CCL:G: ADMP total kinetic energy related to velocities
- From: "Dario Gregorio Perez"
<derwiszck(!)gmail.com>
- Subject: CCL:G: ADMP total kinetic energy related to
velocities
- Date: Fri, 11 Apr 2014 03:56:53 -0400
Sent to CCL by: "Dario Gregorio Perez" [derwiszck|-|gmail.com]
Hello all CCL Subscribers,
I would like to focus on a problem that I obtained with ADMP (Atom Centered
Density Matrix Propagation) implemented in gaussian.
I am trying to calculate the kinetic energies from velocities.
As you know we can define the NKE (nuclear kinetic energy) in the input and
program randomly generate us the velocities. However, generated velocities:
ADMP step 0
GnVelC Random: Velocity
I= 1 X= 3.398782212293D-01 Y= 1.465871681630D-01 Z= 6.424346183195D-01
I= 2 X= 5.013104147695D-01 Y= 3.472184481623D-01 Z= 5.311228776839D-01
I= 3 X= 8.478196138519D-01 Y= 4.055930070705D-01 Z= 6.444282418968D-01
I= 4 X= 1.676431686815D-01 Y= 2.277398149910D-01 Z= 6.024104048879D-02
I= 5 X= 1.732845504177D-01 Y= 9.368130619597D-01 Z= 5.517780556128D-01
I= 6 X= 6.893582811025D-01 Y= 4.348692854433D-01 Z= 3.349183234847D-02
I= 7 X= 7.669745262782D-01 Y= 4.634406180743D-01 Z= 8.149850049870D-01
I= 8 X= 5.653485273988D-01 Y= 6.683079621588D-01 Z= 1.607799839639D-01
I= 9 X= 6.846587566952D-01 Y= 6.774363692626D-01 Z= 9.833239520217D-01
I= 10 X= 6.558141826505D-01 Y= 2.262089648952D-01 Z= 9.651211563677D-01
I= 11 X= 8.738928624306D-01 Y= 9.862285094204D-01 Z= 9.754484548179D-01
I= 12 X= 5.057309487267D-01 Y= 7.118213486021D-01 Z= 8.452797536316D-01
I= 13 X= 7.157120462557D-01 Y= 8.235984112087D-01 Z= 7.521027257725D-02
GnVelC Scaled: Velocity
I= 1 X= 4.697014674563D-02 Y= 2.025790524247D-02 Z= 8.878252977727D-02
I= 2 X= 6.927958979445D-02 Y= 4.798454400516D-02 Z= 7.339958239907D-02
I= 3 X= 1.171661177124D-01 Y= 5.605173227103D-02 Z= 8.905804255253D-02
I= 4 X= 2.316778111110D-02 Y= 3.147295667036D-02 Z= 8.325130400041D-03
I= 5 X= 2.394740307993D-02 Y= 1.294647442673D-01 Z= 7.625406579277D-02
I= 6 X= 9.526723867914D-02 Y= 6.009762578655D-02 Z= 4.628470381238D-03
I= 7 X= 1.059935700474D-01 Y= 6.404609792326D-02 Z= 1.126284736376D-01
I= 8 X= 7.812946413071D-02 Y= 9.235814798704D-02 Z= 2.221930964929D-02
I= 9 X= 9.461777855709D-02 Y= 9.361966636168D-02 Z= 1.358924092220D-01
I= 10 X= 9.063154527979D-02 Y= 3.126139779676D-02 Z= 1.333768376742D-01
I= 11 X= 1.207693621552D-01 Y= 1.362938103084D-01 Z= 1.348040392229D-01
I= 12 X= 6.989049427639D-02 Y= 9.837156696767D-02 Z= 1.168151166779D-01
I= 13 X= 9.890924966787D-02 Y= 1.138188204411D-01 Z= 1.039383319989D-02
GnVelC MW: Velocity
I= 1 X= -7.767235652917D-03 Y= -8.629435985058D-02 Z= -2.943473782054D-02
I= 2 X= -1.051105078791D-03 Y= -3.895073834557D-02 Z= -2.234634745708D-02
I= 3 X= 3.993532120872D-02 Y= -6.509787237278D-03 Z= 3.797451060452D-02
I= 4 X= -7.340768346256D-02 Y= -2.927558997435D-02 Z= -3.158980977140D-02
I= 5 X= -9.115183242473D-02 Y= 4.121349582598D-02 Z= -4.331312607715D-02
I= 6 X= 6.204152852713D-03 Y= -7.243917416043D-04 Z= -3.186792874862D-02
I= 7 X= 8.686570687864D-02 Y= 3.810551973465D-02 Z= 8.400773139458D-02
I= 8 X= 6.267344029835D-02 Y= 5.941458318215D-02 Z= -1.843245336307D-02
I= 9 X= 8.105817639074D-02 Y= 6.647616584633D-02 Z= 1.075468825745D-01
