Modifying the datafile to generating a infinitely length chain

Hello sir/madam,

I want to generate a infinite length chain of (CH2)n. When dealing with the atoms at the boundary, I do as following in the datafile for LAMMPS:

(The configuration of the chain and the atom index are shown in the figure attached.)

(1)Add to the bond list:

BondID BondType m 0

(2)Add to the angle list (totally 6 new angles to be added)

AngleID AngleType m-3 m 1

AngleID AngleType m+1 m 1

AngleID AngleType m+2 m 1

AngleID AngleType m 1 2

AngleID AngleType m 1 3

AngleID AngleType m 1 4

(4)Add to the dihedral list (totally 15 new dihedrals to be added):

DihedralID DihedralType m-3 m 1 2

DihedralID DihedralType m-3 m 1 3

DihedralID DihedralType m-3 m 1 4

DihedralID DihedralType m+1 m 1 2

DihedralID DihedralType m+1 m 1 3

DihedralID DihedralType m+1 m 1 4

DihedralID DihedralType m+2 m 1 2

DihedralID DihedralType m+2 m 1 3

DihedralID DihedralType m+2 m 1 4

DihedralID DihedralType m 1 4 5

DihedralID DihedralType m 1 4 6

DihedralID DihedralType m 1 4 7

DihedralID DihedralType m-6 m-3 m 1

DihedralID DihedralType m-2 m-3 m 1

DihedralID DihedralType m-1 m-3 m 1

If I do these things in the data file, will I generate a infinitely length chain?

Thank you in advance.

Best wishes,

Jiayuan Miao

CH2chain.png

I've said everything I have to say on this topic.
I don't think you are thinking about an infinite
chain correctly.

Steve

I am sorry to bother you so many times.

I am just not clear about your statement in your email:
(1)Do you mean that the "infinitely long chain" produced by repeating one segment of a polymer chain periodically is in fact a ring? Because both ends of the segment must match each other properly?
(2) I could not understand this sentence in your last email:
"You can also fill your box with N monomers and bond them as

one long chain, so long as every (N,N+1) pair is “close” to each

other in a periodic sense, so that the long chain crosses the box

many times. But if you also bond 1 to N, then you have a ring."

Could you give a graph to illustrate this?

Thank you for your help.

Best,
Jiayuan
dear jiayuan, 

i think your concept of periodic boundary conditions is not quite correct - the copies of your simulation box are not actual replicas of your system, but only virtual ones 
that behave in the same way as your simulation box. so you need to know how much this artifact influences your results. 

thus you can not place any particles into the copies of your box. maybe you should have a look at the concept in textbooks or - even better - code a quick and dirty md engine 
yourself that works with and without pbc. this is quite a bit of work but not terribly much and it would strongly contribute to your understanding of the ongoings in lammps. 

best, 

nikita