Re: How are normalized the AO's in the Seward output?

Posted by Vincent Brems on September 11, 2002 at 16:45:56:

In Reply to: Re: How are normalized the AO's in the Seward output? posted by Roland Lindh on September 11, 2002 at 13:41:06:

: The AO's in MOLCAS in case of real sphereical Gaussians are normalized to unity. For cartesian AO functions beyond p the normalization is not unity. The normalization of the primitive Gaussians is handled internally by Seward, i.e. the contraction coefficients listed by Seward are those of the input. The primitive Gaussians are renormalized internally since Seward work with unnormalized Gaussians.

: -roland

OK. So it means that the SO's are not normalized to unity.
Am I right?

I don't intend to account for cartesian functions above p.

Moreover I think 3d2+ and 3d2- correspond respectively
to the real and imaginary parts of the complex 3d2+
function. Is this correct?

i.e. I have programmed for the angular part of the
AO's of d type

3d0 = (3*z^2-r^2)/r^2 *sqrt(1/4)*sqrt(7/(4*pi))
3d1+ = -z*x /r^2 *sqrt(3/2)*sqrt(7/(4*pi))*sqrt(2)
3d1- = -z*y /r^2 *sqrt(3/2)*sqrt(7/(4*pi))*sqrt(2)
3d2+ = (x^2-y^2) /r^2 *sqrt(3/8)*sqrt(7/(4*pi))*sqrt(2)
3d2- = 2*x*y /r^2 *sqrt(3/8)*sqrt(7/(4*pi))*sqrt(2)

Is this correct? (I have some doubt about the signs)

Thank you.

Vincent Brems
University of Bonn
http://www.thch.uni-bonn.de/tc/people/brems.vincent/

Follow Ups:

Post a Followup

Name:
E-Mail:

Subject:

Comments:
: : The AO's in MOLCAS in case of real sphereical Gaussians are normalized to unity. For cartesian AO functions beyond p the normalization is not unity. The normalization of the primitive Gaussians is handled internally by Seward, i.e. the contraction coefficients listed by Seward are those of the input. The primitive Gaussians are renormalized internally since Seward work with unnormalized Gaussians. : : -roland : OK. So it means that the SO's are not normalized to unity. : Am I right? : I don't intend to account for cartesian functions above p. : Moreover I think 3d2+ and 3d2- correspond respectively : to the real and imaginary parts of the complex 3d2+ : function. Is this correct? : i.e. I have programmed for the angular part of the : AO's of d type : 3d0 = (3*z^2-r^2)/r^2 *sqrt(1/4)*sqrt(7/(4*pi)) : 3d1+ = -z*x /r^2 *sqrt(3/2)*sqrt(7/(4*pi))*sqrt(2) : 3d1- = -z*y /r^2 *sqrt(3/2)*sqrt(7/(4*pi))*sqrt(2) : 3d2+ = (x^2-y^2) /r^2 *sqrt(3/8)*sqrt(7/(4*pi))*sqrt(2) : 3d2- = 2*x*y /r^2 *sqrt(3/8)*sqrt(7/(4*pi))*sqrt(2) : Is this correct? (I have some doubt about the signs) : Thank you. : Vincent Brems : University of Bonn : http://www.thch.uni-bonn.de/tc/people/brems.vincent/