VTST 2.x code updates

[graeme]

Ah, this update may have been premature. We'll release version 2.04 with the solid-state NEB soon.

[graeme]

V2.04 (14/07/11) released.
- generalized solid-state NEB method for solid-solid phase transitions

[graeme]

V2.04a (08/24/11) released.
- small changes to the generalized solid-state NEB method

[dano]

It should be able to handle this case:
You should set:
ICHAIN=0
IMAGES =?
LNEBCELL=.TRUE.
IOPT = 3
http://theory.cm.utexas.edu/vtsttools/neb/

[graeme]

V2.04b (10/26/11) released.
- more small changes to the generalized solid-state NEB method
- vasp5 CENTCAR file format

[roing]

[quote="dano"]It should be able to handle this case:
You should set:
ICHAIN=0
IMAGES =?
LNEBCELL=.TRUE.
IOPT = 3
http://theory.cm.utexas.edu/vtsttools/neb/[/quote]

Dear Sir,
I want to do a SS-NEB calculation, and I set
ICHAIN=0
IMAGES = 4
LNEBCELL = .TRUE.
IOPT =3
ISIF =3 ,

should I set a pressure using PSTRESS = ? in my INCAR file ? is it right for the SS-NEB ?

[graeme]

Yes, you can set PSTRESS in the INCAR file to find a barrier for a solid-solid transition under pressure with the solid-state NEB.

[xjtumse]

[quote="graeme"]Yes, you can set PSTRESS in the INCAR file to find a barrier for a solid-solid transition under pressure with the solid-state NEB.[/quote]
Dear Prof. graeme,
I tried to set PSTRESS in the INCAR file to find a barrier for a solid-solid transition under pressure with ssNEB. The phase transition path was hcp phase to omeg phase of zirconium. But the results seems not like what you have said. Due to the convergence criterion is FORCES, so when I finished the NEB calculation, the external stress was -100, just offsetting the value that I applied to the system. How should I handle this problem, can you give me some advice?
I have tried images for 1,3, 12 etc,only IMAGES=1 finished, and I checked the volume of system and found it expanded quite a lot and was the same as zero pressure.
Best regards

The following part is the INCAR file that I used.
general:
SYSTEM = Zirconium
ISYM = 0
ISTART = 0 ; ICHARG=2
PREC = Accurate
ENCUT = 400
PSTRESS = 100
NSW = 150
IBRION =3
ISIF = 3
POTIM=0

ISMEAR =2 ; SIGMA = 0.05
EDIFF = 0.1E-06
NELMIN = 6
ICHAIN = 0
SPRING =-5
LCLIMB =.FALSE.
IMAGES = 12
LNEBCELL = .TRUE.
IOPT = 3
EDIFFG = -0.02
EFIRST = -50.91147858
ELAST = -50.92364994

[dano]

PSTRESS is added to the stress tensor after VASP calls the solid-state NEB. Thus it is not accounted for in the projections or the optimizers which work with the NEB. I am working on including this and will post again once it is solved.

[dano]

The SS-NEB has been updated to include external pressure set by the PSTRESS tag. A word of caution here is because of the way the VASP's optimizers are bypassed (IBRION=3, POTIM=0.0) VASP determines convergence based only on force components so it is important to check convergence of the stress tensor printed after the line "stress matrix after NEB project (eV)". This issue will be addressed in a later version.

[butterflyyh]

[quote="graeme"]Can you make sure that the new objects:

dimer.o dynmat.o neb.o lanczos.o sd.o cg.o qm.o lbfgs.o bfgs.o fire.o opt.o

are all listed before chain.o in the makefile. My guess is that they are after, so that chain.o is being built before opt.o, and it is complaining about not finding the opt module.

These instructions are also at:
http://theory.cm.utexas.edu/vtsttools/downloads/[/quote]
Hi, I came to a problem when compiling VTSTcode, and I copied the object files in orders as the instuctions. The makefile I used is in the attachment.
#mpif90 -FR -lowercase -assume byterecl -O2 -ip -ftz -c chain.f90
#chain.f90(90): error #7002: Error in opening the compiled module file. Check INCLUDE paths. [INSTANTON]
# USE instanton
#--------^
#compilation aborted for chain.f90 (code 1)
#make: *** [chain.o] Error 1
what's the possible reason for the error?
Thanks in advance!

[graeme]

There is a new instanton module which should be listed in your list of objects. See the instructions at:

http://theory.cm.utexas.edu/vtsttools/downloads/

to include instanton.o in your makefile

[butterflyyh]

[quote="graeme"]There is a new instanton module which should be listed in your list of objects. See the instructions at:

http://theory.cm.utexas.edu/vtsttools/downloads/

to include instanton.o in your makefile[/quote]
Sorry, there is the instanton.o in the list of objects of my makefile. So, Is there other reasons to result in the error?

[graeme]

Ah right, I see what you mean. The object list that you posted in your message is not the same as in your Makefile.

To debug this, see if instanton.F is present and if instanton.o (and instanton.mod) were build before chain.F.

Also, try making clean, or remove *.f, *.f90, *.o and do a clean build.

[lcyin]

I have a trouble to incorporate TSTS 2.04b into newest VASP code as following,

dynconstr.f90(164): error #6404: This name does not have a type, and must have an explicit type. [DIR_APP]
OPEN(UNIT=g_io%STRUCTINPUT,FILE=DIR_APP(1:DIR_LEN)//'HILLSPOT',STATUS='UNKNOWN')
------------------------------------------------^
dynconstr.f90(164): error #6514: A substring must be of type CHARACTER. [DIR_APP]
OPEN(UNIT=g_io%STRUCTINPUT,FILE=DIR_APP(1:DIR_LEN)//'HILLSPOT',STATUS='UNKNOWN')
------------------------------------------------^
dynconstr.f90(164): error #6404: This name does not have a type, and must have an explicit type. [DIR_LEN]
OPEN(UNIT=g_io%STRUCTINPUT,FILE=DIR_APP(1:DIR_LEN)//'HILLSPOT',STATUS='UNKNOWN')
----------------------------------------------------------^
dynconstr.f90(1946): error #6514: A substring must be of type CHARACTER. [DIR_APP]
OPEN(UNIT=g_io%STRUCTINPUT,FILE=DIR_APP(1:DIR_LEN)//'HILLSPOT',STATUS='UNKNOWN',POSITION='APPEND')
------------------------------------------^
compilation aborted for dynconstr.f90 (code 1)
make: *** [dynconstr.o] Error 1

Can the developers say something about this or show me how to fix it?