<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:΢ÈíÑźÚ
}
--></style>
</head>
<body class='hmmessage'><div dir='ltr'>
<pre>Dear Prof. Blaha<br><br>Thank you very much for your reply!<br><br>To set the NX,NY,x-len,y-len in case.spaghetti_ene<br><br>The first-line of fort.2 created by hex_fs_mesh is:<br> 1 0 0 0 120 1.00-7.00 1.50 61 121<br>Thus I just set NX=61 and NY=121, is it correct?<br><br>According to the manual, the unit of x-len and y-len is bohr^(-1), thus originally I think it is the reciprocal lattice. But when I set x-len=0.3382<br>, I get very small plot. Thus, I think x-len and y-len is the size of the plot. Actually both <br>x-len=10 and x-len=20 can work, and x-len=20 give twice bigger plot than x-len=10. So I confuse why you said "the length of rec.space must be ok"?<br><br>Thank you again for your help!<br><br>Bests<br><br>Xiangang<br>-----------------------------------------------------------<br>I don't quite understand why you say that 6 bands cross EF along G-M ???<br>According to your bandstructure there are only 4 bands, but two of them cross<br!
>EF twice. But still, this is the "same" band.<br><br>><i> With this setup I finish the calculation, attached please find the six-band which contribute to the Fermi surface.<br></i>><i><br></i>><i> 1) The output-plot is square, could you please tell me the coordinate of the four corner in of the square. I think one is (0,0,0); another is (1/2,0,0). How about other two? Where<br></i>><i> is the x-axis in the plot? I guess the Gamma-point is located at the left-bottom corner of the square, and M-point is the right-bottom corner, right?<br></i><br>This depends a lot on the settings in case.spaghetti_ene is :*<br> > 19 13 0.33820 0.19526 27 23 0 0<br><br>The first two numbers must be set properly according to your mesh.<br>Also the length of rec.space must be ok.<br>The "interpolation" numbers should be 2N-1 and not "arbitrary numbers".<br>and finally, the "0 0" is very important.<br>If you don't understand them, experiment with them and you should be able<br>to!
understand them.<br><br>For better orientation you should do a bandst
ructure plot, but not in the<br>directions you attached (going into z>0 ???), but plot several directions in the<br>Gamma-X-M plane. Then it should be easy to orient you.<br><br>Also, use the fs option of xcrysden to get an 3D overview over the different<br>sheets and then it is easier to "imagine" how the cuts should look like.<br><br>><i> 2) Why the lines in save_filename_64.plot.ps, save_filename_63.plot.ps is dot-line; and the lines in other files is full-line?<br></i><br>plotgenc plots negative contors as dashed, zero as dotted and positive ones as full<br>lines.<br>I don't know how you produced these plots.<br>I also don't quite understand why you have save_filename_63, 64, but then 77,78,79,80 ???<br>This does not look ok.<br><br>><i> 4) It is not convenient to plot the fermi-surface separately. May we combine the sheet from different bands to a single plot?<br></i><br>cat save_filename_63.plot.ps save_filename_64.plot.ps > combined.ps<br>and edit combined!
.ps<br>I attach the edited combined.ps file, where I simply commented (%) a few lines<br>and added a color switch to plot the second sheet in a different color.<br><br>With this tamplate you should be able to make it for other examples.<br><br><br> P.Blaha<br><br></pre> </div></body>
</html>