Tensil test and boundary conditions

i am beginner in lammps i see that lammps offer many options doing tensil test : set velocity add force or fix move , in my test i fixed one end of the simulation cell and used fix move linear but my brittle material somehow become ductile and it took long time and very high strain to visualize the fracture so my question is what fix is recommanded to do a tensil test in vacuum or periodic boundary on the axis of deformation, does fix erate with combinaison with nve and time/rescale to controle the temperature is okay , or it’s better with nvt since i can’t control the pressure.

The choice of thermostat and method (ie either imposed stress or imposed displacement) should NOT influence the result. Therefore here I would say that your best choice is to test several combinations of methods / thermostats and ensure that you always recover the same result. On the side note, temp/rescale is known to be a bad choice, as discussed many time in this forum :