atoms lost when atom size is less than 0.001m

Hi,

I have a problem when trying to model atoms which are 100um in diameter. When the atom size is 100um in diameter, the atoms keep lost during the process, and it won't happen when atom size is larger than 0.001m. Could anyone explain why it happen and how to fix it? I tried to change the units to cgs, but it did not work...

atom_style granular
atom_modify map array
boundary f f f
newton off

communicate single vel yes

units si

region reg block -0.05 0.05 -0.05 0.05 0. 0.15 units box
create_box 1 reg

neighbor 0.05 bin
neigh_modify delay 0

#Material properties required for new pair styles

fix m1 all property/global youngsModulus peratomtype 5.e6
fix m2 all property/global poissonsRatio peratomtype 0.45
fix m3 all property/global coefficientRestitution peratomtypepair 1 0.051
fix m4 all property/global coefficientFriction peratomtypepair 1 0.05
fix m5 all property/global characteristicVelocity scalar 2.
fix m6 all property/global cohesionEnergyDensity peratomtypepair 1 500000

#New pair style
pair_style gran model hertz tangential history cohesion sjkr #Hertzian with cohesion
pair_coeff * *

timestep 0.00001

fix gravi all gravity 9.81 vector 0.0 0.0 -1.0

fix zwalls1 all wall/gran model hertz tangential history primitive type 1 zplane 0.0
fix zwalls2 all wall/gran model hertz tangential history primitive type 1 zplane 0.15
fix cylwalls all wall/gran model hertz tangential history primitive type 1 zcylinder 0.05 0. 0.

#region for insertion
region bc cylinder z 0.01 0.01 0.025 0.05 0.0603 units box

group nve_group region reg

#distributions for insertion
fix pts1 all particletemplate/sphere 15485863 atom_type 1 density constant 2500 radius constant 0.00005
fix pdd1 all particledistribution/discrete 15485867 1 pts1 1.0

#particle insertion
fix ins nve_group insert/pack seed 32452843 distributiontemplate pdd1 vel constant 0.0 0.0 -0.002 &
maxattempt 1000 insert_every 1000 overlapcheck yes all_in yes &
region bc particles_in_region 250 ntry_mc 10000

#apply nve integration to all particles that are inserted as single particles
fix integr nve_group nve/sphere

#output settings, include total thermal energy
compute rke all erotate/sphere
compute str all stress/atom
thermo_style custom step atoms ke c_rke vol
thermo 1000
thermo_modify lost ignore norm no

#insert the first particles so that dump is not empty
run 1
dump dmp all custom/vtk 1000 post/cohesion_*.vtk id type type x y z ix iy iz vx vy vz fx fy fz omegax omegay omegaz radius c_str[3]

#insert particles
run 5000
unfix ins

#run
run 30000 upto

Thanks a lot!