Input/Output files¶
Input Files¶
filename¶
filename is a simple, single line text file containing the path to the main input file, relative to the QXMD executable, in between single quotes. Below shows the entire contents of a sample filename
CONFIG¶
An initial configuration file must be included for any new QXMD run. It is best practice to name this file CONFIG. Below shows a sample configuration file for a water molecule.
3
1 0.579029 0.305583 0.264540
1 0.420969 0.305577 0.264541
2 0.500002 0.244625 0.264685
The first line should contain the total number of atoms in the simulation system. In this case, the water molecule has 3 atoms: 2 hydrogen and 1 oxygen. Each of the following lines (one line for each atom) contains a keyword representing the species the atom, as well as the atom’s x, y, and z coordinates. These may be given in real space of fractional coordinates (you define which kind are provided in the IN.PARAM file). In the example above, the second and third lines give the spatial coordinates for the hydrogen atoms, using ‘1’ as the keyword for hydrogen, while the last line give the spatial coordinates for the oxygen atom, using ‘2’ as the keyword for oxygen. Integer numbers should be used for keywords, though it is arbitrary which numbers are chosen for each species, as long as what you choose is used consistently with atomic data entered in the main input file, IN.PARAM.
IN.PARAM¶
A main input file is required for all QXMD runs, which contains all the parameters settings. It is best practice to name the main input file IN.PARAM. There are on the order of a few hundred input parameters that may be tailored for various kinds of QXMD simulations, though most parameters have default settings, many of which need not be changed for most routine QXMD simulations. Details of these parameters may be found in section 5.
VELOC¶
A file defining the intial velocities of the atoms may optionally be provided. It is best practice to name this file VELOC. If this is not provided, random inital velocities will be assigned to each atom, in accordance with the initial system temperature. Below shows a sample initial velocity file for a water molecule.
3
2.7271510E-03
1 0.058157 0.043210 0.003117
2 -1.000000 -0.688863 -0.057829
2 0.076852 0.002967 0.008360
The first line should contain the total number of atoms in the simulation system. In this case, the water molecule has 3 atoms: 2 hydrogen and 1 oxygen. The next line is a scaling factor by which all of the following veloctiy components should be multiplied. Each of the following lines (one line for each atom) contains a keyword representing the species the atom, as well as the atom’s x, y, and z velocity components in atomic units. The integer keywords for the atomic species should be consistent with those provided in the IN.CONFIG and IN.PARAM files.
Output Files¶
All output data files are dumped to the data directory during QXMD runs. Output files with names beginning with a lower case letter are human-readable text files detailing simulation data, while output files beginning with upper case letters are binary files to be used for restarting a QXMD run. These files must be present in the data directory when attempting to restart a job. After a simulation, all output files that you wish to save should be moved out of the data directory and store elsewhere as a new QXMD run will write over all files present in data.
md_box.d¶
This file gives the simulation box size for a classical MD simulation. A sample of output file md_box.d is shown below.
# MD cell vectors
# L_1 L_2 L_3 angle(2-3) angle(3-1) angle(1-2)
0 3.7280514E+01 3.7280514E+01 3.0613561E+01 90.000000 90.000000 120.000000
L_1, L_2, and L_3 are the lengths of three cell vectors in Bohr units, while angles between these vectors, in degrees, follow. The ‘0’ at the beginning of the last line represents the MD step number. In this example, the simulation box size is held fixed, and thus this is the only line that appears in the output. If these numbers were updated at any point, a new line would appear under the last line with the step number in which the box size was altered, followed by the new box dimensions.
Note that there is also an output file named qm_box.d, which provides the supercell size for a quantum MD simulation.
md_cel.d¶
Similar to md_box.d, md_cel.d holds the cell vector data for the simulation cell in which molecular dynamics take place, however instead of giving the lengths and angles of the vectors, this file gives the x, y, and z components of the three cell vectors. A sample of output file md_cel.d is shown below.
# MD cell vectors
# L_(1:3,1:3)
0 3.7280514E+01 0.0000000E+00 0.0000000E+00 -1.8640257E+01 3.2285872E+01 0.0000000E+00 1.8745400E-15 3.2467985E-15 3.0613561E+01
After the two comment lines, the file lists the step number, in this case ‘0’, and then x, y, and z components of the first cell vector, followed by those of the second and third cell vectors. In this case, there were no changes in the cell vectors during the simulation, however, if changes do occur, additional lines will be written beginning with the step number in which the cell vectors were changed, followed by their new values.
md_eng.d¶
This file contains the Hamiltonian, potential, and kinetic energies [Hartree] of the total system, as well as the temperature [K] at every time step. A sample of the output is shown below.
