Dear Colleagues!
Recently I have been trying to calculate the electronic band structure of a organic-inorganic lead halide perovskite structure, namely CH3NH3PbBr3.
I am not an expert in band structure calculation and I only worked with Wien2k years ago. Unfortunately I don't have access to that license anymore and I am now learning Quantum Espresso.
I tried following the work of Y.H. Chang (Journal of the Korean Physical Society, Vol 44, No 4, April 2004, pp 889~893) and a more recent one by Mosconi et al. (Phys Chem Chem Phys 2014, 16. 16137), however when I am trying to structurally optimize my starting position the calculation goes haywire and all my hydrogen atoms vanish and the calculation runs until nstep.I tried aligning the C-N bond of H3C--NH3 along the cubes basic (100 direction) axis, and alongside the diagonal (111 direction), but this does not help. Mosconi gives a different starting configuration but I did not figure out how to use it properly (I am really new to this kind of material)!
If someone could help me with a reliable starting position that would extremely glad (if this results in anything useful proper credit will be, of course, given, but so far I am mainly playing around out of curiosity).
For the MD calculation I used the example03 that comes with QE.
Best,
Chris
Hi Christoph
May this paper and site can help you very much to know the answer of your question:
1.Journal of Physical Chemistry C
Principles of Chemical Bonding and Band Gap Engineering in Hybrid
Organic−Inorganic Halide Perovskites
Aron Walsh*
Centre for Sustainable Chemical Technologies and Department of Chemistry, University of Bath, Claverton Down, Bath BA2 7AY,
U.K.
2.https://thelostelectron.wordpress.com/2014/06/22/crystal-structures-of-hybrid-perovskites-are-not-0k/
Best regard
I don't know the specific configuration, but what I do is I build my molecules (I just make the approximate molecule) using Gabedit. Then, after building it, Gabedit allows me to run a small molecular dynamics optimization (using either quasi-newton or steepest descent), and then I use that optimized molecule as the input into my quantum chemistry program. It usually helps. Avogadro has the same feature, and they're both free programs. Avogadro is a little easier to start using, I think.
Hi christoph, if you are using a potential, you can write by your self a program which can generate the initial configuration by MD simulations apply Verlet algorithm. The second possibility is to build your molecule using gaussian , then you optimized and collect the optimized configuration , then use it in the program that you want to use
Sincerely yours