NEB Calculation using VASP

Dear Dr. Garvey,

I am trying to use Nudged Elastic Method to get oxygen vacancy pathways in pure Zirconia. Some researcher recommend to use SPRING = -5, while the some forum site recommend to SPRING =0 . Which one is the best? 

I would like to make sure my results. When we use nebmake.pl POSCAR00 POSCAR06 5 command, in these case POSCAR00 and POSCAR06 are optimized structure, isn't it? I allowed to all relaxation when I create optimized initial and final POSCAR files.

In second step; while create 00,01,02.. sub directories, we tell to VASP use NEB theory with,

SPRING = -5 (or 0, I don't know)

IMAGES = 5

LCLIMB = TRUE

Am I right?

After the sub directories are created  in same path, we rearrenge INCAR file. My opinion is use ISIF =7 for not allow to ion relaxation.

I have mention about three type INCAR file,

1. Preparing to Initial and Final position file,

2. Using Nudged elastic band, while we were creating some images

3. In the sub directories (00,01,02...)

What is the INCAR of differences of these three steps?

Thank you,

Berna 

 Dear Berna Akgenc,

I will try my best to answer your question.

You need to have your initial and final structures well relaxed. In this case, it means POSCAR00 and POSCAR06 . Then you use the VTST script nebmake.pl to create the number of images. I understand you want to have 5 images thus you do:

nebmake.pl POSCAR00 POSCAR06 5. 

thus you will have 7 folder (00,01,02,03,04,05,06). For further analysis you need to put the OUTCARs for POSCAR00 and POSCAR06 in the folders 00 and 06.

Then you need to make sure that your vasp executable is complied with VTST (http://theory.cm.utexas.edu/vtsttools/neb.html#neb). unless you can not make a CI-NEB.

Then in your parent directory where you have all the (00,01,02,03,04,05,06) directories, you put  your INCAR, KPOINTS and POTCAR file. But this time, your INCAR file will be a little be different as you need to include the NEB tags on it. And me personally use the spring constant be -5.

A sample of INCAR file is here in the following:

-------------------------------------------------------

SYSTEM = NEB-Calc

ISYM = 0 # switch on symmetry

Start parameter for this Run:

ISTART = 0

ICHARG = 2

Electronic Relaxation 1

ENCUT = 500 eV

IALGO = 38

EDIFF = 1.0E-6

NELMIN = 5

Ionic Relaxation

EDIFFG = -1.0E-3

NSW = 500

DOS related values:

ISMEAR = 0

SIGMA = 0.05

Cell Volume + Cell shape

ISIF = 3

Other Calculation parameters

PREC = ACCURATE # Advised to do so

LREAL = .FALSE.

LWAVE = .FALSE.

LCHARG = .FALSE

NEB Related

IMAGES = 5 # Number of NEB images between the fixed endpoints

SPRING = -5.0 # The spring constant, in eV/Ang^2 between the images; negative value turns on nudging

ICHAIN = 0 # Indicates which method to run. NEB (ICHAIN=0) is the default

LCLIMB = .TRUE. # Flag to turn on the climbing image algorithm

LNEBCELL = .TRUE. # Flag to turn on SS-NEB. Used with ISIF=3 and IOPT=3

# Must set IOPT = 3 or 7 when using LNEBCELL=.TRUE.

IOPT = 3 # QM (Quick-Min) force based optimizers is used and good for high force far from minmum

# O(default vasp),1(LBFGS),2(CG),4(SD),7(FIRE)

IBRION = 3 # Disbale VASP default optimizers which are basesd on Energy

POTIM = 0 # Disbale VASP default optimizers which are basesd on Energy

#MAXMOVE = 0.2 # Maximum allowed step size for translation(Default comes with IOPT=3)

TIMESTEP = 0.1 # Dynamical time step (Default comes with IOPT=3)

LSCALAPACK = .FALSE.

NPAR = 4

-------------------------------------------------------

This INCAR is for calculation allowing a variable CELL with Climbing NEB.