I= 10 X= 5.858619789816D-02 Y= 6.284739985947D-04 Z= 8.940120953276D-02
I= 11 X= 8.018795918517D-02 Y= 1.128080348882D-01 Z= 1.008359087201D-01
I= 12 X= 3.293437281697D-02 Y= 7.532804416915D-02 Z= 8.543617786663D-02
I= 13 X= 7.472863614520D-02 Y= 1.030715738732D-01 Z= 9.256829738788D-03
GnVelC Scaled: Velocity
I= 1 X= -1.083126389307D-02 Y= -1.203358602458D-01 Z= -4.104618775637D-02
I= 2 X= -1.465746244412D-03 Y= -5.431607134159D-02 Z= -3.116156083960D-02
I= 3 X= 5.568905360863D-02 Y= -9.077775750015D-03 Z= 5.295474013503D-02
I= 4 X= -1.023656326255D-01 Y= -4.082425908095D-02 Z= -4.405139502076D-02
I= 5 X= -1.271095143043D-01 Y= 5.747144404967D-02 Z= -6.039933890757D-02
I= 6 X= 8.651574354571D-03 Y= -1.010150646367D-03 Z= -4.443922669866D-02
I= 7 X= 1.211325929203D-01 Y= 5.313742990063D-02 Z= 1.171471999116D-01
I= 8 X= 8.739693261438D-02 Y= 8.285251771665D-02 Z= -2.570370920794D-02
I= 9 X= 1.130341009867D-01 Y= 9.269976180145D-02 Z= 1.499721030871D-01
I= 10 X= 8.169735003318D-02 Y= 8.763951594743D-04 Z= 1.246683036384D-01
I= 11 X= 1.118205994761D-01 Y= 1.573086809428D-01 Z= 1.406137763870D-01
I= 12 X= 4.592636287524D-02 Y= 1.050435394784D-01 Z= 1.191391416251D-01
I= 13 X= 1.042076762733D-01 Y= 1.437313693547D-01 Z= 1.290847480292D-02
do not match to nuclear kinetic energy given in the input. Moreover, they are
few times bigger (around 3.5 times than given energy). I tried to calculate KE
including for both: "pure" and scaled. Without success.
Due to that problem I would like to ask you:
1. Total kinetic energy is: translation kinetic energy + rotation KE and
vibration KE. Gaussian output print in the ADMP step 0 this:
Summary information for step 0
Time (fs) 0.000000
EKinC = 0.1567570; EKinPA = 0.0000000; EKinPB = 0.0000000
EKin = 0.1567570; EPot = -322.7887432; ETot = -322.6319862
ETot-EKinP = -322.6319862
Integrand for the Velocity-Velocity Auto-Correlation Function =
2.939453785855D-01
Angular momentum
JX = -0.2072213981D-13 JY = -0.5055706342D-14 JZ = 0.1946845813D-13
Jtot = 0.2887885224D-13 H-BAR; J (Quantum Number) = 0.0000000000D+00
Total energy -3.226319862D+02 A.U.
Total angular momentum 2.887885D-14 h-bar
So, EKinC is the total kinetic energy. So keyword NKE in referring to total
kinetic energy. Does it mean that include also not only vibrational part but
also translational and rotational.
Rotational looks negligible (Jtot = 0.2887885224D-13 H-BAR) so does it mean
that velocities "contain" translational and vibrational part?
Later to do dynamic velocities are scaled again by factor of 10^15. Why?
Is it just to minimized the numerical errors?
Summary.
Total kinetic energy calculated as:
KE=SUM[i=1 to N](0.5*m"i"*|V"i"|) gives me value
0.5563189665 in a.u.
Which is much bigger than printed in gaussian.
Than I remove translational part considering that TKE is momentum^2/2m so,
TKE=1/2M SUM[i=1 to N] (m"i"^2*v"i"^2), where M is total
mass.
TKE=0.0691092388
Than when I substract (KE-TKE) I should get vibrational KE, which is still much
bigger that the one printed by gaussian.
Please, can you drive me through this problem, where is the error?
Units follow by ADMP maunal should be in a.u (??bohr/sec??).
Thank you,
Darek