<div dir="ltr"><div>Dear all,</div><div><br></div><div>
<div>I am trying to relax a large structure (about 100 atoms) containing
 organic molecules. Since vc-relax using the BFGS algorithm takes a long time to converge, I want to relax the structure using the 
Car-Parrinello method (QE_6.7). I am wondering if 
the following procedure is correct:</div><div><br></div><div><b>1) electron relaxation:</b></div><div>&control<br>    calculation = 'cp'<br>    restart_mode = 'from_scratch'<br>    nstep=10000, iprint=10, isave=100,<br>    etot_conv_thr=1.d-7,<br>    ekin_conv_thr=1.d-7,<br>    dt=1.0,<br>    ndr=50, ndw=50,<br>/<br>&system<br>    ecutwfc = 40, ecutrho = 300,<br>    nr1b=20, nr2b=20, nr3b=20,<br>/<br>&electrons<br>    electron_dynamics='damp', electron_damping=0.05,<br>    emass=300,<br>    orthogonalization='ortho', ortho_eps=1.d-11, ortho_max=200,<br>    conv_thr = 1.0D-6,<br>/<br>&ions<br>    ion_dynamics = 'none'<br>/<br>&cell<br>    cell_dynamics = 'none'<br>/</div><div>......................................................<br></div><div><b><br></b></div><div><b>2) random displacement:</b></div><div>&control<br>    calculation = 'cp'<br>    restart_mode = 'restart'<br>    nstep=10000, iprint=10, isave=100,<br>    etot_conv_thr=1.d-7,<br>    ekin_conv_thr=1.d-7,<br>    dt=1.0,<br>    ndr=50, ndw=51,<br>/<br>&system<br>    ecutwfc = 40, ecutrho = 300,<br>    nr1b=20, nr2b=20, nr3b=20,<br>/<br>&electrons<br>    electron_dynamics='damp', electron_damping=0.05,<br>    emass=300,<br>    orthogonalization='ortho', ortho_eps=1.d-11, ortho_max=200,<br>    conv_thr = 1.0D-6,<br>/<br>&ions<br>    ion_dynamics = 'none'<br>    tranp = .true.<br>    amprp = 0.03,<br>/<br>&cell<br>    cell_dynamics = 'none'<br>/</div><div>
...................................................... <br></div><div><br></div><div><b>3) running vc-cp including ion_dinamics and cell_dynamics:</b></div><div>&control<br>    calculation = 'vc-cp'<br>    restart_mode = 'reset_counters'<br>    nstep=500000, iprint=10, isave=100,<br>    etot_conv_thr=1.d-7,<br>    ekin_conv_thr=1.d-7,<br>    dt=5.0,<br>    ndr=51, ndw=52,<br>/<br>&system<br>    ecutwfc = 30, ecutrho = 200,<br>    nr1b=20, nr2b=20, nr3b=20,<br>/<br>&electrons<br>    electron_dynamics = 'verlet'<br>    emass=700,<br>    orthogonalization='ortho',<br>    conv_thr = 1.0D-6,<br>/<br>&ions<br>    ion_dynamics = 'verlet'<br>    ion_velocities = 'zero'<br>/<br>&cell<br>    cell_dynamics = 'damp-pr'<br>    cell_dofree = 'xyz'<br>/<br></div><div><br></div><div>I
 ran three inputs based on the above steps, but the kinetic energy is 
rising. I ran an scf using pw.x with the positions/cell_parameters at 
0th, 1000th, 2000th and ... NFIs, but total_energy and total_force is 
rising.<br></div><div><br></div><div>Any help will be greatly appreciated.</div><div><br></div><div>Best regards,</div><div><br></div><div>Mohammad Moaddeli</div><div>ShirazU</div>

</div></div>