Top page (← 10/2024 )
Molecular dynamics is useless
(Fig.5) Even when two atoms hit each other, quantum mechanical fictitious electrons and nuclei cannot move realistically, which are different from two actual balls hitting !
Due to the useless unsolvable multi-electron Schrödinger equations, quantum mechanics had No choice but to rely on ad-hoc approximation treating multi-electron atoms or molecules as one pseudo-electron density functional (= Kohrn-Sham ) theory (= DFT, KS ) with artificially-chosen pseudo-potential.
↑ In this current mainstream quantum mechanical one-pseudo-electron DFT or quasiparticle model, it is impossible to give actual shapes to individual atoms or explain the motion of those individual atoms realistically.
The current only method of simulating or explaining dynamical motion of molecules is the extremely-time-consuming molecular dynamics (= MD ).
Ab-initio or first-principle molecular dynamics (= AIMD, FPMD ) such as Car-Parrinello molecular dynamics (= CPMD ) or Born-Oppenheimer MD (= BOMD ) is based on quantum mechanical unphysical one-pseudo-electron DFT approximation ( this p.7-lower~p.8, this p.26, this p.1-middle-last~p.1-right-upper ) with artificially-chosen pseudo-potential ( this 1st-paragraph ) without giving actual shapes to individual atoms.
In this ab-initio MD (= CPMD, BOMD ), physicists have to artificially choose (or guess ) the DFT's one-pseudo-electron density consisting of unphysical plane-wavefunction with (artificially-chosen) coefficients ( this p.14-22 ), and (fictional) pseudo-potentials ( this p.2, this p.10-right-10.1 ).
In each time step (= each time step Δt updating nuclear positions and electron's coefficients must be very short = less than 1 femtosecond or 1 × 10-15 s ) of BOMD, they have to find (adjust) the coefficients giving the lowest (ground-state) energy within the artificially-chosen pseudo-electron's wavefunctions or plane-wave basis sets (= ab-initio MD cannot move electrons realistically, this p.32-33 ), and move only nuclei classically, which takes unrealistically too much time.
Car-Parrinello molecular dynamics (= CPMD ) using fictitious electron mass (= μ, this p.2-left-(1), this p.39 ) and DFT pseudo-potential ( this p.28-32, this p.12 ) slightly modifies the coefficients (= c ) of the artificially-chosen plane-wavefunctions ( this p.12 ) in each time step, repeatedly to try to arrive at the lowest energy state in the end ( this p.16 ) = trying to find the coefficients giving the lowest-energy ground-state.
This paper ↓
p.35- Ab-initio molecular dynamics (= AIMD ) relies on (one-pseudo-electron) DFT model.
p.40- Ab-initio MD tries to move only a nucleus with mass M by (fictitious) classical Newtonian motion.
p.41- Ab-initio MD expresses electrons as nonphysical plane wave with artificially chosen basis sets, coefficients c, and pseudo-potential.
p.44- Ab-initio MD (= CPMD ) tries to change the coefficients c of the unphysically-spreading plane wave or pseudo-electron, instead of moving real electrons.
This p.2-left-(5)-middle says
"The ion (= nuclear ) dynamics
in Eqs. (5) may have a real physical meaning, whereas
the dynamics associated with the (electron's) ψ's and the α's
is fictitious and has to be considered only as a tool to
perform the dynamical simulated annealing (= electron's wavefunction's motion or its coefficients' change is fictitious in ab-initio molecular dynamics )."
This paper ↓
p.19-20 In Born-Oppenheimer molecular dynamics (= BOMD ), they restrict the wave function ψ to be the ground state adiabatic wave function (= fictitious electron's wavefunction's coefficients are adjusted to give the lowest ground state energy within the chosen wavefunction or basis set ) at each instant of time, and only nuclei are moved classically.
p.21-22 uses one-pseudo-electron DFT.
p.27-28 Car-Parrinello molecular dynamics tries to keep electron's wavefunction (= coefficients ) close to ground-state by using fictitious electron's mass.
This paper ↓
p.3-1.1.1 Limitations say
"This is generally true for insulators and semiconductors, but not for metals, for
which CPMD (= ab-initio Car-Parrinello molecular dynamics ) cannot be directly applied. In addition, a small enough fictitious
mass for the electrons must be chosen to allow the orbitals to follow the ions
adiabatically.
CPMD also suffers from the typical drawbacks that DFT suffers from, such as
the inability to describe van der Waals forces, the band gap problem and the
inability to accurately describe highly correlated electrons with localized d and
f orbitals."
p.4-6 Ab-initio MD or CPMD just changes coefficients (= c ) of the unphysical artificially-chosen pseudo-electron's wavefunction.
This paper ↓
p.16-(45) uses fictitious mass μ
p.36--2nd-paragraph says "one–electron equations involving an effective one–particle Hamiltonian of Kohn-Sham DFT"
p.41-lower & p.48-3.1.5 mention pseudo-potentials
This ab-initio molecular dynamics (= AIMD ) based on the unphysical mainstream quantum mechanical one-pseudo-electron DFT model is unreal, useless, far more time-consuming ( this p.12, this p.2, this p.1-abstract, this p.2-2nd-paragraph ) than the (pseudo-)classical MD based on empirical pseudo-potential or force field ( this p.2-left-1st-paragraph ).
Even in the latest researches, the (time-consuming) ab-initio molecular dynamics (= MD ) can simulate only impractically-short ps (= picoseond ) motion of molecules with a less than 1 fs time step ( this p.3-left-1st-paragraph ), which cannot simulate much longer important biological or chemical reactions (= seconds ~ hours ).
This p.7-right-ab-initio molecular dynamics simulations used (DFT) pseudo-potential to conduct only 10-20ps-motion simulation with 1.0fs time step.
This unphysical fictitious ab-inito molecular dynamics of CPMD is unreasonably used to express even the atoms actually observed by microscopes ( this p.10-calculations, this p.5-3. this p.8(or p.7)-DFT calculation ), which quantum mechanical unphysical model hampers nano-technology ( this introduction-4th-paragraph ).
Feel free to link to this site.
