From owner-chemistry@ccl.net Sun Nov 20 18:23:00 2005 From: "Yingbin Ge yingbin.ge**gmail.com" To: CCL Subject: CCL: Is CASSCF appropriate for study of PES? Message-Id: <-30013-051120182139-27291-n4bCnJaStMmo681Ftu9Naw:server.ccl.net> X-Original-From: Yingbin Ge Content-Type: multipart/alternative; boundary="----=_Part_11499_24677866.1132528892393" Date: Sun, 20 Nov 2005 17:21:32 -0600 MIME-Version: 1.0 Sent to CCL by: Yingbin Ge [yingbin.ge(_)gmail.com] ------=_Part_11499_24677866.1132528892393 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Thanks for your kind reply! I need to study all possible reaction pathways of a small molecular system for example: A --> B + C A --> D + E D + E --> F + G F + G --> H H --> A ... And then I need to get the relative energies of local minima and transition states (TS) of the above reactions. Different sigma bonds are broken and/or formed in the above reactions, whic= h makes it hard to choose the active space for every reaction consistently. However I need to get the relative energies right, at least within 2 kcal/mol accuracy. CCSD(T) or QCISD(T) does not do well for TS structures with streching bonds= . That's why I resort to the CAS methods. I plan to use the full valence active space, but still choosing the "correct" virtual orbitals is not easy especially fo= r transition states. What do you think? Thanks again, Yingbin On 11/20/05, errol lewars elewars%x%trentu.ca wrote: > > Sent to CCL by: errol lewars [elewars(!)trentu.ca] > 2005 Nov 20 > > If your reaction is A--> B size-consistency (SC) should not be a > problem. If it is A--> B + C, e.g. the energy of a dimer compared to the > separated monomers, you have to worry about SC. > > "No good way... virtual orbitals". CASSCF is not a model chemistry (see > e.g. Foresman and Frisch, "Exploring Chemistry with Electronic > Structutre Methods", so there in no _automatic_ way to choose the MOs of > the active space. But people still use CASSCF and get useful results. > > E. Lewars > =3D=3D=3D > > > Yingbin Ge yingbin.ge^gmail.com wrote: > > >Sent to CCL by: Yingbin Ge [yingbin.ge ]![gmail.com > ] > >------=3D_Part_4657_175608.1132432456569 > >Content-Type: text/plain; charset=3DISO-8859-1 > >Content-Transfer-Encoding: quoted-printable > >Content-Disposition: inline > > > >Hi CCLers, > > I am studying the PES of a small molecule and wondering > >whether CASSCF or CASPT2//CASSCF is an appropriate > >method for the PES study since these methods are not > >size consistent. Another downside of using CAS is that there > >is no good way of choosing virtual orbitals into the active space. > > Any comments? > > Thanks a lot! > >Yingbin > > > >------=3D_Part_4657_175608.1132432456569 > >Content-Type: text/html; charset=3DISO-8859-1 > >Content-Transfer-Encoding: quoted-printable > >Content-Disposition: inline > > > >
Hi CCLers,
> >
 
> >
I am studying the PES of a small molecule and wondering
> >
whether CASSCF or CASPT2//CASSCF is an appropriate
> >
method for the PES study since these methods are not
> >
size consistent. Another downside of using CAS is that > there > > > > >
is no good way of choosing virtual orbitals into the active > space. >v> > >
 
> >
Any comments?
> >
 
> >
Thanks a lot!
> >
Yingbin
> > > >------=3D_Part_4657_175608.1132432456569--> > > > > > > > > > > > > -=3D This is automatically added to each message by the mailing script = =3D-> > > > ------=_Part_11499_24677866.1132528892393 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline
Thanks for your kind reply!
I need to study all possible reaction pathways of a small molecular sy= stem for example:
A --> B + C
A --> D + E
D + E --> F + G
F + G --> H
H --> A
...
And then I need to get the relative energies of local minima and trans= ition states (TS)
of the above reactions.
Different sigma bonds are broken and/or formed in the above react= ions, which
makes it hard to choose the active space for every reaction = consistently.
However I need to get the relative energies right, at least within 2 k= cal/mol accuracy.
CCSD(T) or QCISD(T) does not do well for TS structures with streching = bonds.
That's why I resort to the CAS methods.  I plan to use the full v= alence active space,
but still choosing the "correct" virtual orbitals is not eas= y especially for transition states.
What do you think?
 
