Molcas Forum
Support and discussions for Molcas and OpenMolcas users and developers
You are not logged in.
- Topics: Active | Unanswered
Announcement
Please note: The forum's URL has changed. The new URL is: https://molcasforum.univie.ac.at. Please update your bookmarks!
You can choose an avatar and change the default style by going to "Profile" → "Personality" or "Display".Pages: 1
#1 2016-11-23 22:07:04
- debu_molcas
- Member
- Registered: 2016-11-13
- Posts: 9
Several questions
Thank you very much for the help.
In my last question "How to fix the direction of axis throughout the calculation?" I am not asking for constrained optimization. I need to fix the axis. Suppose if I use the following coordinate
C 0.00000000 0.00000000 0.00000000
C 0.00000000 0.00000000 1.51201280
C 1.20353907 0.00000000 -0.72069363
C -1.20323480 -0.02706114 -0.72069363
C -1.20596641 -0.05975407 -2.11770233
C 0.00071671 -0.06374287 -2.82256196
C 1.20700507 -0.03262324 -2.11770233
H 2.14799623 0.02815333 -0.18188142
H -2.14808621 -0.02015071 -0.18188142
H -2.15040090 -0.07836066 -2.65487304
H 0.00097063 -0.08632567 -3.90861443
H 2.15161917 -0.02998986 -2.65487304
H 0.01151829 -1.02441836 1.90588964
H 0.88007116 0.51324811 1.91242929
H -0.89138887 0.49333027 1.91242929It defines that the Z-axis is orienting through C1-C2 bond. Only this axis need to fix through that bond during the optimization but not any coordinate.
I have tried to compute a TS using RP-Coor input. Here is my input
&GATEWAY
RP-Coor
15
E1=-271.5344786
C -0.01931600 0.02509300 0.12681800
C -0.04157000 -0.17192300 1.46051500
C 1.28395700 0.03458900 -0.66480400
C -1.24853900 0.26480700 -0.62793000
C -1.27540400 0.20865200 -1.98025600
C -0.07279900 -0.10174200 -2.75063700
C 1.12236100 -0.21518600 -2.14386300
H 1.75361900 1.02655700 -0.54760200
H -2.16322400 0.43539400 -0.06529700
H -2.21114000 0.35665400 -2.51227800
H -0.16361100 -0.24383500 -3.82406800
H 2.01831200 -0.43873800 -2.71774300
H 0.86186700 -0.38929000 2.02334200
H -0.96987600 -0.11684300 2.02217000
H 1.99837300 -0.67417200 -0.22952600
E2=-271.5892346
C 0.00000000 0.01656200 0.01240000
C 0.00000000 -0.00286400 1.52428800
C 1.20346300 0.01229100 -0.70840800
C -1.20346300 0.01229100 -0.70840800
C -1.20656200 -0.00241800 -2.10572100
C 0.00000000 -0.01091600 -2.81074800
C 1.20656200 -0.00241800 -2.10572100
H -0.88578600 0.49527100 1.93113800
H -2.14817700 0.02290000 -0.16941500
H -2.15114600 -0.00350300 -2.64295000
H 0.00000000 -0.01954500 -3.89700100
H 2.15114600 -0.00350300 -2.64295000
H 0.00000000 -1.03232300 1.90497000
H 0.88578600 0.49527100 1.93113800
H 2.14817700 0.02290000 -0.16941500
Basis=C.6-31+G**.., H.6-31+G**..,
Title=single point in Hartree Fock
Group=Full
End of input
>>>Do While
&SEWARD
Title=Seward data
End of input
&SCF
Title=SCF data
KSDFT=B3LYP
End of input
&ALASKA
&SLAPAF
>>>EndDo
&MCKINLEY
&GRID_IT
ASCII
ALL
End of inputUnfortunately, It is not running. Could you please tell me what is the problem?
Could you please help me little bit to understand the CASSCF output. In the output the following sections are printed
printout of CI-coefficients larger than 0.05 for root 1
energy= -269.797777
conf/sym 11111111 Coeff Weight
1 22220000 -0.92488 0.85540
18 22202000 0.16315 0.02662
20 22200200 0.06319 0.00399
29 22udud00 -0.15252 0.02326
32 22ud0ud0 -0.06908 0.00477
40 2u2d0200 -0.05647 0.00319
41 2u2du0d0 -0.14195 0.02015
58 22022000 0.06735 0.00454
60 22020200 0.09042 0.00818
61 2202u0d0 0.05568 0.00310
69 2ud2ud00 0.07627 0.00582
72 2ud20ud0 0.08021 0.00643
88 20222000 0.06116 0.00374
93 20220020 0.05030 0.00253
117 02220002 0.08315 0.00691Please let me know the significance of Coeff and Weight? When we can able to say the system possess sufficient multi reference character?
