From owner-chemistry@ccl.net Mon Sep 19 15:32:00 2011 From: "Wang, Yixuan yixuan.wang:-:asurams.edu" To: CCL Subject: CCL: Pure DFT functionals and ORCA Message-Id: <-45481-110919152400-13889-HPX2Rwt9q5nYQeRkWcA16Q[A]server.ccl.net> X-Original-From: "Wang, Yixuan" Content-Language: en-US Content-Type: multipart/alternative; boundary="_000_FDCE932536422147B5592DDC55D653842B8B19E3EXCH02asuramsed_" Date: Mon, 19 Sep 2011 19:23:45 +0000 MIME-Version: 1.0 Sent to CCL by: "Wang, Yixuan" [yixuan.wang|*|asurams.edu] --_000_FDCE932536422147B5592DDC55D653842B8B19E3EXCH02asuramsed_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable I did not well follow your question and answers. If you have not tried Slow= Conv, may try; otherwise need to adjust your initial geometry. Good luck. Yixuan Wang, Ph.D Associate Professor Department of Natural Science Albany State University Albany, GA 31705 Tel: 229-430-7843 > From: owner-chemistry+yixuan.wang=3D=3Dasurams.edu!^!ccl.net [mailto:owner-ch= emistry+yixuan.wang=3D=3Dasurams.edu!^!ccl.net] On Behalf Of Jason D Acchioli= jdacchio:gmail.com Sent: Monday, September 19, 2011 12:26 PM To: Wang, Yixuan Subject: CCL: Pure DFT functionals and ORCA Hi all, I tried both NRSCF and AHSCF in ORCA, but I still have SCF prob= lems. I'll try NWChem when I get the chance, but I don't see why this would= n't work in ORCA. Any more thoughts? Jason On Sep 16, 2011, at 2:10 PM, Olexandr Isayev olexandr.isayev#case.edu wrote= : Jason: DIIS is usually robust enough for most "normal" molecules. Perhaps, yours is unlucky one, please try NRSCF or AHSCF instead. Hopefully, this will help, Olexandr ________________________________ Olexandr Isayev, Ph.D. Department of Chemistry Case Western Reserve University 10900 Euclid Avenue Cleveland, OH 44106-7078 USA Phone: 769 218-9812 Fax: 216 368-3006 http://olexandrisayev.com On Fri, Sep 16, 2011 at 10:25, Jason D Acchioli jdacchio[a]gmail.com > = wrote: Diego, Frank, and John, Many thanks for your suggestions. I tried Frank's suggestion, and things we= re going smoothly until about SCF cycle 25, where the wild oscillations aga= in took over. I used the VerySlowConv keyword as well. Any more thoughts? Jason On Sep 15, 2011, at 3:38 AM, Frank Neese neese#%#mpi-muelheim.mpg.de wrote: Dear Diego, That is - in principle - correct, But what Jason should have done is to specify it like: ! RKS PW91 ZORA def2-SVP def2-SVP/J TightSCF PrintMOs Grid5 The "ZORA" in the input line is important so that ORCA uses the relativisti= cally recontracted all electron basis sets. Of course, alternatively one co= uld use ECPs. The rest of the %rel block is just duplicating defaults and i= s not necessary. Good luck, FN Am 15.09.2011 um 09:25 schrieb DIEGOI GOMEZ darkego21(-)yahoo.com: Hello Jason.. I'm missing the ECP for the Pd atoms in the input file you provide us. To my knowledge, the def2-TZVP is not an all electron basis set, then you w= ould have to add in your input something like 'ECP{def2-TZVP=3DPd}' to us= e in ORCA the ECP that the program uses for this atom by default. Without t= he ECP you can get a result but probably (if not surely) it will be