Hope it helps.

Regards,

Abel

Dear Abel 

Thank you so much your quick answer and sharing your INCAR file with us. I have quick question here, why do you use ISIF=3 at sub-directrory. I mean that you have already create new POSCAR files in 00,01,02 sub- directories using NEB. And at the beginning , you have allowed to all relaxation (Initial and Final POSCAR's, via ISIF=3) 

My question is coming here actually, 

I totally agree with you.  In sub-directory's INCAR file should be different. 

"Then in your parent directory where you have all the (00,01,02,03,04,05,06) directories, you put  your INCAR, KPOINTS and POTCAR file. But this time, your INCAR file will be a little be different as you need to include the NEB tags on it."

II THINK, we don't allow ion relaxation as well  (with ISIF=7 or other ion relaxation don't allow) we should use frozen ions in these process. 

I hope that I could express myself. I am just trying to understand INCAR tags with its physical meaning. According to your experience, what do you think about my idea?

Am I right or wrong? 

Dear Berna,

First of all in all the directories you do not need to put any INCAR POSCAR and KPOINTS.

Let's say if your parent directory is named NEB-CALC. Thus 

in NEB-CALC id you do ls (assuming you use linux machine) you will have 00 01 02 03 04 05 06.

Your subdirectories will  only have POSCARS.Thus in Your NEB-CALC folder you will have

00/POSCAR  00/OUTCAR

01/POSCAR

02/POSCAR

03/POSCAR

04/POSCAR

05/POSCAR

06/POSCAR 06/OUTCAR

INCAR , KPOINTS and POTCAR

And the ISIF tag is 3 in my case because i relax ion positions and cell(shap+ volume).

But you want to do only ionic motions (ISIF=2) as is stated in here http://cms.mpi.univie.ac.at/vasp/guide/node112.html.

I hope I answered your question. Also check this http://cms.mpi.univie.ac.at/wiki/index.php/Collective_jumps_of_a_Pt_adatom_on_fcc-Pt_(001):_Nudged_Elastic_Band_Calculation.

Regards,

Abel

Dear Abel,

I am sharing my files at attachment, could you take a look please?

I followed your directories.

Between Initial and Final states are 01,02,..09 sub=directories. They included just POSCAR file. 

How i can run it?

Hello Berna,

It seems for me that the way you set up the  calculation is just fine. But I have a couple of questions.  What exactly are you trying to study? I see there is one O-vacancy in your system.

But what  is strange for me is that along your path,  almost starting from image 1 to image 8, distance between some neighbouring Oxygen is not right.

If you check it gets as low as 0.0748 Ang. which is really difficult for the calculation to run.

And the other thing is, do you really want to change the cell's shape and volume during your calculation ? In an other words do you really want to do a SS-NEB calculation?

Because as far as I know, One does SS-NEB  to study phase changes.

if You have your initial structure with a1,b1,c1  and if your final state a2,b2,c2 lattice vectors and if (a1=a2,b1=b2,c1=c2), i guess you don't need to do a SS-NEB, so the tag ISIF must be changed.  I guess in your INCAR change it to ISIF = 2 or something else according to the purpose of your calculation.

The other thing I checked your script for submitting the calculation, there is a part which is not necessary. ( ./nebmake.pl 00/POSCAR 10/POSCAR 9).

Just use your normal script to run one vasp calculation but increase the number of nodes you use according to the number of images you have.

Once you created your intermediate images in which all of them have the POSCARs inside them, you run a normal vasp in the parent directory not in every director. But in the parent directory you need of course the INCAR, KPOINTS and POTCAR files which you have.

I hope I helped.

Regards,

Abel

Dear Berna Akgenc,

This me Abel, I noticed that the ordering your atoms is not correct.

Before sending a NEB calculation you need to make sure the ordering of the atoms between

your initial and final structures is correct, which I noticed in your case is not.

