LIGGGHTS - Vibration Effect on Particles Stuck on Wall

Submitted by egebat7 on Wed, 02/24/2021 - 15:40

I am trying to simulate an ultrasound vibration effect on stuck particles on wall by using "fix move all move/mesh mesh cad1 wiggle" command. I know this line is not creating a wave vibration but still I would like to see if "wiggle" command can remove the particles from the wall if I use low amplitude with very high frequency. I use cohesion model to stick particles on the wall.

However, the particles are not being removed from the wall with this vibration effect. My question is, am I working on something that LIGGGHTS has some limitations or should I use a different command to vibrate my walls. Or maybe my model is wrong?

Note: My particles are very small so I needed to decrease the volume limit to 1e-15.

I am sharing my input file:

atom_style granular
boundary f f f
units si
communicate single vel yes
newton off

#Definition of boundaries

variable xmin equal -1
variable xmax equal 1
variable ymin equal -1
variable ymax equal 1
variable zmin equal 0
variable zmax equal 0.6

#Definition of the timestep
variable dt equal 5e-8

variable natoms equal 2 lid

####Young Modulus####

variable youngmodulus1 equal 2e8 #N/mm²
variable youngmodulus2 equal 2e8 #N/mm²

####Poission ratio####

variable poission1 equal 0.45
variable poission2 equal 0.45

####coefficient of restitution####

variable CoR11 equal 0.2
variale CoR12 equal 0.2
variable CoR21 equal 0.2
variable CoR22 equal 0.2

####sliding friction coefficient####

variable sf11 equal 0.3
variable sf12 equal 0.36
variable sf21 equal 0.36
variable sf22 equal 0.0

####rolling friction coefficient####

variable rf11 equal 0.3
variable rf12 equal 0.75
variable rf21 equal 0.75
variable rf22 equal 0.75

####cohesion energy density####

variable ce11 equal 300000
variable ce12 equal 60000000
variable ce21 equal 60000000
variable ce22 equal 300000

variable nradii equal 1
variable radius1 equal 0.000040 #m
variable frac1 equal 1 #100%
variable density equal 700 #kg/m³

variable filltime equal 1 #seconds
variable fillmass equal 0.0000001 #kg
variable fillmassrate equal ${fillmass}/${filltime} #kg/s
variable fillsteps equal ${filltime}/${dt} #Transform time to iteration steps

region reg block ${xmin} ${xmax} ${ymin} ${ymax} ${zmin} ${zmax} units box

create_box 2 reg

neighbor 0.004 bin
neigh_modify delay 0
pair_style gran model hertz tangential history rolling_friction epsd2
pair_coeff * *

timestep ${dt}

fix integrator all nve/sphere
fix gravi all gravity 9.81 vector 0.0 0.0 -1.0

fix m1 all property/global youngsModulus peratomtype ${youngmodulus1} ${youngmodulus2}
fix m2 all property/global poissonsRatio peratomtype ${poission1} ${poission2}
fix m3 all property/global coefficientRestitution peratomtypepair ${natoms} ${CoR11} ${CoR12} ${CoR21} ${CoR22}
fix m4 all property/global coefficientFriction peratomtypepair ${natoms} ${sf11} ${sf12} ${sf21} ${sf22}
fix m5 all property/global coefficientRollingFriction peratomtypepair ${natoms} ${rf11} ${rf12} ${rf21} ${rf22}
fix m6 all property/global cohesionEnergyDensity peratomtypepair ${natoms} ${ce11} ${ce12} ${ce21} ${ce22}

fix silo all mesh/surface file Silo.stl type 2 scale 0.001
fix walls all wall/gran model hertz tangential history cohesion sjkr mesh n_meshes 1 meshes silo
fix pts1 all particletemplate/sphere 10487 atom_type 1 density constant ${density} volume_limit 1e-15 radius constant ${radius1}
fix pdd1 all particledistribution/discrete 32452867 ${nradii} pts1 ${frac1}
fix ins_mesh all mesh/surface/planar file Insertionsface.stl type 1 scale 0.001
fix ins all insert/stream seed 86028157 distributiontemplate pdd1 &
mass ${fillmass} massrate ${fillmassrate} overlapcheck yes all_in yes vel constant 0 0 -0.75 &
insertion_face ins_mesh extrude_length 0.08

fix ultraSound all move/mesh mesh silo wiggle amplitude 0.001 0. 0. period 1.00E-04
run ${fillsteps}
unfix ultraSound
unfix ins

Best Regards,

Ege

egebat7 | Thu, 04/01/2021 - 16:12

Yes, particles are falling with no cohesion model.

I am trying different cohesion values to see how the simulation will react. For low cohesion values, particles are randomly changing positions on the angular wall, even with this I cannot remove them from the wall with proper vibrating frequencies. For extreme high vibrating frequencies, particles are exploding to everywhere in extreme high velocities. So I tried many different cohesion values with many different frequencies and amplitudes but couldn't end up with satisfying results, nothing is happening between particles still being stuck and exploding.

Okay, I will try to add a cohesion model to pair_style and hope to see some promising results.

Thanks,
Ege