RE: [ESPResSo] Fwd: analyze nbhood & Bond-angle Interactions

[Limbach, Hans Joerg, LAUSANNE, NRC-FS]

Dear Lorenzo,

(1) A bond (or dihedral) angle potential should always be used in
connection with two body bond potentials which will set the interaction
range. E.g. if you place a bond angle between the particles 0, 1 and 2
(1 being the middle particle) you schoud set bonds between the particles
0 and 1 and between 1 and 2. Otherwise Espresso will crash at some
point.
(2) nbhood does not give the particles in a specific order. I mentioned
it to you since I thought nbhood could be a starting point to solve your
problem. After you got the list you have to find out on the tcl level
how the particles are arranged towards each other.

Best regards,
Hanjo

> -----Original Message-----
> From: address@hidden 
> [mailto:address@hidden On Behalf Of 
> Lorenzo Isella
> Sent: vendredi, 19. octobre 2007 17:38
> To: address@hidden
> Subject: [ESPResSo] Fwd: analyze nbhood & Bond-angle Interactions
> 
> Dear All,
> Thanks for the very useful bits of advice I got on the list.
> I am now running some experiments with really simple systems, 
> made up of 3 particles.
> The idea is how suggested on this list: when particles get 
> "close enough", I stick them together with a potential which 
> depends on the angle (a rigid bond).
> I have a few questions for the list.
> (1) if I look at Eq. 5.16 of the Espresso manual 
> (bond_angle_cosine) I do not see how the length of the bond 
> is defined. I mean: the particles will have a preferred 
> orientation but what is the range of the interaction? Can I 
> specify it?
> (2) the command analyze nbhood  tells me how many particles I 
> have within a given distance from a certain particle. It 
> returns a list of particle numbers, but what is the meaning 
> of the order in which the numbers appear? I mean, say I am 
> looking for the particles surrounding particle 1 and I get 
> the array 0 1 2. Does that tell me anything about which 
> particle is in the middle?
> Particles 0 and 2 could be on the same side wrt particle 1, 
> or particle 1could be somewhere in between. This is important 
> seen that I need to know which is the central particle to set 
> up the bond-angle interaction.
> (3)In the end it all boils down to the problems I am 
> experiencing in setting up a system with a few particles able 
> to interact via a bond-angle interaction.
> In the code below, everything should work (in the end I check 
> that my particle 1 has the bonded interaction), but from VMD 
> that does not seem to be the case...
> Cheers
> 
> Lorenzo
> 
> set n_part 3; set density 0.007
> set box_l [expr pow($n_part/$density,1./3.)]
> 
> setmd box_l $box_l $box_l $box_l
> setmd periodic 1 1 1
> 
> inter 0 angle 700.0 0.8
> set q 0; set type 0
> 
> # I now set the particle charge to 0 and see if the code 
> works the same of not
> 
> for {set i 0} { $i < $n_part } {incr i} { set posx [expr 
> $box_l*[t_random]] set posy [expr $box_l*[t_random]] set posz 
> [expr $box_l*[t_random]] part $i pos $posx $posy $posz q $q 
> type $type }
> 
> puts "[part ]"
> 
> setmd time_step 0.01; setmd skin 0.4
> set temp 0.0005; set gamma 0.01
> thermostat langevin $temp $gamma
> set integ_steps 200
> set vmd "yes"
> 
> if { $vmd == "yes" } {
>     prepare_vmd_connection tutorial 3000
>     exec sleep 4
>     imd positions
> }
> 
> set activate_fene 1
> set fene_warning 1
> set integ_step 20
> 
> for {set i 0} { $i < 2000 } { incr i} {
> set temp [expr [analyze energy kinetic]/(1.5*$n_part)] #puts 
> "t=[setmd time] E=[analyze energy total], T=$temp"
> integrate $integ_steps
> 
> if { $vmd == "yes" } { imd positions }
> set acti [analyze nbhood 1 4.5]
> 
> 
> set neighbor [analyze nbhood 1 4.]
> #puts " the neighbors are, $neighbor "
> set mylen [llength $neighbor]
> 
> 
> if { $mylen == 3 && $activate_fene == 1 } {
> 
> part 1 bond 0 0 2
> puts "activated angle bond!"
> set activate_fene 0
> }
> }
> 
> puts "[part ]"
> 
> puts "So far so good"
> 
> _______________________________________________
> ESPResSo mailing list
> address@hidden
> https://fias.uni-frankfurt.de/mailman/listinfo/espresso
>