In this example, we’ll build a gold nanoparticle and equilibrate it to a temperature of 300K. In future posts, we’ll add ligands and solvent to this structure. Here’s a sneak preview of the nanoparticle we’ll make here:
- Start with an input .omd file for the metal of your choice. We’ll be creating gold.omd for this example.
<OpenMD>
<MetaData>
molecule{
name = "Au";
atom[0]{
type = "Au";
position(0.0, 0.0, 0.0);
}
}
component{
type = "Au";
nMol = 1;
}
forceField = "SC";
forceFieldFileName = "SuttonChen.QSC.frc";
</MetaData>
</OpenMD> - Now to build the nanoparticle. We’ll choose a radius of 15 Å and use the lattice constant of gold, 4.08 Å. We’ll call our initial structure NP15.omd.
nanoparticleBuilder --latticeConstant=4.08 --radius=15 gold.omd -o NP15.omd
- To take a look at the structure we’ve just created, you can use the following command:
Dump2XYZ -i NP15.omd
to create a file called NP15.xyz, which can be viewed in VMD, Jmol, or any other chemical structure viewer.
- Add the following lines to the new NP15.omd file below the forceFieldFileName line. This sets some details for the simulation.
ensemble = "LHull";
targetTemp = 5;
targetPressure = 1;
viscosity = 0.1;
dt = 4.0;
runTime = 2E5;
sampleTime = 2000.0;
statusTime = 4;
seed = 985456376;
usePeriodicBoundaryConditions = "false";
tauThermostat = 1E3;
tauBarostat = 5E3; - NanoparticleBuilder carves a nanoparticle of our chosen radius out of a perfect gold crystal. We need to give the atoms some initial velocities before we start equilibrating. We’ll start it out at 5K:
thermalizer -t 5 NP15.omd -o NP15_5K.omd
- For the first step in the equilibration we need to let the gold lattice structurally relax. NP15_5K.omd can now be run:
openmd NP15_5K.omd
- Running the simulation will create several new files. NP15_5K.dump contains the trajectory of the simulation. Statistics such as temperature, pressure, and energy will be recorded in the NP15_5K.stat file and can be viewed using:
xmgrace -nxy NP15_5K.stat
- The end-of-run file NP15_5K.eor stores the last configuration of the simulation. We’ll copy it to a new .omd file.
cp NP15_5K.eor NP15_100K.omd
- To continue with the equilibration we need to change the targetTemp of NP15_100K.omd. We’ll increase it to 100 and run the NP15_100K.omd file.
- We’ll continue the procedure of copying the .eor file to a new .omd file and increasing the temperature until we’ve reached 300K. Temperature increases of 50 – 100K and simulation times of 100 – 200 ps are reasonable.
#1 by Hussain on March 23, 2014 - 12:29 pm
Quote
I’m having trouble running nanoParticleBuilder. It says command not found.
I’ve done this install process (from the README)
1) Get, build, and test the required pieces above.
2) mkdir build
3) cd build
4) cmake ..
5) make
6) sudo make install
#2 by Dan Gezelter on March 23, 2014 - 1:13 pm
Quote
1) Make sure that the binary install directory is in your path. By default, OpenMD installs binaries in /usr/local/openmd/bin so you might need to add this line to your .bashrc:
PATH=${PATH}:/usr/local/openmd/bin; export PATH
If you use csh or tcsh: add this to your .cshrc or .tcshrc
set path = ($path /usr/local/openmd/bin)
2) You also appear to have a capitalization issue. The program is nanoparticleBuilder (with no capitalized P).
#3 by Hussain on March 23, 2014 - 5:40 pm
Quote
Thank you! It’s fine now.
#4 by Amanda on May 14, 2014 - 3:03 pm
Quote
Hello, OpenMD is saying
“OpenMD FATAL ERROR: SimCreator: Cannot open file: gold.md”
Could someone please email me a copy of the md. file to
amanda.cameron@hotmail.com
Thanks
#5 by Dan Gezelter on May 14, 2014 - 4:23 pm
Quote
Amanda,
The gold.md file just contains the definition of the molecule and some parameters. You can cut and paste this as a gold.md:
<MetaData>
molecule{
name = "Au";
atom[0]{
type = "Au";
position(0.0, 0.0, 0.0);
}
}
component{
type = "Au";
nMol = 1;
}
forceField = "SC";
forceFieldFileName = "SuttonChen.QSC.frc";
</MetaData>
</OpenMD>
#6 by Amanda on May 16, 2014 - 12:22 pm
Quote
Thanks, I was only missing the colon in .
