Maurya,

What kind of a UV spectrum is it? - Transmission (absorption), or reflection, or photoluminescence? You need to make your question more informative to get helpful responses.

Anyway, in general, it is not easy to identify the optical bandgap unless you know the calculated band structure, or at least the density of states diagram of the material. This is because, say for example, for the absorption spectrum, a peak or shoulder is due to a transition between two bands, but not necessarily at the same value of k. You can, however, make guesses. For a direct (optical bandgap) transition, the threshold of transition as a function of wavelength is much steeper than in case of an indirect transition where there is involvement of phonons.

Firstly you have to obtain the UV-vis difuse reflectance spectrum. Once you have the data of %R you can transform the data to Kubelka-Mung Function. F(%R) is the way to obtain the band gap. May be even you can obtain the F(%R) from the equipment. You make after, another transformation to energy using plank constant (hv). Like that you could have on the X axis energy on electron volts and on the y axis [(F(%R)hv)1/n].

In crystalline semiconductors, where crystal momentum is conserved and electron transitions obey well-defined selection rules, n is 1/2, 3/2, 2, and 3 when the transitions are direct-allowed, direct-forbidden, indirect-allowed, and indirect-forbidden, respectively. After that you make a line to fit the curve that you obtained

and the intersection with x axis wil be the band gap energy. You can see J. Phys. Chem. B, Vol. 103, No. 4, 1999 for the extended method. :D

The simplest way is to take your film, record the UV-vis-NIR absorption spectrum, take the wavelength when the the absorption intensity start to take off (If the intensity does not take off sharply, you can plot the intensity in logarithm). Then you can derive the optical bandgap using equation: Eg (eV) = 1240/(wavelength in nm). For example, if the intensity arises from the baseline at 620 nm, then you have a material with bandgap of 2 eV. Cheers!

In order to find the value of Eg make use of the Tauc relation

αhν = A (hν - Eg )^m

where α is absorption coefficient given by α = 2.303 log (T/d)(d is the thickness of the sample and T is the transmission ), hν is the photon energy.

Plot the graph between (αhν)^2 versus hν . The values of Eg have been estimated by taking the intercept of the extrapolation to zero absorption with photon energy axis i.e. (αhν )^2 → 0 .

Hi Guys,

I think that the question is not just about a formula to calculate a transition energy from an absorption spectrum, but to find the 'optical band gap'. Two things are important here.

1) It needs to be identified if a threshold is associated with a direct (k=0) transition,

2) Which bands are involved in the transition?

It is essential to compare the experimental spectrum with a band structure of the material concerned to get a correct answer.

go through this journal,

1983 J. Phys. E: Sci. Instrum. 16 1214

Thanks to all

I too have HfO2 thin films on Si. I think UV-vis method is hardly useful for very wide bandgap material, maybe 'far' or 'near UV -UV-vis' spectrometry may help (if there's one). In normal UV-vis, the data become unreliable for wavelength shorter than 150 nm. But I did try UV-vis photo spectrometry and all I get was a weird data. I have some absorbance peaking near about 3.5 - 5 eV but then it falls down at higher energy (above 5 eV). Doing a Tauc plot (although one can always do) is physically meaningless unless I explained why it falls down at higher energy (shorter wavelength). Another way though I heard is that of ellipsometry. Anyway i am attaching some spectra if you want to look into it. One is absorbance and the other is corresponding Tauc plot (assuming direct transition). And I think XPS can also give bandgap to some estimate. Suggestions and feedbacks are welcome.

I agree with Shahid Husain and calculate the same way as he introduced.

How about the thin film on Si substrate?

by using DRS UV we can take UV spectra for solid sample

respected sir

Kuan Sun, Shahid Husain and Donya Ramimoghadam please send the reference.

Dear Siva, 

This is the reference for Eg=1240/wavelength.

A. S. Bhadwal, R. M. Tripathi, R. K. Gupta, N. Kumar, R. P. Singh and A. Shrivastav, RSC Advances, 2014, 4, 9484-9490.

what is the basic difference between indirect or direct band gap clearly

 the optical bandgap is calculated using equation: Eg (eV) = 1240/(wavelength in nm)

.

Take the first derivative of Transmittance or Reflectance and look for peak or Valley as the case may be near the region identified throgh literature. If you take the derivative for the whole spectra it will be meaningless as lot of noise will be involved

The spectral data showed the strong cut off at 720 nm(say); where the absorbance value is minimum.

The data is corroborated in the % Reflectance mode.

Calculations

Band Gap Energy (E) = h*C/λ 

h = Planks constant = 6.626 x 10^-34 Joules sec

C = Speed of light = 3.0 x 10^8 meter/sec

λ = Cut off wavelength = 410.57 x 10^-9 meters

H                       C                      λ                     E                               eV

6.63E-34     3.00E+08     7.2E-07      0.000276E-19               1.723

Where 1eV = 1.6 X 10-19 Joules (conversion factor)

 If a cuvette is used for liquid sample to record the absorption vs wavelength spectrum, how do we calculate the sample thickness, d?  

I would like to suggest the tauc relation for the exact calculation of optical band gap experimentally. The uv-visible spectra should be measured for solid sample instead of solution so that the effect of solvent on sample band gap can be avoided. The thickness of solid film should be kept minimum. By avoiding the scattering coefficient the absorption coefficient can be calculated using the equation

α = (1-R)^2/2R where R is the % of reflectance of the sample. I think this is much easier way to measure the absorption coefficient than the relation α = 2.303 log (T/d).

