Two Element Simulations in LAMMPS

[@rizalsinaga5803]

Hello, why do you choose SC for both Al and Cu atoms, instead of FCC as in library.meam?

[@mustafaaeea2722]

your lectures are very impressive ,great job

[@catcherinthesky788]

Dear EnCodeVentor, Thank you for this tutorial. In case of more complex assemblies like ice and substrate (eg: metals like Aluminum and Titanium), how would you propose bringing the individual structures together to form an adhesive interface? I read that structures can be brought in contact by an "energy minimization" step but I do not grasp what it is. Could you kindly explain? Thanks.

[@jeongwonkim5706]

Thanks for your tutorial. what if an alloy system has 7 - 8 elements (such as Al, Mg, etc) then where do you get the inter-atomic potentials coefficient for different pairs in such system?

[@Jere5120]

Thanks for watching! That's a tough one. As far as I know, no one has made a seven element potential in meam. If you use a simpler pair potential, you can define the interactions between each atom type separately. However, that is just a first order approximation.

[@jeongwonkim5706]

Thank you for your reply. Do you know what are the steps to create a new potential file that contains our desired elements?

[@peterodedeyi3366]

@@jeongwonkim5706 Hi, did you find answer to your question?

[@thescienceboard1581]

I would be very interested to see a data file being made outside of lammps and imported into the program for a simulation. If you have already made a video with that process please let me know.

[@saab182]

How do you set up an orthorhombic simulation box for crystals that are not orthorhombic such as hexagonal lattice? I am specifically interested in md simulation in quartz and performing voronoi tesselation using voro++, which is designed to compute in an orthonormal box. Quantumwise's atomistik tool kit page can transform the unit cell to orthonormal before molecular dynamics. are there free packages and method that can help us set up rectangular super cell for non orthorhombic lattice? I have tried vesta and atomistick, but dont know if that is possible in those programs. Thanks for these videos. I am just starting to learn MD and these videos are very helpful.

[@noor-ul-ainmurtaza7117]

Hi, I need some help to build Au-S interaction. I have separate data files for Au and thiolated polyethylene glycol (PEG) which contain S,C,H,O. How to merge these two data files? lennard jones (LJ) potential is required and how to get this file? Kindly help me...

[@peterodedeyi3366]

Hi , what a brilliant tutorial. Thanks for sharing.

[@LucasValdiviaAraya]

Where can I find and example of more of these two ellements systems?

[@wondwosenkenea]

how is it possible to couple potential functions from DFT to the MD LAMMPS simulation

[@santoshshiva1356]

In my problem I have two blocks one sliding over other how to find out friction force between the surfaces

[@largemax3672]

Link to the code is not working.

[@anjalishankar]

This is the calculation for what is available. is there any way that i can substitute my own potential parameters in this library and then use it?

[@xXicecream9995Xx]

Your links to text files do not seem to work

[@omkmyp4473]

I am trying to download the file but not able to do so. Kindly need help.

[@sharadb9958]

Hello, useful video as well as your discussion in comment section. Thank you. I would like to know about the supercell used in LAMMPS if one wish to obtain density or lattice parameters. Does energy minimization or fix npt of "single unit cell of crytal" sufficient to validate the lttice parameter or simulations with "large super cell" is must? Also, is single unit cell suffiecient, if we want to calculate lattice parameters at non zero Temp?

[@hamidahbrillianamaisun3987]

Sir, why you set all the lattice with sc?

[@ArunKumar-jo8is]

Thanks for sharing the video. Is there a way to define a specific composition of an alloy?

[@Jere5120]

Thanks for watching! LAMMPS does have a built in way of defining a lattice structure for an alloy crystal phase. This works if your alloy creates a structure together (as opposed, for example, to one element being in solution). You can define your own lattice (lammps.sandia.gov/doc/lattice.html - use the "custom" keyword) and then when you create atoms you can decide which basis atom in your lattice is which type (lammps.sandia.gov/doc/lattice.html - see the "basis" keyword). If one or more of your elements is in solution, then you will probably have to use an external program to decide how and where to put in your solute atoms (interstitials, substitutional, etc.).

[@mustafaaeea2722]

I am trying to simulated Fe-Y2O3 model with an edge dislocation in Fe matrix, How can I create dislocation? I did by deleting one layer in Fe but it seems dislocation is not created,what can be the reason?

[@Jere5120]

Thanks for watching! In general, it sounds like that should work. Basically, as long as you delete a half-plane of atoms, you should create some kind of dislocation. Of course, it is important to delete the right plane, and make sure you start your deleted plane in the right place. What did your system do instead of creating a dislocation?

