Hi,

 Forgot the "return" statement ?.
 R.


> Hi all
> I am trying to optimize a crystal with some restrictions, such alpha,
> beta, and gamma angles are constants. Moreover, I also want to fix the
> unit-cell side lengths to their initial values for b and c. This is my
> file constr.f:
>
>  subroutine constr( cell, na, isa, amass, xa, stress, fa, ntcon )
> c *****************************************************************
> c User-written routine to implement specific geometric constraints,
> c by orthogonalizing the forces and stress to undesired changes.
> c Arguments:
> c real*8  cell(3,3)    : input lattice vectors (Bohr)
> c integer na           : input number of atoms
> c integer isa(na)      : input species indexes
> c real*8  amass(na)    : input atomic masses
> c real*8  xa(3,na)     : input atomic cartesian coordinates (Bohr)
> c real*8  stress( 3,3) : input/output stress tensor (Ry/Bohr**3)
> c real*8  fa(3,na)     : input/output atomic forces (Ry/Bohr)
> c integer ntcon        : total number of positions constr. imposed
> c *****************************************************************
>       implicit         none
>       integer          na, isa(na), ntcon, ia
>       double precision amass(na), cell(3,3), fa(3,na),
>      .                 stress(3,3), xa(3,na)
>
>       double precision fav, stressav
>
> c Write here your problem-specific code.
>
>       stress(1,2) = 0.0d0
>       stress(1,3) = 0.0d0
>
>       stress(2,1) = 0.0d0
>       stress(2,2) = 0.0d0
>       stress(2,3) = 0.0d0
>
>       stress(3,1) = 0.0d0
>       stress(3,2) = 0.0d0
>       stress(3,3) = 0.0d0
>
>       end

Responder a