wrong! Regards.. Diego. Hi all, I posted the following question on the ORCA forum, but haven't heard anythi= ng (yet), so I'll throw it out here. The output file is below the message. Jason Hi all, I'm trying to run a single point / TDDFT calc on a metal-organic system. Ho= wever, the SCF does not converge after 125 cycles. I've tried the PBE, RPBE= , and revPBE XC functionals, and the same thing happens-I get an oscillatin= g SCF cycle. The same thing happens when I try the hybrid B3LYP/G functiona= l and the RIJCOSX approximation. However, if I do a B3LYP/G without the RIJ= COSX, convergence is fine. What am I doing wrong? If it is a problem with RI, can I still use any of P= W91 or PBE XC functionals (which, I believe, require auxiliary basis sets?)= ? The input is attached as a zip file. Thanks, Jason !RKS PW91 def2-SVP def2-SVP/J TightSCF PrintMOs Grid5 %output Print[P_OrbPopMO_M] 1 Print[P_FragPopMO_M] 1 end %rel method ZORA #or IORA modelpot 1,1,1,1 modeldens rhoZORA velit 137.0359895 #speed of light used end %tddft nroots 15 maxdim 64 end ! PAL4 * xyz 0 1 H(1) -4.56675 -2.48229 -0.00028 C(1) -4.57359 -1.39699 -0.00028 C(1) -5.77738 -0.69952 -0.00035 H(1) -6.71624 -1.24540 -0.00040 C(1) -5.77739 0.69938 -0.00034 H(1) -6.71628 1.24523 -0.00039 C(1) -4.57363 1.39688 -0.00028 H(1) -4.56682 2.48218 -0.00029 C(1) -3.34414 0.71339 -0.00022 C(1) -2.12293 1.43662 -0.00018 C(1) -3.34412 -0.71346 -0.00021 C(1) -2.12289 -1.43666 -0.00014 C(1) -1.05872 2.02393 -0.00013 C(1) 0.16183 2.74774 -0.00010 C(1) -1.05866 -2.02394 -0.00006 C(1) 0.16190 -2.74772 -0.00004 C(1) 0.16567 4.15310 -0.00038 C(1) 0.16578 -4.15308 -0.00027 H(1) -0.78318 -4.67691 -0.00043 C(1) 1.37444 -4.83715 -0.00028 H(1) -0.78331 4.67691 -0.00066 C(1) 1.37431 4.83720 -0.00031 C(1) 2.56419 4.10860 0.00002 H(1) 1.38849 5.92301 -0.00055 C(1) 2.49611 2.71947 0.00026 H(1) 3.53196 4.59786 0.00008 N(1) 1.33136 2.05360 0.00019 H(1) 3.38890 2.10605 0.00051 C(1) 2.56430 -4.10851 -0.00008 H(1) 1.38865 -5.92296 -0.00045 C(1) 2.49618 -2.71939 0.00013 H(1) 3.53208 -4.59774 -0.00008 N(1) 1.33141 -2.05354 0.00016 H(1) 3.38895 -2.10595 0.00028 Pd(2) 1.33817 -0.00000 0.00023 newgto "def2-TZVP" "def= 2-TZVP/J" end Cl(3) 1.43325 -0.00001 -2.35531 Cl(3) 1.43091 0.00002 2.35584 * ******************************************* Jason D'Acchioli Assistant Professor of Chemistry University of Wisconsin-Stevens Point 2001 Fourth Avenue Stevens Point, WI 54481 http://chemdac.uwsp.edu ******************************************* Jason D'Acchioli Assistant Professor of Chemistry University of Wisconsin-Stevens Point 2001 Fourth Avenue Stevens Point, WI 54481 http://chemdac.uwsp.edu ******************************************* Jason D'Acchioli Assistant Professor of Chemistry University of Wisconsin-Stevens Point 2001 Fourth Avenue Stevens Point, WI 54481 http://chemdac.uwsp.edu ******************************************* Jason D'Acchioli Assistant Professor of Chemistry University of Wisconsin-Stevens Point 2001 Fourth Avenue Stevens Point, WI 54481 http://chemdac.uwsp.edu --_000_FDCE932536422147B5592DDC55D653842B8B19E3EXCH02asuramsed_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

