If you need a dense k-mesh to calculate thermoelectric properties, you can use the utility program BoltzTrap which is compatible with most of the DFT codes such as WIEN2K, VASP, QE, etc.:

https://www.imc.tuwien.ac.at/forschungsbereich_theoretische_chemie/forschungsgruppen/prof_dr_gkh_madsen_theoretical_materials_chemistry/boltztrap/

Alternatively, you can try Wannier90 which does this for you using maximally localized Wannier functions in real space and without any k-mesh sampling: http://www.wannier.org

I personally prefer the latter.

Respected Mohammad Saeed Bahramy , Thanks for your consideration and time. I haven't use wannier before. How to do this with wannier. Your help and guidance will be appreciated.

Dear Nisar Muhammad

Wannier90 has a very well-documented manual and some examples helping you to set up your own inputs: http://www.wannier.org/support/

In brief, you first need to perform an standard self-consistent ab-inito calculations using VASP, QE, WIEN2K, ABINIT, ELK and etc. After that you will perform a non-self-consistent calculation to create a tight-binding hamiltonian for a set of bands relevant to your purpose using maximally localized Wannier functions. For this step you need to prepare one (or depending on the ab-inito program you use, a few) input file(s) that will be processed by Wannier90 to generate the output files you need.

One of the interesting features of the latest version of Wannier90 is that It can calculate all sorts of transport properties including electrical conductivity and Seebeck in the real space and without any need to k-sampling. Of course, you can also use this program to plot DOS and other band structure properties on a fine-kmesh as well. Here is a full list of features available in the latest version of the program:

http://www.wannier.org/features/

Respected Mohammad Saeed Bahramy . Thanks for your time and help.