# step H [hartree] P.E. [hartree] K.E. [hartree] T[K]
0 -1.4792707116E+04 -1.47928610E+04 1.53902180E-01 300.0000
1 -1.4792706554E+04 -1.47928611E+04 1.54538843E-01 301.2410
2 -1.4792705828E+04 -1.47928609E+04 1.55091556E-01 302.3184
3 -1.4792704962E+04 -1.47928605E+04 1.55563674E-01 303.2387
4 -1.4792703950E+04 -1.47928599E+04 1.55970999E-01 304.0327
5 -1.4792703306E+04 -1.47928596E+04 1.56301244E-01 304.6765
6 -1.4792702504E+04 -1.47928591E+04 1.56548702E-01 305.1588
7 -1.4792701327E+04 -1.47928580E+04 1.56710424E-01 305.4741
8 -1.4792700224E+04 -1.47928570E+04 1.56777438E-01 305.6047
9 -1.4792699401E+04 -1.47928562E+04 1.56763969E-01 305.5785
10 -1.4792698443E+04 -1.47928551E+04 1.56675161E-01 305.4053
md_log¶
This file provides of log of the simulation, including values of input parameters set and lengths of computation time for force calculations for each time step.
MD_mts0¶
This is a binary file that is required for restarting a QXMD simulation.
MD_mts0¶
This is a binary file that is required for restarting a QXMD simulation.
md_spc.d¶
This file lists the species keyword for each atom at each time step. A sample of the output is shown below.
# Atomic species
2 42 34
0 12
1 1 1 1 2 2 2 2 2 2 2 2
1 12
1 1 1 1 2 2 2 2 2 2 2 2
2 12
1 1 1 1 2 2 2 2 2 2 2 2
3 12
1 1 1 1 2 2 2 2 2 2 2 2
4 12
1 1 1 1 2 2 2 2 2 2 2 2
5 12
1 1 1 1 2 2 2 2 2 2 2 2
In the second line, ‘2’ represents the total number of species in the system. The following numbers,‘42’ and ‘34’, are the atomic numbers of the two atomic species in the system ordered by the keywords representing those species. In this example, molybdenum has atomic number 42, and is represented by keyword ‘1’, while selenium has atomic number 34, and is represented by keyword ‘2’. The next line lists the step number, followed by the total number of atoms in the system, in this case ‘12’.
md_str.d¶
This file outputs components of the stress tensor [GPa] of the system, computed classically, at given time steps intervals (the interval is defined in IN.PARAM). This file will only be written if dumping stress data is set to true in IN.PARAM. Below is sample output.
# Stress in [GPa]
# Pxx Pyy Pzz Pyz Pzx Pxy
0 1.33846254E+01 1.32248154E+01 5.38055547E+00 2.74846400E-02 4.09607470E-02 -9.31727041E-01
5 1.37099707E+01 1.34067898E+01 6.81653796E+00 1.21177457E-02 2.32549066E-01 -1.03833531E+00
10 1.52252127E+01 1.57211516E+01 9.64813968E+00 3.34044020E-02 3.45617372E-01 -8.50283630E-01
15 1.68630437E+01 1.59302869E+01 1.19623442E+01 1.09971448E-01 4.41730570E-01 7.93058521E-01
20 1.78613247E+01 1.62869327E+01 1.36732341E+01 1.16372513E-01 4.25461913E-01 1.08703422E+00
After the first two comment lines, the following lines list the step number followed by the components of the stress tensor for that time step. Here, stress data is dumped every 5 time steps. Note that contributions from the kinetic energy of the ions are included in these values, while in qm_str.d they are not.
md_str_diag.d¶
This file outputs the diagonalized stress tensor, along with the respective eigenvectors at given time step intervals. Below is sample output.