I did not well follow your question and answers. If you have not tried S= lowConv, may try; otherwise need to adjust your initial geometry.

Good luck.

 <= /p>

Yixuan Wang, Ph.D

Assoc= iate Professor

Depar= tment of Natural Science

Alban= y State University

Alban= y, GA 31705

Tel: = 229-430-7843

 <= /p>

From: owner-ch= emistry+yixuan.wang=3D=3Dasurams.edu!^!ccl.net [mailto:owner-chemistry= 3;yixuan.wang=3D=3Dasurams.edu!^!ccl.net] On Behalf Of Jason D Acchioli jdacchio:gmail.com
Sent: Monday, September 19, 2011 12:26 PM
To: Wang, Yixuan
Subject: CCL: Pure DFT functionals and ORCA

 

Hi all,

 

   &nb= sp;        I tried both NRSCF and= AHSCF in ORCA, but I still have SCF problems. I'll try NWChem when I get t= he chance, but I don't see why this wouldn't work in ORCA. Any more thought= s? 

 

Jason

 

 

On Sep 16, 2011, at 2:10 PM, Olexandr Isayev olexand= r.isayev#case.edu wrote:



Jason:

DIIS is usually robust enough for most "normal" molecules.
Perhaps, yours is unlucky one, please try NRSCF or AHSCF instead.


Hopefully, this will help,
Olexandr


________________________________
Olexandr Isayev, Ph.D.
 
Department of Chemistry
Case Western Reserve University
10900 Euclid Avenue
Cleveland, OH 44106-7078 USA

Phone:   769 218-9812
Fax:       216 368-3006

http://olexandrisa= yev.com


On Fri, Sep 16, 2011 at 10:25, Jason D Acchioli jdac= chio[a]gmail.com <owner-chemistry.:.ccl.net> wrote:

Diego, Frank, and John,

 

Many thanks for your suggestions. I tried Frank's su= ggestion, and things were going smoothly until about SCF cycle 25, where th= e wild oscillations again took over. I used the VerySlowConv keyword as wel= l. Any more thoughts?

 

Jason

 

On Sep 15, 2011, at 3:38 AM, Frank Neese neese#%#mpi-muelheim.mpg.de<= /a> wrote:



Dear Diego, 

 

That is - in principle - correct, 

 

But what Jason should have done is to specify it lik= e: 

 

! RKS PW91 ZORA def2-SVP def2-SVP/J TightSCF PrintMO= s Grid5

 

The "ZORA" in the input line is important = so that ORCA uses the relativistically recontracted all electron basis sets= . Of course, alternatively one could use ECPs. The rest of the %rel block i= s just duplicating defaults and is not necessary. 

 

Good luck,

FN

 

 

 

 



Hello Jason..<= /o:p>

 

I'm missing the ECP= for the Pd atoms in the input file you provide us.

 

To my knowledge, th= e def2-TZVP is not an all electron basis set, then you would have to add in= your input  something like  'ECP{def2-TZVP=3DPd}' to use in ORCA the ECP that the program uses for this atom by default. Wit= hout the ECP you can get a result but probably (if not surely) it will be w= rong!

 

Regards..

 

Diego.

 

Hi all,

 

I posted the following question on the ORCA forum, but haven't heard any= thing (yet), so I'll throw it out here. The output file is below the messag= e. 

 

Jason

 

Hi all, 

I'm trying to run a single point / TDDFT calc on a metal-organic system. Ho= wever, the SCF does not converge after 125 cycles. I've tried the PBE, RPBE= , and revPBE XC functionals, and the same thing happens-I get an oscillatin= g SCF cycle. The same thing happens when I try the hybrid B3LYP/G functional and the RIJCOSX approximation. Ho= wever, if I do a B3LYP/G without the RIJCOSX, convergence is fine.  What am I doing wrong? If it is a problem with RI, can I still use any of P= W91 or PBE XC functionals (which, I believe, require auxiliary basis sets?)= ? The input is attached as a zip file. 