[@mustafaaeea2722]

@@Jere5120now I created dislocation in oriented x[111],y[-1-12],z[1-10] Fe along but in ovito, it is showing ?is there something relatd to EAM or other reason ?

[@neerajkumarbhoi660]

Respected Sir, Can you help me with the potential file for Y2O3 material?

[@akashtalapatra6072]

How can I identify which is for Al atom and which is for Cu atom from the simulation box? Is there any way to know that? Thank you.

[@Jere5120]

Yes, there is a way. It will depend somewhat on the data file, but in this case, you can look at the "type" property and it will have different numbers corresponding to each atom type. Now, that doesn't necessarily tell you which is Al and which is Cu unless you have the input file and determine it.

[@LucasValdiviaAraya]

@@Jere5120 Hi! Can you upload the files again please? 🙏All links are dead

[@devshah1751]

Can Fe3O4 modelling be done using lammps? If so, please help!

[@keks7712]

Hi, it's very nice video. Could you make a video review of the application of the Monte Carlo method in Lammps?

[@Jere5120]

Thanks for watching! This is a topic I would really like to get to, actually. I have not explored it too much yet in LAMMPS, but I have been itching to. Hopefully I will get to it and can make a video on it soon!

[@neerajkumarbhoi660]

Thanks for the brilliant video for two-body simulation. Sir, can you help how to make a fraction of the second element in the first element also with different lattice structure and lattice parameter? For example Aluminum with 10wt. of Silicon carbide.

[@letslearnsomething324]

Sir can you provide a video creating a metal-metal interface eg. CuZr?

[@kidding686]

If I define 2 regions within my simulation box, will the atoms be able to diffuse into the other regions after the simulation starts or will it act like a solid boundary?

[@Jere5120]

After you start the simulation, the atoms will be able to move however the potential and such dictates - by default regions are only useful pre-simulation. There are some fixes you can use if you WANT to use them to affect the simulation, but in general, they do not.

[@yasingoksal6906]

Thanks for educational sharings. I am studying on investigation of mechanical properties ofgraphene reinforced nanocomposites. I am wonder if it possible to set up single layer graphene sheet and aluminium with vdW bond (as an interface interaction) in LAMMPS? If it is not with LAMMPS, how can i achieve this?

[@Jere5120]

Thanks for watching! That is actually a great question. The simple answer is that LAMMPS definitely has the ability to simulate the simulation you are describing. However, it may not be terribly easy. There are two things that you will need to look into for your problem: (1) Does LAMMPS have a built in potential that will simulate the interactions you want? I am pretty sure it does, but you may have trouble finding the constants for your material system. You will have to do a literature search on that topic. But, you can find all of the potentials that LAMMPS has built in right here: (lammps.sandia.gov/doc/pairs.html) or for bonds (lammps.sandia.gov/doc/bonds.html). (2) Actually creating the geometry of a graphene sheet may not be possible within LAMMPS. (I don't think it is, without some complicated manipulation) Probably your best bet is to create the geometry outside of LAMMPS, and then load it in. If you know any scientific programming languages (MatLAB, Python, etc.) this should be straight forward, since the graphene geometry is actually quite straightforward. You could also use a tool that I am looking forward to creating some videos for, namely Atomic Simulation Environment (wiki.fysik.dtu.dk/ase/). This is a python module that can ease the creation of certain molecules and lattice geometries. Good luck with your research!

[@BhargavAkkinepally]

Hello, your videos are really educational and helpful. I have a question on potentials. I am currently working on diffusivity problem for Lithium ion batteries. How can I write the Pair_coeffs for each and every atom for OPLS force field?

[@Jere5120]

Hello! Thanks so much for watching. I don't have any experience with an OPLS type force-field, but you should take a look at this page: sites.google.com/site/akohlmey/software/topotools/topotools-tutorial---part-2 If you look at just the links to the input scripts on this page, I think it shows you how to set up an OPLS force field. Good luck!

[@anjalishankar]

please make a video on how to find structure of alloy in its solid phase through lammps.

[@vpr1422]

The download links for lammps are down, could any of you share the windows installer with me? :D

[@faheemsultan1908]

Could you please help with building the same system for Si-Ge instead of AlCu. How can I find the Si Ge meam potential?