# Diagonalized stress in [GPa]
# Pxx Pyy Pzz Eigenvectors
0 1.42398826E+01 1.23699022E+01 5.38021140E+00 7.3672E-01 -6.7620E-01 1.3084E-03 6.7618E-01 7.3671E-01 6.8597E-03 -5.6024E-03 -4.1690E-03 9.9998E-01
5 1.46113490E+01 1.25135917E+01 6.80835783E+00 7.5737E-01 -6.5263E-01 2.1581E-02 6.5207E-01 7.5764E-01 2.8228E-02 -3.4773E-02 -7.3072E-03 9.9937E-01
10 1.46051510E+01 1.63637815E+01 9.62557143E+00 7.9488E-01 6.0384E-01 5.9491E-02 -6.0340E-01 7.9698E-01 -2.7090E-02 -6.3770E-02 -1.4364E-02 9.9786E-01
15 1.73523452E+01 1.54808731E+01 1.19224564E+01 8.6781E-01 4.9024E-01 8.1122E-02 -4.8915E-01 8.7153E-01 -3.4170E-02 -8.7452E-02 -1.0028E-02 9.9612E-01
20 1.84552850E+01 1.57357781E+01 1.36304284E+01 8.8838E-01 4.5019E-01 8.9996E-02 -4.4823E-01 8.9293E-01 -4.2081E-02 -9.9304E-02 -2.9550E-03 9.9505E-01
After the first two comment lines, the following lines list the step number followed by the components of the diagonalized stress tensor and the x, y, and z components of the three eigenvectors for that time step. Here, diagonalized stress data is dumped every 5 time steps.
MD_Tocontinue¶
This is a binary file that is required in the data directory to restart a QXMD simulation.
md_vel.d¶
This file contains scaled atomic velocities for each atom at each time step. Below is sample output.
# Atomic scaled velocities
0 12
1.7316159E-06
-6.94686 -2.92359 -1.82411 5.50162 -1.28242 2.31888 0.35217 -0.34956 1.04083
-2.29616 2.29092 0.53871 -0.03100 -5.38400 -1.78311 2.52055 1.50530 1.29961
1.94346 1.81671 -2.48381 5.25972 2.70791 1.54716 5.36706 8.68984 1.05675
-5.95609 -9.99900 -0.80332 -3.99628 1.60613 -0.32696 -0.98935 1.80875 -1.02671
The second line gives the step number, in this case ‘0’, and the total number of atoms in the system, in this case, ‘12’. The next line gives the number by which to scale the components of velocity given in the following lines (i.e. multiply the following numbers to get the absolute velocity values). The following lines give the x, y, and z components of velocity for the first atom, followed by those for the second atom, ending with those for the last atom.
qm_box.d¶
This file holds the cell vector data for the supercell for a quantum molecular dynamics simulation. A sample of output file is shown below.
# supercell (FFT cell) vectors (lengths & angles)
# L_1 L_2 L_3 angle(2-3) angle(3-1) angle(1-2)
0 1.2426838E+01 1.2426838E+01 3.1114338E+01 90.000000 90.000000 120.000000
L_1, L_2, and L_3 are the lengths of three supercell vectors in Bohr units, while angles between these vectors, in degrees, follow. The ‘0’ at the beginning of the last line represents the step number. In this example, the simulation box size is held fixed, and thus this is the only line that appears in the output. If the supercell size were changed, a new line would appear under the last line with the step number in which the size was altered, followed by the new supercell dimensions.
Note that there is also an output file named md_box.d, which holds the box size for a classical MD simulation. For purely quantum MD simulations, md_box.d and qm_box.d will hold the same information (except in cases where the double-grid method is used). However, it is possible to perform a hybrid classical-quantum MD simulation in which case the system is treated classically, except for a defined area inside which is treted quantum mechanically. In this case, md_box.d gives the box size of the entire system, while qm_box.d defines the supercell area that is to be treated with QM.
QM_cds¶
This is a binary file that is required in the data directory to restart a QXMD simulation, containing charge density data for the system at the last step of the previous run.
qm_cel.d¶
Similar to qm_box.d, qm_cel.d holds the cell vector data for the supercell, however instead of giving the lengths and angles of the vectors, this file gives the x, y, and z components of the three supercell vectors. Below is sample output.
# supercell (FFT cell) vectors
# L_(1:3,1:3)
0 1.2426838E+01 0.0000000E+00 0.0000000E+00 -6.2134191E+00 1.0761957E+01 0.0000000E+00 1.9052037E-15 3.2999097E-15 3.1114338E+01
After the two comment lines, the file lists the step number, in this case ‘0’, and then x, y, and z components of the first supercell vector, followed by those of the second and third supercell vectors. In this case, there were no changes in the supercell vectors during the simulation, however, if changes do occur, additional lines will be written beginning with the step number in which the supercell vectors were changed, followed by their new values.