So I tried my best to correct it for you. Check it and all you need is to out the

INCAR,POTCAR and KPOINTS files in the parent directory and then run the calculation.

You just simply have to use the same script for submitting a normal VASP run, but since

you have many images, u need to increase the nodes.

Good luck and let me the result is.

I will try to help as much as I can.

Regards,

Abel

Hi Abel,

First of all, I really appreciated your help and time. 

I wanna to calculate activation energy path for Yttria Stabilized Structure (%8 mol Yttria) using Nudget Elastic Band Method.

My initial and final states refer to hopping an oxygen from one site to other site.

You are right, It is not necessary to use Solid-State NEB so far. I just wanna show you my result to more easier compare than yours.

The good sign that my program is running right now :) 

But I don't still understand how I can fix the problem of ordering of atoms. Do you mean that use InterpolatePOSCAR script?

Thank you,

Berna

Dear Berna Akgenc ,

It is a good news that your calculation is running. And What i meant by ordering of atoms, it simply mean atom 1 in POSCAR-00 must be the same in all the other POSCAR files. Previously  in the one you attached, that was not the case.

Regards,

Abel

Could everyone know how to turn on or show the "FORCES: max atom, RMS" in OUTCAR when perform NEB in vasp? I have get a trouble when run nebresults.pl because the script can not find "FORCES: max atom, RMS" in OUTCAR to create fe.dat. VASP version which I use is 5.4.4.

Thank you very much!

It has been over 3 years now since I did my last NEB calculation. I am now even in another career path. To be honest, at this moment, I forgot most of the necessary steps that one has to do prior to a NEB calculation. So if someone asked a question and I did not answer back, please understand my situation. Thanks and enjoy what you are doing. My best wishes to all.

vef.pl script written by Henkelman group would do that!@ O My Na

Hi Dear Abel Gebreegziabher Gebresilassie

I was looking at the answers you gave to Berna Akgenc Hanedar

I am really wondering by your detail explanation and it helps me to a high extent to understand what NEB is. I really appreciate your commitment to help for those who are beginners in this field. Currently I am focusing on computational study of solid electrolytes and I need to use NEB to understand the diffusivity of Lithium ion in my electrolyte. The problem is I am starting this technique from zero experience just as you were 5 years a go as I understood from the question you asked.

Now I am writing this to get your help. Would you please share me all your files you used in these directories? I don’t know even what to use and how to arrange. So if possible please attach to my email, yosefni2001@gmail.com.

Hello Berna Akgenc Hanedar , and Tekalign T. Debela would you help me you too please?

Thank you in advance for your cooperation.

Hi Prof,

Thanks for your reply. I think I known the reason why I can not receive "Forces..." in outcar. I have failed in compilation vtstcode. Now I've got it.

Thank you so much.

Hi All,

Have anyone experienced with different results using VTST in different VASP version( 4.6 and 5.4 in my case). Any kind of suggestions will be helpful

I want to calculate the energy barrier of Li-ion above 2d material. So first using neb.pl, I make 6 images and run using vasp. Now I did not understand how to calculate the energy barrier. in my parent directory, I have total of 7 folders (00,01,02,03,04,05,06,). Now except first and the last folder in each folder I have files like OUTCAR, CONTCAR, OSZICAR .... but I did not understand how to calculate energy barrier. In the OSZICAR file, there has E . But I am not sure this E is my barrier or not. Can anyone please help me.

Hi Diptesh Gayen

In thermodynamics, the change in Gibbs free energy, ΔG, is defined as:

ΔG = ΔH−TΔS

If you use T = 0 k

ΔG = ΔH

And as you can see from the attached file,

Ea = ΔH+ +RT

Ea = ΔH+ ( if T = 0 K)

Ea = ΔG ( because ΔG = ΔH )

You can find the value of ‘G’ for each folder from the OUTCAR file.

Hopefully, it helps.

