<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
<span style="scroll-behavior:auto !important;margin:0px;font-size:12pt">Dear All,</span>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt"><br>
</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt">As part of EPW calculations of the EPI matrix lambda on Nb, I am obtaining oscillatory wannier interpolation of both the phonon and electron band structures.</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt"><br style="scroll-behavior:auto !important">
</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt" class="elementToProof">
<span style="scroll-behavior:auto !important;margin:0px">I'm sharing two plots of PW vs Wannier phonon and Fermi-crossing electron bands. For the interpolation, an nscf calculation on a K-Mesh of 10^3 was used to generate .nnkp file. The scf calculations had
 the same settings except a K-Mesh of 20^3 was used, and the phonons a Q-Mesh of 10^3. I<span style="scroll-behavior:auto !important;margin:0px;background-color:rgb(255, 255, 255);display:inline !important">t is possible that I need to increase the nscf K-mesh
 for finer interpolation, but since I'm using Wannier 3.1 with the</span><span style="scroll-behavior:auto !important;margin:0px;background-color:rgb(255, 255, 255);display:inline !important"> tag use_ws_distance = .true. by default, I thought it might help.
 Additionally, whenever I try to wannier interpolate through epw.x on multiple nodes, I run into this parallelization error:</span></span></div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt" class="elementToProof">
<span style="scroll-behavior:auto !important;margin:0px"><span style="scroll-behavior:auto !important;margin:0px;background-color:rgb(255, 255, 255);display:inline !important"><br>
</span></span></div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt" class="elementToProof">
<span style="scroll-behavior:auto !important;margin:0px"><span style="scroll-behavior:auto !important;margin:0px;background-color:rgb(255, 255, 255);display:inline !important">
<p class="p1" style="margin:0px;font:11px Menlo;color:rgb(29, 186, 255);background-color:rgb(9, 9, 9)">
<span class="s1" style="font-variant-ligatures:no-common-ligatures">Error in routine epw_readin (1):</span></p>
<p class="p1" style="margin:0px;font:11px Menlo;color:rgb(29, 186, 255);background-color:rgb(9, 9, 9)">
<span class="s1" style="font-variant-ligatures:no-common-ligatures"><span class="Apple-converted-space">    
</span>Number of processes must be equal to product of number of pools and</span></p>
<p class="p1" style="margin:0px;font:11px Menlo;color:rgb(29, 186, 255);background-color:rgb(9, 9, 9)">
<span class="s1" style="font-variant-ligatures:no-common-ligatures"><span class="Apple-converted-space">    
</span>number of images</span></p>
<p class="p1" style="margin:0px;font:11px Menlo;color:rgb(29, 186, 255);background-color:rgb(9, 9, 9)">
<span class="s1" style="font-variant-ligatures:no-common-ligatures"><span class="Apple-converted-space">    
</span>Image parallelization can be used only in calculations on coarse grid.</span></p>
</span></span></div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt"><br style="scroll-behavior:auto !important">
</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt">I would like to interpolate only the two bands the cross the Fermi-level, out of the 11 in the pseudopotential I used. Is that possible? Or would I need to interpolate more bands and freeze
 the Fermi-crossing ones? If so, I am not sure how to specify the projections, and suggestions are greatly appreciated. <span style="scroll-behavior:auto !important;margin:0px">I am currently using the auto-scdm projection scheme and interpolating all 11 bands.</span></div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt"><br style="scroll-behavior:auto !important">
</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt">Below are the versions used for this calculation, as well as the google drive folder for the phonon and electron band plots and input/output files. nb.epw1.in is the input for wannier interpolation
 with epw.x.</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt"><br>
</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt">Versions:</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt">QE: 6.8</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt">EPW: 5.4.1</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt">Wannier90: 3.1</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt"><br style="scroll-behavior:auto !important">
</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt"><a href="https://drive.google.com/drive/folders/1I-rnjZFgS3BDttcFLnOFKd9RUxar86Q3?usp=sharing" id="LPlnk878952">https://drive.google.com/drive/folders/1I-rnjZFgS3BDttcFLnOFKd9RUxar86Q3?usp=sharing</a><br>
</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt"><br>
</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt" class="elementToProof">
Thank you,</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt" class="elementToProof">
Bander Linjawi</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt" class="elementToProof">
PhD Student Mechanical Engineering</div>
<div style="scroll-behavior:auto !important;margin:0px;font-size:12pt" class="elementToProof">
Duke University</div>
</div>
</body>
</html>