QM_cell¶
This is a binary file that is required in the data directory to restart a QXMD simulation, containing simulation cell data for the system at the last step of the previous run.
QM_eig¶
This is a binary file that is required in the data directory to restart a QXMD simulation, containing energy eigenvalue data for the system at the last step of the previous run.
qm_eig.d¶
This file lists the energy eigenvalues, along with their electronic occupation number for each time step. Below is sample output.
# Eigenvalues
0 7 10
1 -1.79706E+00 2.000
2 -8.95274E-01 2.000
3 -6.37670E-01 2.000
4 -4.86501E-01 2.000
5 -9.48240E-02 0.000
6 1.26780E-01 0.000
7 1.34093E-01 0.000
8 1.69431E-01 0.000
9 1.94460E-01 0.000
10 2.79168E-01 0.000
1 11 10
1 -1.79215E+00 2.000
2 -8.86033E-01 2.000
3 -6.40897E-01 2.000
4 -4.85737E-01 2.000
5 -9.72281E-02 0.000
6 1.20947E-01 0.000
7 1.34115E-01 0.000
8 1.68074E-01 0.000
9 1.93969E-01 0.000
10 2.75769E-01 0.000
The first number in the second line gives the step number, in this case ‘0’. The second number gives the cumulative number of SCF iterations performed, in this case ‘7’ iterations were performed in step 0. Finally, the third number, ‘10’ represents the total number of energy eigenvalues and occupation numbers to follow. This number corresponds to the number of energy bands for the system, defined in IN.PARAM. Thus the next ten lines list the band index number, the energy in eV, and the number of electrons which occupy that energy band. In the second step, 4 SCF iterations were performed since the number ‘11’ follows the step number ‘1’ (7 SCF iterations were performed in the first step, followed by 4 SCF iterations, for a total of 11 SCF iterations after the first two time steps.) In this example, the eight electrons of the system occupy the four lowest energy bands in the first time two steps.
qm_eng.d¶
This file provides various components of the system’s energy [Rydberg] at each time step. Below is sample output.
# Total potential energy and energy parts in [Ryd.] units
# Total(HF) Total(KS) Kinetic External Hartree Exchange Correlation ------ Entropy Onsite E. -------- ------ Ewald E. DFT-D
0 7 -4.4003610122E+01 -4.4003640357E+01 1.370805E+01 -6.080189E+01 2.800936E+01 -7.786772E+00 -5.786949E-01 0.000000E+00 0.000000E+00 -1.533928E+01 0.000000E+00 0.000000E+00 -1.214255E+00 -1.610012E-04
1 11 -4.4002961457E+01 -4.4003371148E+01 1.367946E+01 -6.058249E+01 2.791215E+01 -7.771542E+00 -5.778955E-01 0.000000E+00 0.000000E+00 -1.532657E+01 0.000000E+00 0.000000E+00 -1.336327E+00 -1.631839E-04
2 15 -4.4001366424E+01 -4.4002003283E+01 1.365563E+01 -6.038887E+01 2.782594E+01 -7.758172E+00 -5.771782E-01 0.000000E+00 0.000000E+00 -1.531663E+01 0.000000E+00 0.000000E+00 -1.442551E+00 -1.647241E-04
3 18 -4.3999479244E+01 -4.3999841762E+01 1.363668E+01 -6.023643E+01 2.775745E+01 -7.747680E+00 -5.766015E-01 0.000000E+00 0.000000E+00 -1.530791E+01 0.000000E+00 0.000000E+00 -1.525182E+00 -1.650145E-04
4 22 -4.3997932717E+01 -4.3997243098E+01 1.362538E+01 -6.013514E+01 2.771149E+01 -7.740722E+00 -5.762123E-01 0.000000E+00 0.000000E+00 -1.530223E+01 0.000000E+00 0.000000E+00 -1.579643E+00 -1.636864E-04
5 25 -4.3997129513E+01 -4.3997300982E+01 1.362022E+01 -6.009034E+01 2.769035E+01 -7.737485E+00 -5.760705E-01 0.000000E+00 0.000000E+00 -1.530042E+01 0.000000E+00 0.000000E+00 -1.603399E+00 -1.606360E-04
After the first two comment lines, each line gives the step number, the cumulative number of SCF iterations completed up to that step number, followed by various components of system energy as labeled by the column titles.
qm_fer.d¶
This file gives the Fermi energy of the system at each time step. Below is sample output.