[@Jere5120]

This is own of the hardest parts of atomistic research, in my opinion! The truth of the matter is, there may not be an Si-Ge MEAM potential out there. But, you can check here, including all the links on this site: icme.hpc.msstate.edu/mediawiki/index.php/Category:Nanoscale#Interatomic_Potentials_available_online You can also try a simple google search for MEAM potentials for SiGe. If you can't find one, you could also try creating one. Check out the other videos on my channel - you will see a few by Dr. Doyl Dickel on how to calibrate MEAM potentials that will be helpful. Good luck!

[@lakshmit540]

Sir , can you show how to model asphalt and graphene. Also, how their interaction works?

[@ABHISHEKKUMAR-wr6if]

Thanks for a wonderful video, but I have a doubt I try to create a carbon hydrogen molecule with large molecular structure, what variable style we have to use.

[@Jere5120]

Thanks for watching! So, you probably don't need to be worried about different variable styles for a C-H molecular structure. But, what you will need to look into is bonds. You will need to create those differently than you create the atoms for a metallic structure. You may need a different program to help you create that structure and then write out a data file that you can import into LAMMPS. Unfortunately, I don't do much work on polymers so I don't know of a great one... I have heard Material's Studio is very good, but it is also not free.

[@silentwalk9925]

I know your great work but why the icme site not working nowadays how can we see your scripts and excercise please

[@pricardo333]

Hey! Excellent videos. You mentioned that you can create a lattice outside of LAMMPS... can you emphasize this? I'd also prefer to use a better interface to create my lattice structures. How do you achieve this?

[@Jere5120]

Thanks for watching! And, great question. The LAMMPS interface for creating a lattice structure is really only good for simple structures. There are a lot of different softwares out there for generating more complicated structures, but most of them are pretty expensive (Materials Studio, and others). One free (and open source) software package I have been using is called Atomic Simulation Environment. It is a python module that really enables the creation of different structures. I am planning on making a series of videos exploring it, but they are still in the works. In the meantime, you can check out the software here: wiki.fysik.dtu.dk/ase/ Good luck!

[@harshvardhangavandi7733]

Thank you so much sir, can you please make a video on segregation of different sized particles while flowing down an inclined plane and of the same material. Thank you !

[@mamoonarehman8738]

Sir i am much confused about determing melting temperature of metal alloy. How can i confirm that my Tm calculated using two phase simulation is accurate?

[@anjalishankar]

you can try to plot density vs temp graph using npt ensemble or energy vs temp hysteresis using nvt... also you can visualize it in OVITO.

[@largemax3672]

Could you give us an example how to run code with voronoi/atom? I want example for how to run it. Input file example will be really helpful, too.

[@jashwanthreddyearla4413]

Very informative video. I am trying to crrate library.meam and NiTi.meam files for monoclinic B19' crystal could you assist me in this context.

[@翟建辉]

Do you know how to generate a FCC with three different kinds of atoms or more?

[@Jere5120]

Hello! Thanks for watching! So, the process for generating crystal structures is relatively straight forward, but you have to know what structure you are going for. Technically, FCC only has one type of atom. So, how you want to arrange your three or more different kinds of atoms in an FCC-like structure. One thing you can do is specify the fcc lattice with the lattice command like before, but then use the "basis" option to change which atoms are included in the basis structure. The FCC lattice in LAMMPS used the cubic basis with 4 basis atoms, so by specifying just one of those 4 atoms, you can create just one type of atom in one of the locations. By repeating this process for all of the different basis positions, you can create your FCC lattice with different atom types in each position of the fcc basis. I hope this helps!

[@翟建辉]

Thank you very much. I will try your method later!

[@mukeshkumarsharma8692]

Thanks for such video, can you please tell me how to set the temperature , i want to calculate for different temperatures, i tried but some how it is fixed for whole iteration cycle...

[@Jere5120]

There are several ways to add a temperature actually. What I typically do is use the "create velocity" command in lammps. It will create a random velocity on each atom with a magnitude that will represent the temperature. You typically need to do an equilibration step (with a thermostat) after that to even out all the velocities to the temperature you want. As long as you have a thermostat active though, the temperature is usually not constant (even if you want it to be!)

[@mareeduveeramohanrao7603]

without custom lattice, we can not create sio2 glass with 1:2 , with group id give to si and other to o

[@kidding686]

kzbin.info/www/bejne/h5XcmJyYgrxkd6s

[@kidding686]

Watch this video it will address your specific problem

[@Blitzkreig261]

Great, helpful tutorial!

[@faheemsultan1908]

You have used simple cubic system for both Al and Cu whereas both are FCC. Can you tell me please why have you used sc?