Yosef

Hello,

I am a complete beginner in this field, and I was wondering how to download the VTST files and integrate them into VASP calculations. The VTST says to modify the chain.F file but I am not quite sure what to do. Thank you in advance, and any help is greatly appreciated!

Neha Basu This is what you need to do. After downloading the zip folder, transfer the folder to your main directory where you run your VASP simulation/jobs and unzip the folder. Open the unzipped folder, select all the files, right-click, select "Properties", and tick on all the "Execute boxes", click Confirm/OK. There and then, you're good to go.

Cheers!

Shedrack G. Akpe Thank you so much!

Dear all,

I am able to process the NEB calculation. However, the calculation didn't work well which is only run 4 out of 8 images I've input. Could anyone help me with this problem as I have no idea where is my mistakes? Thank you very much.

Cheers

This was a question I raised, during my PhD time. I am happy to still see people are refereting to answers written in here. I have to be honest, it has been a while since I did such kind of simulaitons.

Hi Lim Je Sie

I was looking at your INCAR file, I couldn't see what the problem is. Maybe you can try the attached INCAR file. It was the one I used in my case for c-NEB. hope it will be helpful

Dear Lim Je Sie

The problem is from your POSCAR file, I tried to see your poscar files and 00, 05, 06, 07, 08 and 09 have problems. you need to fix this problem first. Either it is from building the initial and final states or there is something you made a mistake during generating the images. I recommend you to recheck all the poscar files first before processing. Thanks

@all I want to ask is ISPRING and SPRING tag same in while calculating transition state using NEB in VASP

I would like to complement the answer by telling that a regular NEB calculation is usually run before the flag LCLIMB is turned on.

I also recommend to watch the video below in order to understand a bit more about the theory and details of the calculations. The talk was given by Dr. Manaschai in a workshop. It is very good!

https://www.youtube.com/watch?v=2PotR4_sV_0

Thanks Abel Gebreegziabher Gebresilassie , your answer helped me

Lim Je Sie Hi, I think the IBRION tag value in your INCAR is wrong, it should be set to 1 or 3 for better convergence. And I advise you use less images for starting NEB calculation for faster convergence, then increasing the number. Some details see below: https://cms.mpi.univie.ac.at/vasp/vasp/Elastic_band_method.html

Hello Everyone,

I am new to VTST, I want to use the nebmake.pl to generate my intermediates POSCAR. I would appreciate it if someone please walk me through the process. I have download the nebmake.pl code, I do not know what flag should I have in my INCAR to read the nebmake code, and also how many directories should I made. Should I make a separate folder for the initial and final POSCAR or all are in the same folder? Thank you

Fateme Molajafari

You need to relax the initial and final positions and take the CONTCARS from those runs and put them in the same folder. For example, CONTCAR_00 (initial) and CONTCAR_06 (final) and you need to create 5 images.

Type "nebmake.pl CONTCAR_00 CONTCAR_06 5"

This will create 5 images between the initial and final positions.

Finally,

Have a look at this video: https://www.youtube.com/watch?v=2PotR4_sV_0

Ravindra Kempaiah

Thank you for your answer. I ran the job but this is the error I got:

ERROR: the following files does not exist POSCAR

In my directory I had:

INCAR, KPOINTS, POTAR, POSCAR_00, POSCAR_06, and nebmake.py

This is my INCAR:

SYSTEM = HZSM5

PREC = Accurate

ENCUT = 500

ALGO = All

ISMEAR = 0

SIGMA = 0.01

# parallelization

LPLANE = .TRUE. # distribute real-space mesh

#NPAR = 8 # number of cores per computing node

LORBIT = 11

#LREAL=.FALSE.