Thanks, 

Jason

 

!RKS PW91 def2-SVP def2-SVP/J TightSCF PrintMOs Grid5

%output

Print[P_OrbPopMO_M] 1

Print[P_FragPopMO_M] 1

end=

%rel method ZORA #or IORA

modelpot 1,1,1,1

modeldens rhoZORA

velit 137.0359895 #speed of light used

end=

%tddft nroots 15

maxdim 64

end=

! PAL4

* xyz 0 1

H(1)         -4.56675    =   -2.48229       -0.00028

C(1)         -4.57359    =   -1.39699       -0.00028

C(1)         -5.77738    =   -0.69952       -0.00035

H(1)         -6.71624    =   -1.24540       -0.00040

C(1)         -5.77739    =    0.69938       -0.00034

H(1)         -6.71628    =    1.24523       -0.00039

C(1)         -4.57363    =    1.39688       -0.00028

H(1)         -4.56682    =    2.48218       -0.00029

C(1)         -3.34414    =    0.71339       -0.00022

C(1)         -2.12293    =    1.43662       -0.00018

C(1)         -3.34412    =   -0.71346       -0.00021

C(1)         -2.12289    =   -1.43666       -0.00014

C(1)         -1.05872    =    2.02393       -0.00013

C(1)          0.16183   &= nbsp;    2.74774       -0.00010

C(1)         -1.05866    =   -2.02394       -0.00006

C(1)          0.16190   &= nbsp;   -2.74772       -0.00004

C(1)          0.16567   &= nbsp;    4.15310       -0.00038

C(1)          0.16578   &= nbsp;   -4.15308       -0.00027

H(1)         -0.78318    =   -4.67691       -0.00043

C(1)          1.37444   &= nbsp;   -4.83715       -0.00028

H(1)         -0.78331    =    4.67691       -0.00066

C(1)         1.37431     =    4.83720       -0.00031

C(1)          2.56419   &= nbsp;    4.10860        0.00002

H(1)          1.38849   &= nbsp;    5.92301       -0.00055

C(1)          2.49611   &= nbsp;    2.71947        0.00026

H(1)          3.53196   &= nbsp;    4.59786        0.00008

N(1)          1.33136   &= nbsp;    2.05360        0.00019

H(1)          3.38890   &= nbsp;    2.10605        0.00051

C(1)          2.56430   &= nbsp;   -4.10851       -0.00008

H(1)          1.38865   &= nbsp;   -5.92296       -0.00045

C(1)          2.49618   &= nbsp;   -2.71939        0.00013

H(1)          3.53208   &= nbsp;   -4.59774       -0.00008

N(1)          1.33141   &= nbsp;   -2.05354        0.00016

H(1)          3.38895   &= nbsp;   -2.10595        0.00028

Pd(2)         1.33817    =   -0.00000        0.00023 newgto "def2-TZVP&= quot; "def2-TZVP/J" end

Cl(3)         1.43325    =   -0.00001       -2.35531

Cl(3)         1.43091    =    0.00002        2.35584

*

 

 

*******************************************
Jason D'Acchioli
Assistant Professor of Chemistry
University of Wisconsin-Stevens Point
2001 Fourth Avenue
Stevens Point, WI 54481
http://chemdac.uwsp.= edu

 

 

**= *****************************************
Jason D'Acchioli
Assistant Professor of Chemistry
University of Wisconsin-Stevens Point
2001 Fourth Avenue
Stevens Point, WI 54481
http://chemdac.uwsp.= edu


 

 

 

 

*****************************= **************
Jason D'Acchioli
Assistant Professor of Chemistry
University of Wisconsin-Stevens Point
2001 Fourth Avenue
Stevens Point, WI 54481
http://chemdac.uwsp.= edu


 

 

 

*******************************************
Jason D'Acchioli
Assistant Professor of Chemistry University of Wisconsin-Stevens Point
2001 Fourth Avenue
Stevens Point, WI 54481
http:/= /chemdac.uwsp.edu


 

--_000_FDCE932536422147B5592DDC55D653842B8B19E3EXCH02asuramsed_--