I have one more question. I
I am sorry for the long E-mail. Your help will be highly appreciated. have started a CASSCF Single point calculation from the previous DFT optimization geometry and guess " FileOrb=opt.ScfOrb" after defining the active orbital. Now the CASSCF single point produces two orbital file ".RasOrb" and "RasOrb.1". Which orbital we need to call for further calculation e.g., CASSCF optimization?
Thanks & Regards
Debasish
Offline
#2 2016-11-24 09:39:25
- Ignacio
- Administrator
- From: Uppsala
- Registered: 2015-11-03
- Posts: 1,085
Re: Several questions
Here you have three different questions that are unrelated to the original thread. Please stick to the topic of the thread and create new ones when necessary.
It defines that the Z-axis is orienting through C1-C2 bond. Only this axis need to fix through that bond during the optimization but not any coordinate.
It still can be done through constraints (and I don't know of any other way), just set x(C1)=x(C2), y(C1)=y(C2), so they can only differ in the z coordinate, i.e., their bond is along the z axis. You'd do this with (untested):
Constraints
x1 = Cartesian x C1
x2 = Cartesian x C2
y1 = Cartesian y C1
y2 = Cartesian y C2
Values
1.0 x1 - 1.0 x2 = 0.0
1.0 y1 - 1.0 y2 = 0.0
End of ConstraintsYou may need to add VarR to &GATEWAY too.
I have tried to compute a TS using RP-Coor input. [...] Unfortunately, It is not running. Could you please tell me what is the problem?
The manual says: "Note that this option will overwrite the coordinates which have already been specified with the normal input of the molecular geometry. However, this does not make that input section redundant and should always be included." This means that you must still specify a Coord keyword, as in a normal calculation. Other than that, "it is not running" is not very helpful to say what's wrong.
Please let me know the significance of Coeff and Weight? When we can able to say the system possess sufficient multi reference character?
The CASSCF wavefunction is a linear combination of CSFs (each of which is a fixed linear combination of Slater determinants). Coeff are just the coefficients of the corresponding CSFs in the wavefunction (where the 222ud000 numbers indicate occupation and spin coupling of the active orbitals), and Weight are the squares of the coefficients (the sum of weights should be 1).
As for the multi-reference character, it may be related to the weight of the most significant configuration. If it's high, you can say it's almost single-reference, if it's low, it depends on the orbitals: if the orbitals were optimized for that particular root, you may say it's multi-reference, if they were optimized for other root or for a state-average, it could still be single-reference with the proper orbitals, or maybe not. What is "high" or "low", I can't say.
Now the CASSCF single point produces two orbital file ".RasOrb" and "RasOrb.1". Which orbital we need to call for further calculation e.g., CASSCF optimization?
In general, .RasOrb
Offline
#3 2016-11-28 14:20:04
- debu_molcas
- Member
- Registered: 2016-11-13
- Posts: 9
Re: Several questions
Thank you for your reply. I have tried to prepare the RP-Coordinate transition state calculation input according to the manual. Please see below the manual and my comments are in bracket.
"This activates the Saddle method for TS geometry optimization. The line (RP-Coordinate) is followed by an integer specifying the number of symmetry unique coordinates to be specified (For my input it should be 15 as the molecule possess 15 atoms with no symmetry) . This is followed by two sets of input - one line with the energy (No format or unit has been shown. So, I used the energy printed in the output file in au) and then the Cartesian coordinates in bohr - for each of the two starting structures of the Saddle method(The Cartesian coordinate in Bohr also given). Note that the order of the coordinates must always match the order specified with the conventional input of the coordinates of the molecular system (I think this line is for The order of the atoms must be identical within all molecule).
Another things you suggested in last mail "This means that you must still specify a Coord keyword, as in a normal calculation." I have tried to put 'Coord' Keyword in many ways but nothing works.
Here is the Gateway section of my input.
