Hi Igor,
one obvious difference between your experimental and calculated densities is 
that the diffraction experiment averages over zero-point and thermal vibrations 
(what was the temperature ?).  The DFT density was (presumably) calculated for 
the equilibrium structure, i.e., for atoms at rest.
For a better comparison, you could run an MD and average the the densities of a 
series of snapshots.
Alternatively, you could repeat the experiment at lower temperature.

Regards, Ulrich.
----- Original Message -----
From: Igor .
[mailto:[email protected]]
To: [email protected]
[mailto:[email protected]]
Sent: Tue, 18 Feb 2014 18:51:20 +0100
Subject:
[SIESTA-L] Doubt about charge density by rho2xsf and MEM (maximum entropy
method)‏


> Hi siesta users,I work with x-ray diffraction in crystals, with the
> diffraction data I can generate electron density maps using the technique of
> MEM (maximum entropy method), I would like to do some tests and compare
> electronic densities generated by MEM and DFT by Utility rho2xsf.I have
> chosen a simple material and calculated the electron densities by two
> methods. I did it with several materials. The figures of bulk SiO2 and also
> the fdf are attached. I used the pseudopotential on the site.I realized that
> the electron density generated by DFT have lower values compared to the
> isosurface generated by the MEM.What are the reasons for the two results
> being different ? The DFT gives the electronic density for the ground state,
> how different is the electronic density at room temperature in comparison to
> the one ?What can I do to have something more reliable?
> 
> 
> Best regards                                    

---
Max-Planck-Institut für Eisenforschung GmbH
Max-Planck-Straße 1
D-40237 Düsseldorf
 
Handelsregister B 2533 
Amtsgericht Düsseldorf
 
Geschäftsführung
Prof. Dr. Gerhard Dehm
Prof. Dr. Jörg Neugebauer
Prof. Dr. Dierk Raabe
Prof. Dr. Martin Stratmann
Dr. Kai de Weldige
 
Ust.-Id.-Nr.: DE 11 93 58 514 
Steuernummer: 105 5891 1000
-------------------------------------------------

Responder a