LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] ERROR: No pair reax/c for fix qeq/reax
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lammps-users] ERROR: No pair reax/c for fix qeq/reax


From: Rajesh <creativeidleminds@...24...>
Date: Mon, 2 Oct 2017 10:16:43 +0530

Dear RAy,
Is it correct now?

pair_style hybrid lj/cut 2.5 reax/c NULL
pair_coeff * * reax/c ffield.reax.BN B C H N
neigh_modify exclude type 1 2
neigh_modify exclude type 1 3
neigh_modify exclude type 2 4
neigh_modify exclude type 3 4
pair_coeff 1 3 lj/cut 1.0 1.5
pair_coeff 1 2 lj/cut 1.0 1.5
pair_coeff 2 4 lj/cut 1.0 1.5
pair_coeff 3 4 lj/cut 1.0 1.5


On Sun, Oct 1, 2017 at 11:52 PM, Rajesh <creativeidleminds@...24...> wrote:
ok. But how can I define the coeff. for different atom pairs in lj? Please correct me.

On Sun, Oct 1, 2017 at 11:50 PM, Ray Shan <rshan@...1795...> wrote:
Both inputs are wrong. You only have 1 lj/cut so there no need to sequence them in pair_coeff commands.

Ray


On Oct 1, 2017, at 10:02, Rajesh <creativeidleminds@...24...> wrote:

Dear Ray,
I think this will work for me. But I am not sure whether pair_coeff 1 2 none option can be used many times or not? Also I tried with both the options mentioned below but I am getting +ve potential energy in upper one. That means first option is not working. Please advise.

pair_style hybrid lj/cut 2.5 reax/c NULL
pair_coeff * * reax/c ffield.reax.BN B C H N
pair_coeff 1 2 none
pair_coeff 1 3 none
pair_coeff 2 4 none
pair_coeff 3 4 none
pair_coeff 1 3 lj/cut 1 1.0 1.5
pair_coeff 1 2 lj/cut 2 1.0 1.5
pair_coeff 2 4 lj/cut 3 1.0 1.5
pair_coeff 3 4 lj/cut 4 1.0 1.5

or 
pair_style hybrid lj/cut 2.5 reax/c NULL
pair_coeff * * reax/c ffield.reax.BN B C H N
neigh_modify exclude type 1 2
neigh_modify exclude type 1 3
neigh_modify exclude type 2 4
neigh_modify exclude type 3 4
pair_coeff 1 3 lj/cut 1 1.0 1.5
pair_coeff 1 2 lj/cut 2 1.0 1.5
pair_coeff 2 4 lj/cut 3 1.0 1.5
pair_coeff 3 4 lj/cut 4 1.0 1.5

On Sun, Oct 1, 2017 at 9:10 PM, Ray Shan <rshan@...1795...> wrote:
Please reply to the list, not to me.

No, it is not correct. Re-read the pair hybrid doc page.

Ray


On Sep 30, 2017, at 09:28, Rajesh <creativeidleminds@...24...> wrote:

Dear Dr Ray,

Thank you for your response. Please correct me if I am wrong.

pair_style hybrid lj/cut 2.5 reax/c NULL
pair_coeff * * reax/c ffield.reax.BN B C H N
pair_coeff 1 3 lj/cut 1 1.0 1.5
pair_coeff 1 2 lj/cut 2 1.0 1.5
pair_coeff 2 4 lj/cut 3 1.0 1.5
pair_coeff 3 4 lj/cut 4 1.0 1.5

Is it right?

On Fri, Sep 29, 2017 at 6:29 AM, Rajesh <creativeidleminds@...24...> wrote:
Dear Dr Ray,

Thank you for your response. I think I should go by this

pair_style hybrid lj/cut 2.5 reax/c NULL
pair_coeff * * reax/c ffield.reax.BN B C H N
pair_coeff 1 3 lj/cut 1 1.0 1.5
pair_coeff 1 2 lj/cut 2 1.0 1.5
pair_coeff 2 4 lj/cut 3 1.0 1.5
pair_coeff 3 4 lj/cut 4 1.0 1.5

Is it right?


On Fri, Sep 29, 2017 at 1:56 AM, Ray Shan <rshan@...1795...> wrote:
Fix qeq/reax doesn’t support multiple instances of pair_sytle reax/c in hybrid – you can only have one reax/c in pair_style hybrid or hybrid/overlay.

It doesn’t make sense anyway, to have to reax/c instances since you only have one ffield.reax.BN.  You also only need one lj/cut. 

Ray

On 9/26/17, 7:22 AM, "Rajesh" <creativeidleminds@...24...> wrote:

Dear users,

I using hybrid pair style with reax/c. I have 4 type of atoms
1  B
2  C
3  H
4  N

I want pair_style reax/c for B and N and C and H. And 1-2, 1-3, 2-4, 3-4 should be modelled with lj/cut potential. I have defined this as

pair_style hybrid lj/cut 2.5 lj/cut 2.5 lj/cut 2.5 lj/cut 2.5 reax/c NULL reax/c NULL
pair_coeff * * reax/c 1 ffield.reax.BN B NULL NULL N
pair_coeff * * reax/c 2 ffield.reax.BN NULL C H NULL
pair_coeff 1 3 lj/cut 1 1.0 1.5
pair_coeff 1 2 lj/cut 2 1.0 1.5
pair_coeff 2 4 lj/cut 3 1.0 1.5
pair_coeff 3 4 lj/cut 4 1.0 1.5

But when I run my input script I get the error "ERROR: No pair reax/c for fix qeq/reax". Please advise.


A part of my script is

# VARIABLES
variable fname index we.dat
variable simname index we
# Initialization
units real
boundary p p p
atom_style charge
#log  log.${simname}.txt
read_data we.dat
#read_data gp.data add append shift 80.0 80.0 75.0 group gp 
#read_restart restart.pe.dreiding4
# Dreiding potential information

write_restart restart.re.50000

pair_style hybrid lj/cut 2.5 lj/cut 2.5 lj/cut 2.5 lj/cut 2.5 reax/c NULL reax/c NULL
#pair_style reax/c NULL #lmp_control #checkqeq no
pair_coeff 1 3 lj/cut 1 1.0 1.5
pair_coeff 1 2 lj/cut 2 1.0 1.5
pair_coeff 2 4 lj/cut 3 1.0 1.5
pair_coeff 3 4 lj/cut 4 1.0 1.5
pair_coeff * * reax/c 1 ffield.reax.BN B NULL NULL N
pair_coeff * * reax/c 2 ffield.reax.BN NULL C H NULL



neighbor 2 bin
neigh_modify every 10 delay 0 check no

######################################
############### INITIALISATION #######
######################################

fix             spcl all qeq/reax 5 0.0 10.0 1e-6 reax/c #param.qeq

compute           mytemp  all  temp
compute           peratom all  pe/atom
compute           strs all stress/atom NULL
compute           pr all reduce sum c_strs[1] c_strs[2] c_strs[3]
min_style       cg
minimize          0 1.0e-10 10000 10000
min_modify dmax   0.05
#####################################################
# Equilibration Stage 1 (Langevin dynamics at 500 K)
velocity  all create 500.0 1231141
fix     1 all nve/limit 0.05
fix     2 all temp/berendsen 500 500 10 #042900
thermo_style custom step lx ly lz  temp  pe  etotal  pxx pyy pzz press density
dump           1 all custom  1000  EQ_*.dump id type mass x y z vx vy vz c_peratom q
thermo          1000
timestep 0.1
run  50000
unfix 1
unfix 2
undump 1


But when I run my input script I get the error "ERROR: No pair reax/c for fix qeq/reax". Please advise.