&GATEWAY
RP-Coord
15
-271.52287184233336
C -0.021511200 0.021815450 0.126188030
C -0.041221710 -0.125512680 1.463377340
C 1.284241830 -0.001003470 -0.664429650
C -1.254149670 0.229360750 -0.629649960
C -1.281458680 0.185955260 -1.979405570
C -0.071748090 -0.075721730 -2.754920820
C 1.121654140 -0.189180560 -2.151622540
H 1.811951590 0.954293600 -0.504534250
H -2.170404110 0.378859790 -0.063888710
H -2.220451020 0.317563850 -2.509694120
H -0.158258310 -0.182012610 -3.832720200
H 2.021928790 -0.378508970 -2.730772530
H 0.863419620 -0.310999810 2.034972110
H -0.969363180 -0.059865960 2.023248210
H 1.958379000 -0.765028920 -0.257307360
-271.57877382493052
C 0.000000000 0.015350290 0.010347800
C 0.000000000 -0.002474530 1.521230950
C 1.202323600 0.011298940 -0.709145700
C -1.202323600 0.011298950 -0.709145690
C -1.205182920 -0.002396800 -2.104311770
C 0.000000000 -0.010145700 -2.807968000
C 1.205182920 -0.002396820 -2.104311770
H 2.146234890 0.021105620 -0.169263100
H -2.146234890 0.021105650 -0.169263100
H -2.149545250 -0.003011030 -2.641506680
H 0.000000000 -0.017656450 -3.894011400
H 2.149545250 -0.003011050 -2.641506680
H 0.000000050 -1.030048180 1.906105030
H 0.885291940 0.495488600 1.927973560
H -0.885292000 0.495488510 1.927973550
Basis=C.6-31G**.., H.6-31G**..,
Title=single point in Hartree Fock
Group=Full
End of inputUnfortunately, I am unable to catch the proper format and I am really sorry for the same question again. I will be very grateful if you help me to correct the input.
One more Query about TS calculation in RP-Coor method using CASSCF: I am using CASSCF(8,8) method to model a intra-molecular H transfer reaction. I have optimized both reactant and product in that method. Is it possible here to read geometry and also Guess (not only geometry) form the reactant as well as from the product and use them to determine the transition state? Can I follow the trajectory during this TS optimization?
With Best Regards
Debasish
Offline
#4 2016-11-29 16:38:55
- Ignacio
- Administrator
- From: Uppsala
- Registered: 2015-11-03
- Posts: 1,085
Re: Several questions
Unfortunately, I am unable to catch the proper format and I am really sorry for the same question again. I will be very grateful if you help me to correct the input.
It would help if you showed what is the problem / error message you get. Anyway, I recommend using xyz files instead of inline coordinates. Something like this should work:
&GATEWAY
Coord = reac.xyz
RP-Coord
reac.xyz
prod.xyz
Basis = 6-31G**
Group = NoSymwhere "reac.xyz" and "prod.xyz" are standard XYZ files, in ångström, with the particularity that the energy (in hartree) should be given in the second line, like these:
reac.xyz
8
-151.84918514994180
C -0.68417499 -0.77058346 0.00000000
C 0.68417499 -0.77058346 0.00000000
C -0.68417499 0.77058346 0.00000000
C 0.68417499 0.77058346 0.00000000
H -1.45888094 -1.52347960 0.00000000
H 1.45888094 -1.52347960 0.00000000
H -1.45888094 1.52347960 0.00000000
H 1.45888094 1.52347960 0.00000000prod.xyz
8
-151.84918514994180
C -0.77058346 -0.68417499 0.00000000
C 0.77058346 -0.68417499 0.00000000
C -0.77058346 0.68417499 0.00000000
C 0.77058346 0.68417499 0.00000000
H -1.52347960 -1.45888094 0.00000000
H 1.52347960 -1.45888094 0.00000000
H -1.52347960 1.45888094 0.00000000
H 1.52347960 1.45888094 0.00000000To avoid problems the structures and energies should correspond to optimized minima, with the same method you use for the calculation. You can just use the final $Project.Opt.xyz that you get after an optimization.
Is it possible here to read geometry and also Guess (not only geometry) form the reactant as well as from the product and use them to determine the transition state?
The "saddle" method works by setting the environment variable SubProject to ".Reac" and ".Prod" alternately (more or less) and doing two constrained optimizations. You can put two files with names $Project.Reac.RasOrb and $Project.Prod.RasOrb in the scratch directory, and then add "FileOrb = $Project$SubProject.RasOrb" in &RASSCF, it may work.
Can I follow the trajectory during this TS optimization?
You can view the file $Project.geo.molden or $Project.saddle.molden, the latter contains only the intermediate "optimized" structures, and the TS, if successful.
You may consider using the "FindTS" method instead (keyword in &SLAPAF), it's easier to use.
Offline
Pages: 1