Then plot (αhν)^n versus hν (n = 2 to 0) and take the intercept of the extrapolation to zero absorption with photon energy axis.

Normally after taking the UV-visible spectrum u will get data for wavelength verses absorbance(A).from that data find optical absorption coefficient called alpha which is = 2.303A/ thickness of the sample.Then plot a graph with (alpha * h*new)^1/2 along Y axis and (h*new)along X axis .Now extrapolate the straight line portion of the graph to (alpha *h*new)^1/2=0.Measure that value on the X axis in eV which gives u the band gap.  If u have transmitence instead of absorbance find alpha = (-2.303 log T)/ thichness of the sample.

Dear Dr. Savita Maurya

Experimentally, to get  the value of Eg , we usually use of the Tauc relation, which is given by this equation:                                

 αhν = A (hν - Eg )n

where α is absorption coefficient given by              

α = 1/t  ln (1/T)

where t is the sample thickness, T is the transmission and (hν) is the photon energy.

If any one plot a graph between (αhν)1/n  versus (hν), he can get a straight line. This line intersects the X-axis at (αhν)1/n = 0 .

The values of Eg have been estimated from this intercept.  The value of n is dependent on the electronic transition type. Where:

n=1/2 for direct allowed transition, n= 2 for indirect allowed transition, n=3  for direct forbidden transition and n=3/2 indirect forbidden transition. You should try to select the suitable n according to your samples and their preparations.

The appended papers may help you; they are examples of direct and indirect allowed transitions.

Good luck

Your,

Ahmed Saeed Hassanien

https://www.researchgate.net/publication/283338024_Effect_of_Se_addition_on_optical_and_electrical_properties_of_chalcogenide_CdSSe_thin_films

https://www.researchgate.net/publication/281495242_Influence_of_composition_on_optical_and_dispersion_parameters_of_thermally_evaporated_non-crystalline_Cd50_S50-xSex_thin_films

Article Influence of composition on optical and dispersion parameter...

Article Effect of Se addition on optical and electrical properties o...

Excellent elaboration by Prof Dr. Ahmed Saeed Hassanien

Thank you Dr.  Abdul Ghafar Wattoo for your kind comment about my previous answer. I think that every one should explain in details what he knows in order to help Who asks.

Thank you dr Hassanien for your answer. Its good.

Dear Savita Maurya

You are welcome, I hope  my answer help you.

Good Luck and best wishes

Hello dear Kuan Sun and Ahmed Saeed Hassanien, you both were explained in a simple and easily understandable manner, I got the answer what I was looking for in a few minutes, I thank you very much.. And extended thanks to everybody who shared their answers here... Cheers.........

HELLO..DR KUAN SUN..CAN WE FIND THE ENERGY GAP USING ABSPORTION SPECTA IN SOLVANT ??? OR IT NECESARY USE UV SPECTRUM IN SOLIDE SAMPLE

Respected Kuan Sun Sir, Can we eliminate the absorption effect of substrate material if somebody have not used the transparent substrate in UV-Vis. analysis. Is it possible by taking the absorption spectra of substrate material as reference ? Please suggest....

Dear Surender P. Gaur ji yes you can take substrate as a reference. Just run ablnak scan using substrate as  a reference and then used grown sample on the substrate for comaparison

HELLO.....I NEED HELP FROM YOU FREINDS..CAN WE FIND THE ENERGY GAP USING ABSPORTION SPECTA IN SOLVANT ??? OR IT NECESARY USE UV SPECTRUM IN SOLIDE SAMPLE

Use this below link

https://www.researchgate.net/post/How_to_calculate_band_gapEg_of_Uv-vis_spectroscopy_from_absorbance_vs_wavelength

Daer Israel Perez

@Israel Perez

The average value of the tailing parameter (alpha)o can be obtained from the linear fitting of the figure. In my investigations it was found equaled to 0.249 eV.  There is a linear relationship between band gap energy and the width of Urbach tail which can observed for many semiconductors. My observed change values in Eg and EU are successfully analyzed on the basis of Mott and Davis's mode.

Please,for more detail go to my article titled: 

"Effect of Se addition on optical and electrical properties of chalcogenide CdSSe thin films" Available from: https://www.researchgate.net/publication/283338024_Effect_of_Se_addition_on_optical_and_electrical_properties_of_chalcogenide_CdSSe_thin_films 

Good Luck

Article Effect of Se addition on optical and electrical properties o...

U CAN CALCULATE GAP ENERGY IN THIS CASE WITH TAUC FORMULA

Nice, detailed an clear answers @ Prof. Ahmed Saeed Hassanien.

Nice, detailed an clear answers @ Prof. Ahmed Saeed Hassanien.

Thank you very much Dr. Mostafa A. El-Aasser for your comment and recommendation. You are most welcome.

Hi, I noticed there are two methods here:

1) by using the expression of Eg (eV) = 1240/(wavelength in nm).

2) by using tauc plot relation.

May I know is there any differences between these two types of calculation? What is the pros and cons between these two methods?

Thank you.

Hope the attached article helps you.

https://www.sciencedirect.com/science/article/pii/S0921883117304260