[lammps-users] The y-direction vibration is applied by moving wiggle, and there is X-direction displacement in the process of vibration propagation?

Hello everybody:
The sine vibration in Y direction is applied to an region by fix move wiggle, and then the vibration propagates along X direction. It is found that there is displacement in the X direction during the propagation. Why? Ensemble NVE, amplitude 0.1A, period 2ps! Thanks very much!

Cage

##test for Al
##################define variable################
variable a equal 1
variable T equal 0.01
##################Initialization################
units metal
atom_style atomic
boundary m p p
neighbor ${a} bin
neigh_modify delay 0
read_data 1k.data

##################region-group################
region source block INF 50 INF INF INF INF units box
group source region source
region backend block 180 INF INF INF INF INF units box
group backend region backend
region detector block 80 90 INF INF INF INF units box
group detector region detector
group boundary union source backend
group zone subtract all boundary

fix walls all wall/reflect xlo 2 xhi 200 units box
##################potential################
pair_style eam/alloy
pair_coeff * * Al_Zhou04.eam.alloy Al
##################output interatomic information################
thermo 100
thermo_style custom step lx ly lz xy xz yz pxx pyy pzz pxy pxz pyz pe ke temp
##################wave motion process(x)################
velocity source set 0 0 0
fix 1 source setforce 0 0 0
velocity backend set 0 0 0
fix 10 backend setforce 0 0 0

fix 2 source move wiggle 0.0 0.1 0.0 2 units box
fix 3 zone nve
timestep 0.001
dump 1 all custom 100 AL1.lammpstrj id type x y z
timestep 0.001
run 2000
unfix 2
undump 1

dump 2 all custom 100 AL2.lammpstrj id type x y z
timestep 0.001
run 3000
##################output strain and stress################
print "All done"

please see: https://matsci.org/t/the-y-direction-vibration-is-applied-by-moving-wiggle-and-there-is-x-direction-displacement-in-the-process-of-vibration-propagation/36817