#NELECT = 50 #7*nions+1

ISPIN = 2

MAGMOM = 8*0

# relax

ISIF = 2 # 2=atoms, 3=2+lattice, 4=shape+ions

IBRION = 2 # 1 = Quasi-Newton, 2 = CG, 5 = vib modes

NFREE = 20

NELMIN = 4 # minimum number of electronic steps

NELM = 150

EDIFF = 1E-6 # SCF energy convergence 1E-7

EDIFFG = -0.03 # max force for ionic steps -0.01

NSW = 250 # max ionic steps

POTIM = 0.05

NUPDOWN = 0

IVDW = 11

nebmake.pl POSCAR_00 POSCAR_06 5

# outputs

NWRITE = 1 # {0|1|2} default:1

LCHARG = .FALSE.

LWAVE = .FALSE.

LAECHG = .FALSE.

Is this a correct INCAR? I simply wrote:

nebmake.pl POSCAR_00 POSCAR_06 5

in the INCAR.

Thank you.

Fateme Molajafari

Hi!

This is the mistake. The script nebmake.pl should not be put inside the INCAR, since it is used to create the images.

So, first you create the images. In your case 5. The script will generate folders 00 ...06. Start and end points should be previously optimized.

In folder 00 you will add CONTCAR from starting point (rewritten as POSCAR) along with the OUTCAR from this structure. In folder 06, you put the CONTCAR (written as POSCAR) and OUTCAR from the final points. Then, the images will be automatically placed in the folders in between (01 to 05).

Note that your INCAR should have the tags to indicate the method of optimization (different in NEB), the number of images and so on. IBRION should also be set in accordance with that.

I hope I have helped a bit.

Dear Giane B. Damas

Thank you for your answer.

At this time I do not want to do NEB calculations, I just want the nebmake.pl to create 5 images for me. Please tell me:

what flag should I use in my INCAR to read the nebmake.pl script in my folder?

Also, do I have to make that 5 images myself?

I have the optimized initial and final points.

Now, what files should I put in my main direcotry?

INCAR, POTCAR, KPOINTS, POSCAR_00, POSCAR_06, and nebmake.pl?

Anything else? any subdirectories in my main directory?

Thank you very much.

Dear Fateme Molajafari,

First, as you just want to create the 5 images, Make any Work-Test directory and place your optimized POSCAR files of initial and final points (POSCAR-IS and POSCAR-FS).

Next, on command line in that directory, use to create 5 images,

This will create 7 subdirectories with name 00(IS), 01,02,....06(FS), each contains POSCAR created using nebmake.pl script.

Hope this helps.

Fateme Molajafari

As Swapnil Deshpande mentioned, there is no need for INCAR or other VASP files to use nebmake.pl. This script is just to create the images by linear interpolation between the initial and end points. I also recommend that you check each image after you create them, because some chemical bonds can be wrong, specially if there is rotation between one image and another.

May I ask you what kind of calculations you want to perform? Because nebmake.pl is for NEB.

Dear Swapnil Deshpande

Thank you for the explanation, it helped a lot, and actually, I just got my intermediates. I appreciate it.

Dear Giane B. Damas

Thank you very much. It helped a lot. I want to study Methane dissociation. Basically, breaking one C-H bond of the methane.

Dear Giane B. Damas

I am having difficulties running NEB calculations, using nebmake.pl script I made 5 intermediates, and totally I had 7 directories , I provided the OUTCAR of the initial and final optimized POSCARs in their directories. Finally I pvided the INCAR, POTCAR, KPOINTS, and VTST scripts in the root directory, and ran the calculations for 250 ioinc steps. After comleteting 250 steps, I see that the forces on intermediate images did not decrease at all, and in the OSZICAR the energy did not change from the second ionic step to the last one. I am studying a very simple system which is the umbrella inversion of the NH3. Looks like NEB is doing nothing. I would appreciate it if you please help me with this. Thank you. Here is my INCAR:

PREC = Accurate

ENCUT = 600

ALGO = All

ISMEAR = 0

SIGMA = 0.01

# parallelization

LPLANE = .TRUE. # distribute real-space mesh

#NPAR = 8 # number of cores per computing node

LORBIT = 11

LREAL=.FALSE.

