CCL:G: Gaussian error/basis sets

 Sent to CCL by: "Ol  Ga" [eurisco1_+_pochta.ru]
 Dear sachin  Tyagi,
 About your input :  you should use keyword pseudo=read in addition of GEN. I
 made some threads in CCL about this subject, e.g.
 http://www.ccl.net/chemistry/resources/messages/2008/07/14.004-dir/index.html
 http://www.ccl.net/cgi-bin/ccl/message-new?2008+07+14+007+raw
 and, please, dont forget to type 6d 10f (explicit  definitions are more
 reliable) if you use  6-31G* basis set.
 Obviously, you can find in CCL many other messages of CCLers with correct route
 cards in similar cases.
 What ECP + basis set(s) are you better to use? It is difficult to answer. But
 there are general suggestions: dont forget polarization functions; triple
 exponential basis sets are more reliable than bi-exponential basis sets;
 possible, ECP (small core) from Stuttgart-Dresden group in more reliable than
 ECP from Los-Alamos.
 Sincerely,
 Ol Ga
 > Sent to CCL by: "sachin  Tyagi" [styagi2,uic.edu]
 > Could any body please tell me how to assign different basis sets
 > for different atoms. I read in few books and checked some sample
 > input file and I tried using Keyword GEN and at the end of molecule
 > specifications I assigned the basis sets, (6-31G, LanL2DZ). But I am
 getting the following error:
 > General basis read from cards:  (5D, 7F)
 >   WANTED AN INTEGER AS INPUT.
 >   FOUND A FLOATING POINT NUMBER AS INPUT.
 >   1 2 1.0 5 2.0 22 1.0
 >
 >
 >
 >      ?
 >  Error termination via Lnk1e in /usr/common/g03/l301.exe at Fri Nov 21
 > It would be a great help if you could suggest me what basis sets
 > should I be using for Iodine and transition metal atoms (Rhodium and
 Copper).
 > Thanks
 > Regards
 > Sachin