[QE-users] units of q points in matdyn.in files to calculate phonon dispersion

[w]

It is my pleasure to write to you.
I am a postgraduate student in Tsinghua university. 
I am calculating phonon dispersion  with quantum espresso. 
And I am using materialscloud to creat K ponits on line.
I would like to ask some questions about units of q points in matdyn.in.freq files.
For example , H point for MgB2 , is (0.3333 0.3333 0.5) with scaled units created on materialscloud or using Materials Studio.
My question is whether H (0.3333 0.3333 0.5)  be used directly in matdyn.in files to calculate phonon dispersion.
In the matdyn.in files , Kpoint's unit is  "2pi/alat" ,so it's wrong to use H (0.3333 0.3333 0.5) directly?


So I convert H (0.3333 0.3333 0.5) to （0.333333  0.57735  0.437548）with 2pi/alat units , and here is my ways,I don't konw whether it's right ？ Or how to convert it ?
First , scf.out contains information about "reciprocal axes: (cart. coord. in units 2 pi/alat)", it's 
b(1) = (  1.000000  0.577350 -0.000000 )  
b(2) = (  0.000000  1.154701  0.000000 )  
b(3) = (  0.000000 -0.000000  0.875096 )
Then , H point with scaled units is ( 0.3333 0.3333 0.5 ) , 
equals to  0.3333*b(1）+0.3333*b(2)+0.5*b(3)  with unit 2pi/alat，
because in scf.out files :
              b(1)= 1*2pi/alat(1)+0.577350*2pi/alat(2)+0*2pi/alat(3),
              b(2)= 0*2pi/alat(1)+1.154701*2pi/alat(2)+0*2pi/alat(3),
              b(3)= 0*2pi/alat(1)+0*2pi/alat(2)+0.875096*2pi/alat(3),
Finally, H points convert to （0.333333   0.57735      0.437548）with unit 2pi/alat,
which equals to row vector H multify matrix B=[ b(1) ; b(2) ; b(3) ]
 = H*B=（0.333333  0.57735  0.437548）with 2pi/alat units.
Is this the right method to convert high symmetry q points ?
Or how to convert the q points with scaled units to those with 2pi/alat?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20191028/b8b4f7ce/attachment.html>