<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7655.10">
<TITLE>question about LDOS calculation within boxes using pp.x</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->

<P><FONT SIZE=2>Dear QE users,<BR>
<BR>
I have been trying to use pp.x on a Cr(001) slab, and I would like to calculate PDOS in a box above de surface.<BR>
As a matter of fact, I would like to see the decrease of the wavefunctions in the vacuum, by analysing the PDOS in boxes at different heights.<BR>
After a calculation on small boxes, I would like to have, now, an average type box, on the whole surface, so let's say a pavement-like box.<BR>
<BR>
After a pp.x regular PDOS calculation, I have noticed that the segmentation along x and y direction is 30, I thought that regarding this information, I could take a box with a width of 30 on x and y direction for another pp.x calculation with the boxes to get the big box I need.<BR>
<BR>
Hence, pp.x generates a .xsf file for visualizing the boxes. When I would calculate it with a small box (width=2 on x and y directions, ie irmin(1)=1, irmax(1)=3 in the pp.x input file), I could see the box with xcrysden. However, using the big values (30) for my boxes boundaries, I cannot visualize them. But I still get a result about ldos in the boxes.<BR>
<BR>
My questions are the following : how can I be sure that the ldos calculation are actually done within my big boxes boundaries? Is it correct to use that big segmentation in the first place?<BR>
<BR>
I've included the pp.x box input.<BR>
<BR>
Thank you,<BR>
<BR>
<BR>
Parwana Habibi<BR>
PhD student<BR>
CEA-Saclay<BR>
DSM/IRAMIS/SPCSI/LNOSC<BR>
France<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
</FONT>
</P>

</BODY>
</HTML>