How can I fix:
OpenMD FATAL ERROR:
Integrator Factory can not create LHull Integrator
The lines from step 4 were added to the md. file
#7 by Cbu on May 22, 2014 - 9:05 am
Quote
Is there a way to generate a top. file of the nanoparticle? I want to add the particle to an amino acid in Gromacs.
#8 by Sogol on June 23, 2014 - 7:09 am
Quote
I have a problem ! I need a nanoparticle, with 10 nm radius, open MD generates the *.md for me but when I use command ./Dump2xyz , it is killed !!I think the problem is because of the big size of the particle, what should I do !!!!! it converts the one I tested with the radius of 10 A but not 100 A !!
#9 by Dan Gezelter on June 30, 2014 - 11:00 am
Quote
That’s a pretty large nanoparticle. The MD file is 29 MB, and the derived XYZ file is around 11MB. It is likely that you are hitting a memory, disk, or stack size limit. We can upload the 10 nm file if you’d like to use it.
#10 by Carlos D Garcia on July 14, 2014 - 11:49 am
Quote
I would LOVE to get a link for the pdb file (10 nm particle)…
#11 by Carlos D Garcia on July 14, 2014 - 11:50 am
Quote
In the meantime, I’m downloading/installing all the components
#12 by Dan Gezelter on July 15, 2014 - 12:59 pm
Quote
Carlos,
Here are links to the structure files for the 10 nm (100 angstrom) radius particle:
Note that at this size, the gold nanoparticle has 246617 atoms. The field width for the atom serial number in the PDB file format isn’t wide enough to handle structures this large. Typical conversion scripts either mess up the serial number or displace the coordinates once you get above 100000 atoms. You might be able to coerce the XYZ file into a PDB format that your program will accept, but we don’t use that format internally in OpenMD.
#13 by Martin Vala on September 5, 2014 - 7:58 am
Quote
I have same problem as Amanda from #6
OpenMD FATAL ERROR:
Integrator Factory can not create LHull Integrator
How can i fix it?
#14 by Dan Gezelter on September 11, 2014 - 11:18 am
Quote
I think we’ve resolve this on the mailing list, but in general, that error means that OpenMD was built without the qhull library. The LangevinHull integrator requires this library in order to function.
#15 by Lano on November 13, 2014 - 6:51 am
Quote
Hi I am using NAMD and VMD. I have made Au.pdb and now I need to bond them (psf). Does anyone know how to do that or say need a topology file please? I was using inorganicbuilder through VMD to make a bond but it did not work!
Thanks
#16 by Dan Gezelter on November 13, 2014 - 10:31 am
Quote
Lano,
You might want to try asking your question over at the NAMD mailing list.
#17 by otheetyi on May 27, 2015 - 11:55 pm
Quote
Dear Dan, i really have a problem to build 4-5 nm gold crystal nanoparticle. I would really appreciate if you could give me a link to get the pdb or xyz file.
Thank you 🙂
#18 by Martin on July 9, 2015 - 11:49 am
Quote
Dear Dan,
I would like to perform solvent explicit MD simulations with OPENMD, coupling a gold nanoparticle generated in OPENMD with a peptide, do you know where I can find information to perform this type of calculation?
Greetings
#19 by Hirak on September 18, 2015 - 4:49 pm
Quote
Hello Dr. Gezelter,
Thank you for the wonderful program, its so easy to operate that a novice like me could able to install and run only a couple of hours effort.
I have built the nanoparticle as prescribed method, now I want to show their agglomeration to conform a bigger particle in a solvent environment. Please find me a way through your enlightened views.
Thanks,
Hirak
#20 by Rakesh Gupta on November 25, 2015 - 9:48 am
Quote
Dear Dr. Gezelter
Thank you for such amazing package. I have used it to create AuNP for simulation.
I would like to extend my study with surface coated AuNP’s. Are there any scripts or programs available in OPENMD to create surface coated gold nano particles?