IALGO = 38

#NELECT = 50 #7*nions+1

ISPIN = 2

MAGMOM = 4*0

# relax

ISIF = 3 # 2=atoms, 3=2+lattice, 4=shape+ions

NFREE = 20

NELMIN = 4 # minimum number of electronic steps

NELM = 150

EDIFF = 1E-6 # SCF energy convergence 1E-7

EDIFFG = -0.03 # max force for ionic steps -0.01

NSW = 250 # max ionic steps

NUPDOWN = 0

# outputs

NWRITE = 1 # {0|1|2} default:1

LCHARG = .FALSE.

LWAVE = .FALSE.

LAECHG = .FALSE.

ISTART = 0

ICHARG = 2

IMAGES = 5 # Number of NEB images between the fixed endpoints

SPRING = -5.0 # The spring constant, in eV/Ang^2 between the images; negative value turns on nudging

ICHAIN = 0 # Indicates which method to run. NEB (ICHAIN=0) is the default

LCLIMB = .TRUE. # Flag to turn on the climbing image algorithm

IBRION = 3 # Disbale VASP default optimizers which are basesd on Energy

LNEBCELL = .TRUE. # Flag to turn on SS-NEB. Used with ISIF=3 and IOPT=3

POTIM = 0 # Disbale VASP default optimizers which are basesd on Energy

IOPT = 3 # QM (Quick-Min) force based optimizers is used and good for high force far from minmum

#MAXMOVE = 0.2 # Maximum allowed step size for translation(Default comes with IOPT=3)

TIMESTEP = 0.1 # Dynamical time step (Default comes with IOPT=3)

LSCALAPACK = .FALSE.

Hi Fateme Molajafari

I suggest that you remove ISIF=3, because the volume of the cell should not change in this process. Could you put here the results that you obtain by using $nebefs.pl ? I think this optimizer might be too slow for your convergence criteria.

Hi Giane B. Damas

Thank you for your response, I removed ISIF =3 and it helped, the forces are decreasing now. However, my goal is to break one of the C-H bonds in methane using another molecule. In my intermediates, I am consistently increasing the distance between H and C of the methane. However, when I run NEB, H goes back to the CH3! It should get further rather it get closes. Should not NEB freeze the intermediates? I am lost. Can you please comment on this? Thank you.

Hi Fateme Molajafari

Could you please copy the input here?

During the NEB calculations all atoms that are allowed to move (as given in the Selective Dynamics section) will be optimized following the NEB forces (where the force acting on a certain image is a vector combining the component along the band (or parallel) and another that pushes it to the saddle point region (perpendicular) until the convergence criteria has been reached for the atomic forces ( EDIFFG= -0.01, for instance). Thus it is important that the band is built to have smooth changes from one image to another. I have found myself that big rotations from one image to another are quite troublesome, so I recommend to do it in small steps. Do you have rotations in your path?

See more on paper: "Optimization methods for finding minimum energy paths" by Daniel Sheppard et al. J. Chem. Phys. 128, 134106 (2008).

Giane B. Damas

Thank you. This is my INCAR:

SYSTEM = HZSM5

PREC = Accurate

ENCUT = 600

ISMEAR = 0

SIGMA = 0.01

# parallelization

LPLANE = .TRUE. # distribute real-space mesh

#NPAR = 8 # number of cores per computing node

LORBIT = 11

LREAL=.FALSE.

#NELECT = 50 #7*nions+1

ISPIN = 2

MAGMOM = 8*0

# relax

ISIF = 2 # 2=atoms, 3=2+lattice, 4=shape+ions

IBRION = 1

NFREE = 2

NELMIN = 4 # minimum number of electronic steps

NELM = 150

EDIFF = 1E-6 # SCF energy convergence 1E-7

EDIFFG = -0.03 # max force for ionic steps -0.01

NSW = 800 # max ionic steps

NUPDOWN = 0

POTIM = 0.5

IVDW = 11

# outputs

NWRITE = 1 # {0|1|2} default:1

LCHARG = .FALSE.