[@Jere5120]

Thanks for watching! You are correct that on their own, both Al and Cu are FCC materials. But, when they are mixed together in an alloy, they will have to form some different kind of structure together. I don't actually know whether the B2 structure I used here is the most favorable, it was just an easy structure to create, so I used it as an example. If you were going to do something with these materials for real, you would need to see what the most favorable structure was for the alloy.

[@osmanfb1]

Thanks for a nice useful tutorial. I have a question, what does "S" is the element designations mean? I had thought it was for a sulfur atom, but obviously wrong :-)

[@Jere5120]

Thanks for watching! And, good question! I thought the same thing when first saw the designation myself. As you mentioned, it doesn't stand for Sulfur. Unfortunately, I don't actually know the answer. However, I can say that the element designation in the library file can actually be totally arbitrary - LAMMPS only uses that information in order to pick the correct element from your file. That means that you could use any combination of letters, so long as there was a matching entry in the library.meam file. So, my guess is that the researchers who created the potential wanted to have a way of distinguishing these 5 potentials from the other versions of the same element, and chose (for some reason) to put an "S" on the end.

[@Chandansingh-818]

thank you for this video, how to generate eam file for oxides like alumina

[@sparetime6]

hi, it's very useful video you make. i have problem to simulate grain growth process of metal with lammps. can you help me to understand how to do it ? please

[@Jere5120]

+qodi run Thanks for watching! That is a good challenge, and one I have been thinking of myself recently. The difficulty of doing grain growth in LAMMPS (or any MD) is of course the time scale. Your grains have to be so unstable that they are growing in mere nanoseconds. At least, that is the straight forward way of doing it. Another way I have been thinking of, but haven't tried yet, would be to use some form of Monte Carlo MD instead. You can take a look at the LAMMPS documentation for a couple of options for MC. That may get around your time scale issue, though you still have to make sure the results are physical. Good luck!

[@abrarfaiyad285]

Could you please provide the 5 element meam potential file link as well TIA

[@Jere5120]

Thanks for watching! Actually, the 5 element potential should be included in your LAMMPS potential file. The elements are in the library.meam file under their element names with an S. For example, the aluminum for the 5 element potential is under "AlS".

[@asimriaz8740]

Hi Sir, I am trying to do the molecular dynamics of metal oxides (Ce-V-O) under methane (Reduction), CO2 and H2O (oxidation) at high temperature for redox cycling. I couldn't find the potentials for Ce-V, V-O, V-C, and Ce-C. I was wondering if i could get your email or any other contacting details so i can discuss it in details, even to collaborate in the project i'm working on. Thanks

[@Jere5120]

That is a fascinating and tough problem. You can email at [e-mail]

[@saeedalmishal7658]

Can you please help or upload a tutorial on how to plot the radial distribution function using LAMMPS? Thank you very much for everything

[@Jere5120]

Thanks for watching! I would use the modifier available in OVITO. You can find it under "Coordination analysis". You can then output the data to a text file from there.

[@saeedalmishal7658]

@@Jere5120 Thank you very much

[@vadymkovalenko1638]

Great tutorial! Unfortunately, files are no longer available.

[@МихаилУразалиев]

Now add '' .html '' ... icme.hpc.msstate.edu/mediawiki/index.php/File:AlCu.meam.txt.html

[@saeedalmishal7658]

Thank you very much for this video, I'm trying to model CaTiO3 on Lammps and it's giving me hard time, can you help please?

[@Jere5120]

Thanks for watching! I'd be glad to try to help. What difficulties are you having?

[@saeedalmishal7658]

Thank you very much for responding,,, First how can I add the atomic ratio of 1:1:3 of Ca:Ti:O to the input script in Lammps? Secondly, CaTiO3 is a perovskite structure so how can I generate the atomic positions or the atomic structure in Lammps? and are they actually generated in lammps or should they be generated in another software?

[@Jere5120]

By creating the correct structure, you should have the correct atomic ratio. Depending on the geometry you want, you probably could make the structure in LAMMPS - but you will have to specify the basis vectors and atom positions in the base cell manually. Unfortunately I do not know of a package that has the crystal structure for perovskite already built in. For LAMMPS, you will need this command: lammps.sandia.gov/doc/lattice.html then use the a1, a2, a3 options to specify the orthorhombic cell, and repeated uses of the "basis" option to specify the atoms in that cell. Good luck!

[@saeedalmishal7658]

@@Jere5120 I will try my best and see what I can do. Thank you very much for your help.