Thanks
#21 by Dan Gezelter on November 25, 2015 - 11:45 am
Quote
Rakesh,
We typically use Packmol to place ligands near the surface of the particle. Here’s a simple packmol input file that will take two xyz files (one for the particle called NP30.xyz and one for a simple dodecanethiolate ligand called C12.xyz). NP30 was a gold nanosphere with a 30 Å radius. C12.xyz is a structure containing the atom locations for a template ligand molecule.
seed 6329
output NP30_C12.xyz
filetype xyz
structure NP30.xyz
number 1
center
fixed 0. 0. 0. 0. 0. 0.
end structure
structure C12.xyz
number 352
atoms 1
inside sphere 0. 0. 0. 32.
end atoms
atoms 13
outside sphere 0. 0. 0. 47.
end atoms
end structure
After running packmol, we have to convert the structure back to OpenMD’s StuntDouble lines, which we do using a simple awk script:
END {print " </StuntDoubles>\n </Snapshot>\n</OpenMD>"}
Note that packmol doesn’t know anything about our component blocks, so we typically tailor the inputs by hand after getting the coordinates.
#22 by parvin on January 11, 2016 - 2:16 am
Quote
Dear Dr
thanks alot
i want make a hollow nano particle with diametter of 1 micro meter and it’s thickness is 30 nm please guide me?
is it possible?
#23 by Dan Gezelter on January 11, 2016 - 11:52 am
Quote
Parvin,
You can indeed make spherical shells. Here’s one way to make a hollow shell that has an inner radius of 25 Angstroms and an outer radius of 30 Angstroms.
--latticeConstant=4.09 --vacancyPercent=100 \
--vacancyInnerRadius=0.0 --vacancyOuterRadius=25.0 gold.omd
However, the spherical shell you describe would contain roughly 5.5e9 atoms, which is probably more than you would be able to simulate (or visualize).
#24 by mrinal on August 4, 2016 - 5:30 am
Quote
I want to make a nanoparticle coated with other groups.Is it possible?
#25 by Dan Gezelter on August 6, 2016 - 11:11 am
Quote
It is possible to coat a particle with other groups – we typically use packmol to do this step (see response 21 above). I’m working on a separate post that shows how to do this more explicitly.
#26 by mrinal on August 8, 2016 - 11:33 am
Quote
Thanks a lot.But there is an error.
‘At line 332 of file getinp.f90 (unit = 10, file = ‘C12.xyz’)
Fortran runtime error: End of file’
#27 by mrinal on August 9, 2016 - 3:23 am
Quote
Is it possible to get the psf file?please reply.
Pingback: Building and equilibrating gold icosahedra in OpenMD « OpenMD
#28 by Dan Gezelter on August 17, 2016 - 10:45 am
Quote
It sounds like you’ve got an invalid C12.xyz file. We used this file:
S -0.626 1.709 0.0
CH2 0.0 0.0 0.0
CH2 1.54 0.0 0.0
CH2 2.166 -1.407 0.0
CH2 3.706 -1.407 0.0
CH2 4.332 -2.814 0.0
CH2 5.872 -2.814 0.0
CH2 6.498 -4.221 0.0
CH2 8.038 -4.221 0.0
CH2 8.664 -5.628 0.0
CH2 10.204 -5.628 0.0
CH2 10.830 -7.035 0.0
CH3 12.370 -7.035 0.0
OpenMD doesn’t use psf files. If you want to convert the files for use with other codes, you’ll be on your own.
#29 by mrinal on August 17, 2016 - 1:17 pm
Quote
Sir,how to get the c12.xyz file?
#30 by Cesar Rosiles Mendoza on October 1, 2016 - 8:12 pm
Quote
now i am trying without gold and i am using fe3o4 but “Can not find Matching Atom Type for[Fe]” like the liquid example (C3 —CT, O3–OH),then , what is the correct type for [Fe] and all elements of periodic table? i mean where is the list of correct types inside OpenMD folder?
#31 by Amin on October 18, 2016 - 1:43 am
Quote
Dear Dr. Gezelter,
Would you please upload xyz structures of 2, 3, and 4 nm gold nanoparticles?
Thanks
#32 by parvin on October 25, 2016 - 2:28 am
Quote
hi
Dear dr i use
nanoparticleBuilder -o hollowShell.omd –radius=30.0 \
–latticeConstant=4.09 –vacancyPercent=100 \
–vacancyInnerRadius=0.0 –vacancyOuterRadius=25.0 gold.omd in coad in lammps i cant use open md on windows
i need it’s xyz file carbon.
please help me about it or about run open md on win.
i will promise compensate it if i can.
best regards
parvin
#33 by behzad on April 17, 2017 - 2:34 pm
Quote
hi
would yo please answer this question:
i want to simulation Ag and Au nanoparticles with gromacs .but they are not recognized. how can i define these atom types?