# Fermi energy
0 7 -2.90493E-01
1 11 -2.90662E-01
2 15 -2.91483E-01
3 18 -2.92223E-01
4 22 -2.92859E-01
5 25 -2.93001E-01
The first column gives the step number, the second column gives the cumulative number of SCF iterations up to that step number, while the third column gives the Fermi energy in eV.
qm_frc.d¶
This file gives the three components of force on each atom at each time step. Below is sample output for a monolayer of MoSe2 with 12 total atoms.
# Atomic forces in [a.u.]
0 2 4 8
3.7133826E-02
0.14634 1.41724-0.01118-0.15415 0.36847-0.00294 0.87782-0.64926 0.00080
-0.89534-0.92856 0.00524 1.28410 0.95522 8.33389-1.27916 1.17417 9.99839
-0.17476-0.37722 6.37526 0.18425-1.85265 8.15680 1.29026 0.94518-8.33006
-1.28567 1.15640-9.99900-0.17251-0.36263-6.37149 0.17883-1.84636-8.15571
1 2 4 8
3.5751399E-02
-1.40927-0.53578 0.00521 1.21420-0.62553 0.03759 0.23010-0.18164-0.00182
-0.27261 1.29953 0.02477-0.13760-1.72196 9.40196 0.26416 0.31938 6.59092
-1.83555 0.61826 8.18927 1.83320 0.80236 9.98594-0.10811-1.77374-9.39287
0.19630 0.32076-6.62970-1.84943 0.69302-8.21226 1.87463 0.78534-9.99900
After the comment line, the next line gives the step number, the total number of atomic species in the system (‘2’ for molybdenum and selenium), the number of atoms of the species corresponding to keyword ‘1’ (in this case, keyword ‘1’ was used for molybdenum and there are ‘4’ molybdenum atoms), followed by the number of atoms of the species corresponding to keyword ‘2’ (in this case, keyword ‘2’ was used for selenium and there are ‘8’ selenium atoms). The number in the third line is a scaling factor for the following force vector components (multiply all of the following force components by this number to get the true values for force). The following lines give the x, y, and z components of force on the first atom, followed by those for the second atom, and so on.
qm_fsshprob_***to***-u.d¶
This set of files are only written during a Non-Adiabatic QMD simulation (i.e. TD-DFT set to .true.). These files give the probabilities for electrons to hop from one band to another, where these band indices will be given in the title of the file in place of the ‘*’. A sample file name from this set could be qm_fsshprob_29to32-u.d, which will give the probability for an electron to transition from band index 29 to band index 32 at each time step that has a non-zero probability. The ‘-u’ at the end of the file name indicates that the data is for spin-up electrons in the case that spin polarization is used in the QXMD simulation. In this case, spin-down electron data will be stored in files named ‘qm_fsshprob_***to*-d.d’ If spin polarization is not used, all electron transition probabilities will be stored in the spin-up data files by default. Below is a sample of this output from this file.
# step probability accumulation
14 8.19528E-07 8.19528E-07
16 4.54580E-06 4.54580E-06
17 1.01866E-05 1.47324E-05
18 5.77833E-06 2.05108E-05
As the column titles suggest, the first column gives the step number at which there exists a finite probability, given by the second column, of the electron hopping from, in this case band index 29 to 32. Probabilities are only written for step numbers for which there is a finite (non-zero) probability. Thus, in the first 13 steps of the simulation there was zero probability for the electron to hop from band 29 to 32, as well as time step 15. However, at time steps 14, 16, 17, and 18, there were non-zero probabilities for this electronic transition. The accumulation column refers to ???
QM_hrt¶
This is a binary file that is required in the data directory to restart a QXMD simulation.
QM_ion¶
This is a binary file that is required in the data directory to restart a QXMD simulation, containing atomic position data for the system at the last step of the previous run.
qm_ion.d¶
This files gives the three components of the positions for all the atoms at each time step. It can be used for visualization of the atomic trajectories throughout the simulation. Below is sample output for monolayer MoSe2 with a total of 12 atoms.
