[RMCProfile-users] X-ray PDF use

[Christopher J. Wright]

Dear Dr. Tucker,

Sorry for the late reply, I was applying for graduate positions which
consumed most of my time.  As for my system, the nanoparticles are spherical
with a diameter of 3.5 nm and a composition of Ni30Pd70.  I think with a
nanoparticle this small I should be able to simulate the whole particle,
right?  Since the structure is loosely FCC I would like to use a maximal
coordination constraint, for this system a coordination of 12.  Can this be
done in RMCProfile, and if so how?  Also, what is the formula for the
conversion of the neutron weights to x-ray weights?

Thank you very much and Happy New Year,

Chris Wright

From: matt.tucker at stfc.ac.uk [mailto:matt.tucker at stfc.ac.uk] 
Sent: Friday, December 6, 2013 10:40 AM
To: rmcprofile-users at rmcprofile.org
Cc: christopher_wright at brown.edu
Subject: RE: [RMCProfile-users] X-ray PDF use

 

 

Dear Christopher,

 

1)      Can RMCProfile fit x-ray PDF?: Yes

 

What Chris Kerr says is not strictly true. RMCProfile can fit x-ray PDF
data. 

 

The x-ray F(Q) can be fitted with the XRAY_RECIPROCAL_DATA keyword. However,
the x-ray form factors have to be in the data file and it has to be
convolved with the simulation box size. There is a program call xray_to_rmc
that will do this for you if supplied the data in the usual rmc format.

 

For the x-ray PDF, since the q-dependant part has been divided out you can
treat this as a neutron PDF with different weights. These weightings are
defined using the keyword NEUTRON_COEFFICIENTS (sorry a bit confusing I
know).

 

We are writing a tutorial example to show how all this works and it will be
in the minor update 6.5.2. This will be out in the next few weeks.

 

As Chris Kerr says there are plans to fit the x-ray PDF in a more rigorous
way. Once this is implemented it will be useful for all x-ray PDF data, as
it will make a fairer comparison regardless of the atomic radii

 

2)      How do you fit nanoparticles with RMCProfile?

 

Sadly this is not an easy question to answer, there are three possible
approaches. 

a.      The first is to treat the data as a bulk sample if the particle is
large. If not then a modification function for the particle size could be
applied but this isn't implemented at present.

b.      The second (as Chris says) is to put one particle in a large box and
only fit out to a distance where it does not see its periodic repeat (RMC
uses periodic boundary conditions).  

c.      Finally if you can build a starting configuration with particles in
a matrix that is not too large (> ~40,000 atoms), then you can use the
various constraints/restraints to hold things together and model it as
normal. This of course is not trivial but probably the ideal solution. 

 

If you tell us some more about your system we can advise what your best
approach is most likely to be.

 

 

Best wishes,

Matt

 

 

. 

 

 

 

 

 

 

From: rmcprofile-users-bounces at rmcprofile.org
<mailto:rmcprofile-users-bounces at rmcprofile.org>
[mailto:rmcprofile-users-bounces at rmcprofile.org] On Behalf Of Christopher J.
Wright
Sent: 02 December 2013 00:36
To: rmcprofile-users at rmcprofile.org <mailto:rmcprofile-users at rmcprofile.org>

Subject: [RMCProfile-users] X-ray PDF use

 

Hi,

I am working with PDF data collected from an x-ray synchrotron source.  The
samples are nanoparticles of various size and compositions.  Is it possible
to model these particles using RMCprofile?

Thank you very much,

Christopher J. Wright

Brown University Chemistry Department

 

-- 
Scanned by iCritical. 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rmcprofile.org/pipermail/rmcprofile-users/attachments/20131229/eec91bd9/attachment.html>