Pulling force keeps increasing

GROMACS version: 2023.3
GROMACS modification: No
Here post your question

I am trying to calculate the friction force in a system where a molecule is subjected to a harmonic spring in the X and Y dimensions and is pulled along the X-axis on a rough polymer slab (see attached figure). The polymer slab is frozen during this process, and the system was equilibrated before performing the pulling simulation. I expected the force data from pullf.xvg to represent the friction force. However, the force appears to increase linearly over time. I am unsure if there is an issue with my input parameters (provided below). Has anyone encountered a similar problem or have any suggestions for resolving this?
Thank you.

integrator = md
dt = 0.001
nsteps = 2000000

nstxout = 1000
nstvout = 1000
nstenergy = 1000
nstlog = 1000
nstxout-compressed = 10000
compressed-x-grps = System

cutoff-scheme = Verlet
nstlist = 10
rcoulomb = 1.4
rvdw = 1.4
pbc = xyz
DispCorr = no

coulombtype = PME
pme_order = 4
fourierspacing = 0.12

tcoupl =Nose-Hoover
tc-grps = system
tau_t = 0.2
ref_t = 298

pcoupl = no

continuation = no
constraint_algorithm = lincs
constraints = none
lincs_iter = 1
lincs_order = 4
lincs_warnangle = 30
morse = no

gen_vel = no
gen-temp = 298
gen-seed = -1

;*************
; Pull code *
;*************
pull = yes
pull-ncoords = 1
pull-ngroups = 1
pull-nstxout = 1000
pull-nstfout = 1000
pull-xout_average = no
pull-fout_average = no

pull-group1-name = PTM
pull-coord1_type = umbrella
pull-coord1_geometry = direction
pull-print-components=yes
pull-coord1_dim = Y Y N
pull-coord1_groups = 0 1
pull-coord1-origin = 0.0 0.0 0.0
pull-coord1_vec = 1.0 0.0 0.0
pull-coord1_start = yes
pull-coord1_rate = 0.05
pull-coord2_k = 700

freezegrps = PPN
freezedim = Y Y Y


If the polymer slab is frozen it is quite probable that the group you pull will get stuck on it. The high pull force means that the reference position (the other end of the harmonic spring) is far away from the group you pull. Why do you want the polymer slab to be frozen?

The pull group continues to move on the polymer slab. If I don’t freeze the polymer slab, the pull group will drift away from it. I’m unsure if position restraining the polymer is a better option.