Dear Youssef Essam ,

Your graphs are very odd. These steps in the curve are uncommon. Could you share your DRS analysis? Did you used Kubelka-Munk theory? It seems that your DRS scan step is large, and your intensity is waaay above some points.

If you share the DRS spectrum, it will be easier to help you. And which material are you using?

Gustavo Henrique de Magalhães Gomes

Yes, I used the Kubelka-Munk method

I have synthesized 2 different nanocomposites the attached JPGs are for the first one (copper)

The 2 nanocomposites are:

N-Cu-Cd doped TiO2

N-Co-Cd doped TiO2

The UV DRS results that I obtained where wavelength vs absorbance so I had to convert absorbance to % reflectance, but I did not really find a straight forward way to do so. I used R = 1- (A+T) and calculated %R afterwards.

I have attached several files.

The data after conversion to %R along with the original,

And the origin lab file containing the Kubelka-Munk function calculations

Youssef Essam

I now know what is wrong with your analysis.

There is no straightforward way to transform absorbance into Reflectance. In your calculation R = 1- (A+T) you are not considering a lot of light phenomena like diffraction, inelastic scattering of light and so on, so the Reflectance cannot be calculated this way.

I recommend you to use the Absorbance spectrum. I've plotted the graph and it is very well behaved. So you can perform the Tauc method directly by the absorbance graph employing only the Tauc method. Using the Beer Lambert Law you can consider that the molar absorption of your particles are directly proportional to the absorbance, and then plot using the Tauc method. You can follow the method of Swanepoel to estimate the absorption coefficient :

Article Determination of the Thickness and Optical Constants of Amor...

To summarize, do not transform Abs to %R, it is impossible in a simple way. Use Abs to calculate the bandgap.

You can check these articles for bandgap calculation as well:

Article Characterization and application of niobium-doped titanium d...

Regards,

GG.

Gustavo Henrique de Magalhães Gomes

Yeah, i figured that this approximation was highly inaccurate and that it is the thing that is causing the plot to be so noisy. Unfortunately, the data I obtained were for a powder sample and I only have the powder sample. I was aware of the normal tauc plot method using absorption, but all the references I've found including these were performed on thin films not powder, so I can't use it because I don't have a value for thickness. I am kind very short on time so redoing the DRS testing and asking for reflectance this time is probably not an option. Is there any other way to work around this?

Also, If you can direct me to a way to find out the mode of transition from the raw data I would be grateful

Youssef Essam

You can calculate approximating the absorption coefficient from the absorbance values. You can follow a tutorial here:

https://www.youtube.com/watch?v=LLESVSsZmlg

Also there is a discussion here:

https://www.researchgate.net/post/What_calculations_are_available_for_determining_the_Bandgap_of_nanoparticles

Gustavo Henrique de Magalhães Gomes

I have one last question. After I construct my tauc plot How do I know which best fits the X-axis in order to determine the electron's transition mode?

and gain thanks alot for your help

You can take a linear regression on the energy that the slope changes, and it starts to absorb light. Check the R² obtained from the linear, and calculate the Bandgap. Repeat these procedure several times changing the points, to see if there is such a difference, and then you can calculate also a standard deviation.