Hi all,
I am using QE 5.0.3.
I am trying to test if a saddle point from the neb.x calculation is a real
saddle point. I am doing this by looking for the vibrational frequency: If a
state is a saddle point, it has only one imaginary (negative) frequency. If a
state is an energy minimum, it has no imaginary frequency.
My system consists of a crystal structure and a helium impurity in it. The
crystal structures at energy minimum and the saddle point are almost identical,
only the helium atom is displaced. To facilitate the frequency calculation, I
used nat_todo = 1 to only allowing the helium to be displaced:
phonons of initial state at Gamma
&inputph
nat_todo = 1
tr2_ph=1.0d-14,
prefix = "te",
epsil=.true.,
/
0.0 0.0 0.0
25
The last line, 25, indicates only the 25th atom (helium) is displaced during
the ph.x calculation. Because there are 25 atoms total in the system, there
should be 75 modes. The complete ph.x outputs are attached. The energy minimum
state has the frequencies:
omega( 1) = -2.080553 [THz] = -69.399779 [cm-1]
omega( 2) = -2.080553 [THz] = -69.399779 [cm-1]
omega( 3) = -2.080553 [THz] = -69.399779 [cm-1]
???.
???.
omega(73) = 4.621913 [THz] = 154.170425 [cm-1]
omega(74) = 4.621913 [THz] = 154.170425 [cm-1]
omega(75) = 4.621913 [THz] = 154.170425 [cm-1]
with omega(4~72) all equal to zero. For the saddle point one:
omega( 1) = -4.316979 [THz] = -143.998928 [cm-1]
omega( 2) = -2.270462 [THz] = -75.734475 [cm-1]
omega( 3) = -2.270462 [THz] = -75.734475 [cm-1]
???.
???.
omega(73) = 2.168375 [THz] = 72.329209 [cm-1]
omega(74) = 9.304582 [THz] = 310.367434 [cm-1]
omega(75) = 9.304582 [THz] = 310.367434 [cm-1]
My questions are:
1. Why are there 3 additional frequencies? Shouldn?t there only be 3 (omega
73~75) due to the DoF of one atom?
2. Why are the three additional frequencies negative? Shouldn?t there be one
negative frequency for the saddle point, and 2 positive frequencies? and no
negative frequency for the energy minimum?
Please let me know if I understand this incorrectly, or I am missing some
settings.
--
Kan-Ju Lin, Postdoctoral Associate
MIT Materials Science and Engineering
Office 4-053 / kjlin at mit.edu<mailto:kjlin at mit.edu>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://pwscf.org/pipermail/pw_forum/attachments/20140318/bfcf64f1/attachment.html
