CCL: Failure TS opt caused by methyl rotation

 Sent to CCL by: "Ol  Ga" [eurisco1 .. pochta.ru]
 Dear Zheng Xu !
 I intend that correct route sections are
 Step 1.
 #p ub3lyp/6-31++g(d,p) pOpt(qst2,noeigen,ModRedun)
 # scf(xqc,vshift,tight,Maxcycle=512)
 and freeze some (possible all) H-atoms to forbid rotation CH3-groups.
 You need to find out numbers of H-atoms in Z-mat and type in the job file
 number_H-atom_1    F
 number_H-atom_2    F
 ................
 Step 2.
 #p ub3lyp/6-31++g(d,p) opt(calcall,ts,noeigen,gdiis,tight) guess=always
 # scf(xqc,tight,Maxcycle=512)
 Possible a lot of trials is necessary to achieve a saddle point if it exists  on
 the level of theory
 Sincerely
 Ol Ga
 ----- Original Message -----
 > From: "Zheng Xu zjuxuzheng###hotmail.com"
 <owner-chemistry!A!ccl.net>
 To: "Ga, Ol " <eurisco1!A!pochta.ru>
 Sent: Tuesday, July 08, 2008 6:19 AM
 Subject: CCL: Failure TS opt caused by methyl rotation
 >
 >
 > Sent to CCL by: "Zheng  Xu" [zjuxuzheng|-|hotmail.com]
 > Dear CCLers,
 >
 > I'm trying to get the TS of dimethylsilylene (Me2Si) insertion into Si-Cl
 bond of SiH2FCl.
 >
 > I have tried many times with different opt keywords(gdiis,tight or
 verytight ),  howerver alll attempt was a failure. From the plot of "total
 energy" vs. "optimization step number", an unregular energy
 perturbation aroused by the rotation of two methyl groups was detected.
 >
 > Howerver, when ths substrate was changed from SiH2FCl to SiH3Cl, a right TS
 can be obtained.
 > Why? What should I do, is there anyone have good suggestion?
 >
 > ---------------------------------------------------------------------------
 > #p rb3lyp/6-31++g(d,p) opt(calcfc,ts,noeigen,gdiis) freq pop(regular,nbo)
 test
 > ---------------------------------------------------------------------------
 >