[Wien] "x nn" problem

[Jorissen Kevin]

What do you want us to look at?  Apparently this struct-file is okay for nn?
I guess I'd use just two RMT values (one for H, one for N and O ; I would certainly try to avoid using different RMT for the same element (N) as this may compromise comparison of the atoms), but maybe you have a reason for using your values.
 
Kevin.
 

	-----Oorspronkelijk bericht----- 
	Van: Bing Zhou [mailto:umbingz at cc.UManitoba.CA] 
	Verzonden: za 10/11/2003 11:44 
	Aan: Jorissen Kevin 
	CC: wien at zeus.theochem.tuwien.ac.at 
	Onderwerp: RE: [Wien] "x nn" problem
	
	

	Dear Kevin and Lukasz:
	
	Thank you for your great help! Yes, I use A instead of a.u. when I input
	the lattice constants using w2web. Now I attach my case.struct and
	case.outputnn files with this email, could you please take a look for me
	again?
	
	Thanks!
	
	Best wishes!
	
	
	
	On Sat, 11 Oct 2003, Jorissen Kevin wrote:
	
	> Hello Bing Zhou & Lukasz,
	> the answer is much more simple.
	>
	> The "5" in "atom 5" corresponds to the number of an inequivalent atom class in case.struct.  Apparently, this class contains more than one atom in the unit cell, either because there is more than one entry for atom 5 in case.struct, or your lattice type (e.g. F) generates more positions.
	> Nn verifies that NO atoms in the unit cell overlap, whether they belong to the same equivalency class or not.  In your case, the distance between two different atom positions both belonging to equivalency class 5 was calculated.
	>
	> So there is nothing strange about this.
	>
	> 1 a.u. is a rather small nearest neighbour distance, maybe you did input the lattice constants in A instead of a.u., as Lukasz suggested?
	>
	> Kevin.
	>
	> PS Lukasz, in fact, if you go to case.outputnn, you can find out exactly which atom positions correspond to a particular interatomic distance - nn tends to write endless lists of those.
	>
	>       -----Oorspronkelijk bericht-----
	>       Van: Lukasz Plucinski [mailto:pluto at buphy.bu.edu]
	>       Verzonden: za 10/11/2003 1:35
	>       Aan: wien at zeus.theochem.tuwien.ac.at
	>       CC:
	>       Onderwerp: Re: [Wien] "x nn" problem
	>
	>
	>
	>       > Why did my "init_lapw" give the distance between atom 5 and atom 5?
	>       >
	>       >
	>       >    ERROR !!!!!!!!!!!!!!!
	>       >  RMT(  5)=0.70000 AND RMT(  5)=0.70000
	>       >  SUMS TO 1.40000 GT NNN-DIST= 1.05157
	>
	>       Hello,
	>
	>       Difficult to answer. My guess:
	>
	>       Distance 1.05157 is related to your lattice constant (and maybe you set it
	>       e.g. in eV instead of Ry when typing into w2web).
	>
	>       Maybe the problem is also related to space group and inequivalent
	>       positions, I am not sure.
	>
	>       You should probably attach 'struct' file. There was a discussion few days
	>       ago that the users should provide more information about their problems.
	>       Even if Prof. Blaha and few other experts are able to answer almost every
	>       question with minimum information included other guys (like me) may need
	>       some more. And then Prof. Blaha don't have to answer almost every single
	>       question himself :).
	>
	>       What is the material ? How many atoms you have in the primitive cell ? -
	>       if its like 2 or 4 it would be very easy to find out where the distance
	>       of 1.05157 comes from.
	>
	>       Regards,
	>
	>       Lukasz
	>
	>       _______________________________________________
	>       Wien mailing list
	>       Wien at zeus.theochem.tuwien.ac.at
	>       http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
	>
	>
	>
	>
	
	
	
	Bing Zhou
	Dept. of Geol. Sc.
	The Univ. of Manitoba
	Canada R3T 2N2
	Tel: (204)474-8395
	

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 8230 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20031013/472c495f/attachment.bin