[@hafizmuhammadrizwanahmad4896]

ERROR: Lost atoms: original 2000 current 1983 (../thermo.cpp:397) please guide . its give me error in your file

[@Jere5120]

Thanks for watching! This error can happen for a couple of reasons - I am kinda surprised it happened in this script. One cause of this error is when atoms are moving too quickly that the updated positions cross a non periodic boundary and are lost. I think it can also happen if the updated position is an invalid number (could be type overflow, or division by 0 or something). I would check two things: make sure that the pair potential definition matches, and that you have the potential files that I used. If a potential is bad, I could see something like that happening. Also, you could add a minimization step before any other runs to make sure the atoms are not too close to each other.

[@kidding686]

I am modifying the input file and potential to run an AlSi simulation and I am getting a similar error. ERROR: Lost atoms: original 2000 current 1992 (../thermo.cpp:442)

[@anjalishankar]

getting unknown pair style error.......! NEED help

[@ernestobarraza1469]

Can you paste the entire error here?

[@anjalishankar]

@@ernestobarraza1469 thank you for your reply....it turned out that i did not install the meam package properly....eventually, it worked properly

[@ernestobarraza1469]

@@anjalishankar Glad you got it to work!

[@ankuranand3533]

help me in making bi crystalline atom for graphene in lammp

[@Jere5120]

I am not sure I understand the question. Are you looking for a graphene type grain boundary?

[@kidding686]

I am getting a warning to use pair_style meam/c instead of just meam. And the an error: Did not find all elements in MEAM Library file. I have my library file in the same directory as my input file

[@kidding686]

fixes the issue if I delete the library.meam file from the directory

[@Jere5120]

Thanks for watching! It sounds like the library.meam file you are using may not have the elements specified in your input file. It may be that your library file is missing an entry (and when you delete it, it defaults to the the one in the distribution "potentials" file) - or there could be a typo in the pair_coeff command. I am glad it is working!

[@mareeduveeramohanrao7603]

how to create simple cubic of SiO2 structure with 1:2

[@Jere5120]

Thanks for watching! If you are just using LAMMPS to create your geometry, then first you will need to know what the crystal structure is. Then, you can use the custom lattice command (lammps.sandia.gov/doc/lattice.html) to create that structure with basis atoms, one for Si and two for oxygen. Then, when you create the atoms, make sure you assign the correct type to the correct basis atom using the "basis" keyword (lammps.sandia.gov/doc/create_atoms.html).

[@muhammadal-sandoqachi8882]

please you can help me for solve problems at lammps

[@Jere5120]

Sure! You are welcome to ask a question here in the comments, and I will try to help you out!

[@babuj6949]

Thank you very much.

[@mareeduveeramohanrao7603]

thanks for ur help

[@thomasdoconnor]

BRO that keyboard THOCK

[@likose53]

thanks a lot

[@mlorono]

Hello, thanks for the videos!. However, I have tried to follow you, but it always gave me this: read_data CPU = 0.025557 secs Reading potential file library.meam with DATE: 2012-06-29 Reading potential file AlSiMgCuFe.meam with DATE: 2012-06-29 ERROR: MEAM library error 3 (../pair_meamc.cpp:596) Last command: pair_coeff * * library.meam AlS AlSiMgCuFe.meam AlS Thanks.

[@mlorono]

running, i have modified the AlSiMgCuFe.meam! Thanks

[@mareeduveeramohanrao7603]

it is not crystal structure, it is a amorphous

[@Jere5120]

Well, if it is amorphous, you won't have a simple cubic structure. But, that does make it somewhat easier. You can use "create_atoms" with the "random" keyword (lammps.sandia.gov/doc/create_atoms.html) to create atoms with random locations inside your simulation box. If you simply make twice as many oxygen atoms as silicon, you will have the correct ratio. Then, if you want to group them by type, you can use the group command (lammps.sandia.gov/doc/group.html). Also, if you use randomly created atoms like this, you should be sure to run a minimization step before anything else - the atoms can be overlapping and have very high forces. A minimization should make sure they are not overlapping. Good luck!

[@kidding686]

How would you write the code to double the ratio of one of the elements or e.g I need to add 15% of an element. I can't see how the following code will change # Create Al atoms lattice sc 4.0 create_atoms 1 region whole # Create Cu atoms lattice sc 4.0 origin 0.5 0.5 0.5 create_atoms 2 region whole

[@SAPHELPER]

Hi I have some doubts in lammps I need some help plz respond it is very important for me: I can pay u for ur time If u don't want to help for free.

[@Jehosafath]

Sir, all the links are dead now :(

[@LucasValdiviaAraya]

di u find the files?