#34 by Dan Gezelter on April 17, 2017 - 2:37 pm
Quote
You might have better luck getting an answer at the GMX-Users mailing list.
Start here: http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List
(We aren’t regular users of gromacs, but it has a user community that might be able to help you.)
#35 by behzad on April 21, 2017 - 2:13 pm
Quote
thank you
#36 by kedar on June 30, 2017 - 10:30 am
Quote
how can i convert file.xyz to file.pdb?
#37 by Dan Gezelter on July 4, 2017 - 11:57 am
Quote
You can use openbabel to convert between chemical file types.
#38 by kedar on August 1, 2017 - 10:01 am
Quote
I have seen your 21 number response .i want to attach the ligand to specific nanoparticles.what are the preferable nanoparticles for my ligands ? and how to attach the ligand with nanoparticles at AuNP’S surface?
#39 by kedar on August 8, 2017 - 2:41 am
Quote
sir , please suggest me a route.
#40 by Urooj on December 7, 2017 - 11:09 am
Quote
Sir,
In response to #21, because of my little understanding i didnt get that what type of co-ordinates i will mention in internal sphere and outer sphere?
i want to capped a ligand on spherical gold nano-particle.
please suggest me the files format.
thank you.
#41 by Samuel Awelewa on March 22, 2018 - 6:48 pm
Quote
SIr, I have downloaded the OpenMd file and have installed cygwin on my laptop but I am still yet to figure out how to run the OpenMd.
#42 by Awelewa Samuel on April 17, 2018 - 3:00 pm
Quote
Dear Dr. Gezelter,
I have been able to build OpenMD on Ubuntu but I am still finding it hard to figure how to run it. I have been trying to follow the examples listed above but I am not getting it. I will be glad if you would explain to me how to move forward from here because I need OpenMD for my MSc project.
Thanks.
#43 by Mozammel Hoque on May 15, 2018 - 11:01 am
Quote
Dear Sir,
I am really glad to how you guys helping people. I am just a beginner, I have followed the installation process but I don’t know how to run the program. I am trying to simulate different structure for my experimental 4.5 nm gold nanoparticles. I really appreciate your help. Thanks
#44 by Neha Gandhi on December 28, 2018 - 6:15 am
Quote
Is it possible to build a selenium nanoparticle? The unit cell of gold nanoparticle is triclinic whereas the crystal structure of trigonal selenium has following properties (not triclinic).
4.3662 4.3662 4.9536 90 90 120 P3_121
atom x y z B(1,1) B(2,2) B(3,3) B(1,2) B(1,3) B(2,3)
Se .2254 0 1/3 .0338 .0417 .03775 .02085 0 0
Should we only change the lattice constant in nanoparticle builder? How we do we take into account the angle?
Many thanks,
#45 by Dan Gezelter on December 28, 2018 - 1:14 pm
Quote
Currently, the nanoparticleBuilder code assumes the underlying lattice is one of the cubic structures (FCC, BCC, SC), so you couldn’t use it for the selenium structure you want to simulate.
#46 by sokyem on May 31, 2019 - 7:40 pm
Quote
I have the .omd file for 600A gold nanoparticle and I’m trying to convert it o xyz using DUMP2XYZ .. but the turnout is killed. Can you please help out
#47 by Dan Gezelter on May 31, 2019 - 8:22 pm
Quote
That’s a very large nanoparticle. The particle you describe would contain roughly 5.2e7 atoms, which is probably more than you would be able to simulate (or visualize).
The omd file probably around 6.3GB, and the derived XYZ file will be around 2.4GB, so It is likely that you are hitting a memory, disk, or stack size limit.
#48 by tanni on July 4, 2019 - 5:24 am
Quote
can i create quartz sio2 nanoparticle using openmd?
#49 by Mohammad Zohurul Islam on July 18, 2019 - 10:11 pm
Quote
Dear Sir
I would like to work on the ligand coated gold nanoparticle interaction on lipids monolayer in lung surfactant. I started with gold nanoparticle (diameter 5 nm) from your website but I found the problems like
OpenMD FATAL ERROR:
Integrator Factory can not create LHull Integrator
You already solved this problem (#6) and I have to setup qhull library. I am not able to that. Could you please help me.
Regards
Zohurul