T1/T2 Amplitudes & D1/D2 Diagnostics

Nathan DeYonker nate at hartree.ccqc.uga.edu
Mon Mar 15 19:42:56 GMT 2004

It was suggested a few years back in an email by Anthony Scott that the
CCSD T1 Diagnostic reported by Molpro may be incorrect for open-shell

Dylan Jayatilaka and Tim Lee derived a T1 diagnostic for open-shell cases
in JCP V98 p9734 1993, and since then, an improved D1/D2 diagnostic has
been created, published by Neilsen and Janssen in CPL V310 p568 1999 and
CPL V290 p423 1998 for closed-shell systems and by Leininger et al. for
open-shell systems in CPL V328 p431 2000.

Tim Lee has since shown a relationship (CPL v372 p362 2003) between the
different types of diagnostics and suggested that using both the D1 and T1
diagnostic is the most informative. 

In our lab, we've seen that the values of individual T1 and T2 amplitudes
themselves are useful in determining how well CCSD is performing.

Psi 3.2 and ACESII print individual amplitudes by default and Psi now
prints both the updated T1 and the D1 diagnostics. Is there any chance of
molpro updating in this area?

  Nathan DeYonker              | phone: 706 542-7737
  Ctr. for Comp. Chemistry     | fax:   706 542-0406
  University of Georgia        | e-mail: nate at ccqc.uga.edu
  Athens, GA  30602-2556       | http://www.ccc.uga.edu

More information about the Molpro-user mailing list