LWAVE = .FALSE.

LAECHG = .FALSE.

IMAGES = 5 # Number of NEB images between the fixed endpoints

SPRING = -5.0 # The spring constant, in eV/Ang^2 between the images; negative value turns on nudging

ICHAIN = 0 # Indicates which method to run. NEB (ICHAIN=0) is the default

I want to break C-H of methane using a carbide.

I have a small rotation.

In my POSCAR, I have a CH3 that approcing the metal, and H of the CH3 that is approaching a carbon.

Hello Everyone,

I am new to NEB calculations, and I have some questions.

In my system, the goal is to break one of the C-H bond of methane using a carbide molecule. In the initial endpoint, methane is approaching the carbide, and in the product endpoint, one of the H of methane dropped off and forms a bond with the carbon. I have used nebmake.pl script o generate 5 intermediates. I ran the NEB calculations, and the problem is that in 2 of the intermediates, the forces are around 1 eV/angstrom and do not decrease even after 3000 ionic steps. Does NEB ever converge the intermediates? Should I preoptimize the intermediate and then start NEB? This is my INCAR:

ISMEAR = 0

SIGMA = 0.01

# parallelization

LPLANE = .TRUE. # distribute real-space mesh

#NPAR = 8 # number of cores per computing node

LORBIT = 11

LREAL=.FALSE.

#NELECT = 50 #7*nions+1

ISPIN = 2

MAGMOM = 8*0

# relax

ISIF = 2 # 2=atoms, 3=2+lattice, 4=shape+ions

IBRION = 1

NFREE = 2

NELMIN = 4 # minimum number of electronic steps

NELM = 150

EDIFF = 1E-6 # SCF energy convergence 1E-7

EDIFFG = -0.03 # max force for ionic steps -0.01

NSW = 800 # max ionic steps

NUPDOWN = 0

POTIM = 0.15

IVDW = 11

# outputs

NWRITE = 1 # {0|1|2} default:1

LCHARG = .FALSE.

LWAVE = .FALSE.

LAECHG = .FALSE.

IMAGES = 5 # Number of NEB images between the fixed endpoints

SPRING = -5.0 # The spring constant, in eV/Ang^2 between the images; negative value turns on nudging

ICHAIN = 0 # Indicates which method to run. NEB (ICHAIN=0) is the default

Thank you in advance.

Is there any reason you using MAGMOM = 0?

Hi Indrajit Nandi

I use MAGOM =0 because I do not want the system to be spin polarized.

Thanks.

Fateme Molajafari Hi!

I sent you a message on Linkedin.

Hello everyone,

I want to make a NEB calculation (with IMAGES=2), but I have a problem with the POSCAR file. Which POSCAR file should I put to the repertory 01 and 02 ?

Thank you

Hi Abel Gebreegziabher Gebresilassie

I am trying to understand the INCAR used for CI-NEB calculation (as above in your answer comment). Can you please help me clarify that the INCAR will actually perform a self-consistent calculation or ISIF = 3 (Cell Volume + Cell shape) is performed? Then IBRION = 3 is taken which is ionic relaxation (damped molecular dynamics). Could you please help me understand the INCAR for CI-NEB calculation.

Hi Radha N. Somaiya

The value of ISIF will depend on what you are trying to study. If not necessary, use ISIF = 2. You can go through following link, they have explained each tag:

Hi Radha N. Somaiya , the default INCAR (not the Henkelman's group modified one)I use for NEB is:

SYSTEM = catalyst_adsorbate

Startparameter for this Run:

NWRITE = 2; LPETIM=F write-flag & timer

ISTART = 0 job : 0-new 1-cont 2-samecut

ISPIN = 2

# NPAR = 1

POTIM = 0.01

Electronic Relaxation 1

ENCUT = 400.00 eV

NELM = 60;

