Pure shear fracture of Silicon (Mode II)

Dear Lammps Users,

I am studying mode II fracture of Pure silicon using Lammps. I have setup my input file similar to the example given in examples/shear/ directory. The sample consists of an initial edge type v-notch crack of length 10 nm and the crack plane is (010) and crack propagation direction is [100]. Silicon is known to fracture in a brittle way at low temperatures. However, I do not see any crack propagation in the horizontal direction (please see the attached snapshots). The Shear stress (sigma-xy) also stabilizes at a non-zero value after initially increasing to a maximum value and then decreasing (please see the attached plot). Below is my input file:

units metal
boundary s s p

atom_style atomic
read_data notch2.crack.data

region lower block 0.0 543.0 0.0 16.29 0.0 27.15
region upper block 0.0 543.0 152.04 168.33 0.0 27.15
group lower region lower
group upper region upper
group boundary union lower upper
group mobile subtract all boundary

pair_style sw
pair_coeff * * Si.sw Si

temp controllers

compute new3d mobile temp
compute new2d mobile temp/partial 0 1 1
compute peratom all stress/atom

dump 1 all custom 5000 dump1.shear.notch id type x y z c_peratom[1] c_peratom[2] c_peratom[3] c_peratom[4] c_peratom[5] c_peratom[6]

velocity mobile create 50.0 5812775 temp new3d
fix 2 all nve
fix 3 boundary setforce 0.0 0.0 0.0

fix 4 mobile temp/rescale 1 50.0 50.0 2.0 1.0
fix_modify 4 temp new3d

thermo 100
thermo_style custom etotal vol pe ke temp press pxx pyy pzz pxy pxz pyz lx ly lz xy xz yz
thermo_modify line multi flush yes temp new3d

timestep 0.001
run 10000

shear

velocity upper set 1 0 0
velocity mobile ramp vx 0.0 1.0 y 16.29 152.04 sum yes

unfix 4
fix 4 mobile temp/rescale 1 50.0 50.0 2.0 1.0
fix_modify 4 temp new2d

thermo_style custom etotal vol pe ke temp press pxx pyy pzz pxy pxz pyz lx ly lz xy xz yz
thermo_modify line multi flush yes temp new2d
thermo 100

reset_timestep 0
run 50000

I expected to see crack propagation in the horizontal plane ultimately leading to failure (when sigma-xy goes back to zero). I am also simulating Mode I fracture where I see complete fracture along the expected lines. Can anyone tell me if I am missing something in my simulation setup that is leading to this result? I already tried longer initial notches but I still see the same thing. My initial system size is about 100 x 31 x 5 unitcells long in x, y and z directions.

regards,
Dr. Karthik Guda

Synopsys

stress_vs_time.png

shear0000.png

shear0005.png

shear0008.png

shear0010.png

shear0012.png

shear0015.png