Thanks again,
Yingbin
 
On 11/20/05, errol lewars elewars%x%trentu.ca <owner-chemistry^ccl.net> wrote:
Sent to CCL by: errol lewars [el= ewars(!)trentu.ca]
2005 Nov 20

If your reaction is A--> B size= -consistency (SC) should not be a
problem. If it is A--> B + C, e.g. the energy of a dimer compared to= the
separated monomers, you have to worry about  SC.

&= quot;No good way... virtual orbitals". CASSCF is not a model chemistry= (see
e.g. Foresman and Frisch, "Exploring Chemistry with Electronic
Stru= ctutre Methods", so there in no _automatic_ way to choose the MOs ofthe active space. But people still use CASSCF and get useful results.

E. Lewars
=3D=3D=3D


Yingbin Ge yingbin.ge^gmail.com w= rote:

>Sent to CCL by: Yingbin Ge [= yingbin.ge]![gmail.com]
>------= =3D_Part_4657_175608.1132432456569
>Content-Type: text/plain; charset=3DISO-8859-1
>Content-Trans= fer-Encoding: quoted-printable
>Content-Disposition: inline
>>Hi CCLers,
> I am studying the PES of a small molecule and wond= ering
>whether CASSCF or CASPT2//CASSCF is an appropriate
>method fo= r the PES study since these methods are not
>size consistent. Another= downside of using CAS is that there
>is no good way of choosing virt= ual orbitals into the active space.
> Any comments?
> Thanks a lot!
>Yingbin
>
>= ------=3D_Part_4657_175608.1132432456569
>Content-Type: text/html; ch= arset=3DISO-8859-1
>Content-Transfer-Encoding: quoted-printable
&g= t;Content-Disposition: inline
>
><div>Hi CCLers,</div>
><div>&nb= sp;</div>
><div>I am studying the PES of a small molecule= and wondering</div>
><div>whether CASSCF or CASPT2//CASS= CF is an appropriate</div>
><div>method for the PES study since these methods are not<= /div>
><div>size consistent. Another downside of using CAS&a= mp;nbsp;is that there</div=3D
>
>
><div>is no go= od way of choosing virtual orbitals into the active space.</di=3D
>v>
><div>&nbsp;</div>
><div>An= y comments?</div>
><div>&nbsp;</div>
><= ;div>Thanks a lot!</div>
><div>Yingbin</div>
>
>------=3D_Part_4657_175608.1132432456569-->
>
&= gt;
>
>



-=3D This is automatically added to each= message by the mailing script =3D-
To recover the email address of the = author of the message, please change l= ook up the X-Original-From: line in the mail header.

E-mail to subsc= ribers: CHEMISTRY^ccl.net or use:
     http://www.ccl.net/cgi-bin/ccl/send_ccl_message

E-= mail to administrators: CHEMIS= TRY-REQUEST^ccl.net or use
     http://www.ccl.net/cgi-bin/ccl/send_ccl_message
Subscribe/Unsubscribe:
     http://www.ccl.net/chemistry/sub_unsu= b.shtml

Before posting check wait time for next message at: http://www.ccl.net

Job advertisements: http://www.ccl.net/jobs

If your m= ail bounces from CCL with=20 5.7.1 error, check:
     http://www.ccl.net/spammers.txt

-+-+-+-+-+-+-+-= +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+




------=_Part_11499_24677866.1132528892393--