NELMIN = 4

EDIFFG = -5E-02 stopping-criterion for IoR

EDIFF = 1E-06 stopping-criterion for ELM

# RWIGS = 1.73 1.65 1.11

# prec = high

# MAGMOM = 72*1 5*1 5*1 12*1

Ionic Relaxation

NSW = 400 number of steps for IOM

IBRION = 3 ionic relax: 0-MD 1-quasi-New 2-CG

#NFREE = 2

DOS related values:

ISMEAR = 1; SIGMA = 0.05

# ISMEAR = -5

Electronic Relaxation 2

IALGO = 48 algorithm

LDIAG = T sub-space diagonalisation

# ALGO = F

LREAL = A

LWAVE = F

# LELF = T

# LPARD = T

# IBAND = 97

# NGX = 100; NGY = 100; NGZ = 100

# NGXF = 100; NGYF = 100; NGZF = 100

LCHARG = F

NEB calculation

LCLIMB = T

IMAGES = 6

SPRING = -5

This will suffice for any NEB calculations. After the runs, you can then go ahead and start analyzing your results using the various codes in the vtscripts by Henkelman's group.

It has been more than 5/6 years since I am not active in this topi. However, I am happy it is live and people are exchanging their knowledge and experiences. Enjoy VASPing.

Hi Shedrack G. Akpe

I am done with NEB calculations, but while i am running nebresults.pl

it is giving me very very high values as you can see in the last column.

Forces and Energy:

0 0.000000 -362.926700 0.000000

1 0.000000 -359.480900 3.445800

2 0.000000 -322.705500 40.221200

3 0.000000 -210.978400 151.948300

4 0.000000 -302.013200 60.913500

5 0.000000 -357.249700 5.677000

6 0.000000 -365.912100 -2.985400

Extrema 1 found at image 0.000000 with energy: 0.000000

Extrema 2 found at image 1.000000 with energy: 3.445858

Extrema 3 found at image 2.000000 with energy: 40.221269

Extrema 4 found at image 3.000000 with energy: 151.948283

Extrema 5 found at image 4.000000 with energy: 60.913577

Extrema 6 found at image 5.000000 with energy: 5.677067

Extrema 7 found at image 6.000000 with energy: -2.985387

Can you help me with this?

I would appreciate if anyone who is aware of this NEB calculation comes forward and helps me with this.

Thank you in advance.

Depending on what your reaction maybe, from your results above, it is certain that your starting images in between 00 and 06 (since I can see you created 5 images) are not perfectly aligned. There should at least be "partly" smooth flow from 00 via all images to the product at 06.

Make sure after creating the images, you also use the vtscripts (via this command: "nebavoid.pl 1") to space the atoms of the reacting components to at least 1 Angstrom.

After that, create a movie (with the command: "nebmovie.pl") and check the flow from 00 to 06 on your display software (such as jmol) before submitting the job. I hope this will help Radha N. Somaiya

Abel Gebreegziabher Gebresilassie Thank you for doing this outstanding post. I have some basic questions for you, do we need to do some modifcations in the script nebmake.pl? I am trying to run this script, but nothing happened. When I use the prompt chmod +x nebmake.pl POSCAR00 POSCAR06 5" The terminal said "chmod: cannot access ‘5’: No such file or directory", when i make a directory call "5" i am not getting any results

First do this only: chmod 755 nebmake.pl

then use: nebmake.pl POSCAR1 POSCAR2 5 (initial structure final structure no. of images)

Rita Maji Thank you, dear colleague, for your guidelines, btw, once the images are generated and after running vasp properly with the INCAR, KPOINT, and POTCAR file from the main directory, is it normal to face the following bugs?

BUG

internal error in: us.F at line: 1332 |

| |

| internal ERROR SETYLM_AUG: 36 -78 404 36 35 41 -79 |

| |

| If you are not a developer, you should not encounter this problem. |

| Please submit a bug report. |

|

Dr. Sambath Baskaran can you help me in calculating CINEB, i am aving an error. My OUTCARS are not generating in subfolders.