# Atomic scaled coordinates
0 2 4 8
1.0000000E-01
1.66624 3.33424 2.49974 6.66751 3.33318 2.49997 1.66758 8.33529 2.50021
6.66486 8.33089 2.50009 3.33548 1.66849 1.46790 8.33121 1.66281 1.46807
3.33216 6.66685 1.46770 8.33431 6.66561 1.46808 3.33503 1.66953 3.53207
8.33384 1.66623 3.53173 3.33184 6.66746 3.53228 8.33294 6.66581 3.53215
1 2 4 8
1.0000000E-01
1.66036 3.33190 2.49817 6.67225 3.33212 2.50197 1.66794 8.33491 2.50111
6.66273 8.33275 2.50055 3.33569 1.66400 1.46680 8.33330 1.66428 1.46971
3.33378 6.66836 1.46589 8.33873 6.66766 1.46984 3.33989 1.67716 3.53255
8.32863 1.65779 3.53052 3.32835 6.66879 3.53167 8.33197 6.66709 3.53084
After the comment line, the next line gives the step number, the total number of atomic species in the system (‘2’ for molybdenum and selenium), the number of atoms of the species corresponding to keyword ‘1’ (in this case, keyword ‘1’ was used for molybdenum and there are ‘4’ molybdenum atoms), followed by the number of atoms of the species corresponding to keyword ‘2’ (in this case, keyword ‘2’ was used for selenium and there are ‘8’ selenium atoms). The number in the third line is a scaling factor for the following spatial coordinates of each atom (multiply all of the following spatial coordinates by this number to get the true atomic positions). The following lines give the x, y, and z coordinates of the first atom, followed by those for the second atom, and so on.
qm_log¶
This file provides of log of simulation details, including values of input parameters set, time statistics and energies computed for each SCF iteration for every simulation time step, as well as total computation time. This file is the best place to look to determine why a QXMD simulation fails/crashes, and general debugging.
qm_ovp.d¶
This file provides the overlap charge densities between atoms at the given time step intervals. Note that this file is only written if Mulliken analysis is set to true. Below is sample output for a monolayer of MoSe2 with 12 total atoms.
# Mulliken analysis : overlap population between atoms
0 2 4 8 -1
3 0 1 2
3 0 1 2
9
2 -0.0137 -0.0003 0.0171 -0.0002 -0.0983 -0.0017 0.0156 -0.0018 -0.2547
3 -0.0057 -0.0003 0.0163 -0.0002 -0.2605 0.0009 0.0143 0.0009 -0.0488
4 -0.0457 -0.0003 0.0207 -0.0003 -0.0089 0.0004 0.0232 0.0005 -0.0259
5 0.0041 0.0555 -0.0021 -0.0127 0.0859 -0.0138 -0.0024 0.2150 0.0596
6 0.0067 0.0514 -0.0028 -0.0130 0.1780 -0.0220 -0.0024 0.2105 0.0451
7 0.0045 0.0520 -0.0027 -0.0126 0.1058 -0.0041 -0.0024 0.2361 0.0196
9 0.0051 0.0562 -0.0024 -0.0125 0.0870 -0.0137 -0.0019 0.2140 0.0593
10 0.0050 0.0513 -0.0024 -0.0126 0.1794 -0.0219 -0.0025 0.2126 0.0452
11 0.0068 0.0523 -0.0026 -0.0128 0.1068 -0.0042 -0.0020 0.2331 0.0193
After the comment line, the next line gives the step number, the total number of atomic species in the system (‘2’ for molybdenum and selenium), the number of atoms of the species corresponding to keyword ‘1’ (in this case, keyword ‘1’ was used for molybdenum and there are ‘4’ molybdenum atoms), followed by the number of atoms of the species corresponding to keyword ‘2’ (in this case, keyword ‘2’ was used for selenium and there are ‘8’ selenium atoms). The ‘-1’ at the end of the line may be ignored as is it only included for backward compatibility with older versions of the code. The next set of two lines give the total number of angular momenta (orbital types) for each species, in this case both Mo and Se have s-type (‘0’), p-type (‘1’), and d-type (‘2’) orbitals. The ‘9’ in the next line signifies that the following lines will give partial charge overlaps between atoms number 9 and the other atoms. Lines 6-14 begin the the atom number, and then give the overlap between the all combinations of orbital types of atom 9 and all orbital types of the line’s respective atom number. Line 6, for example, gives the partial charge overlaps between atom 9 and atom 2. The 9 charge overlap numbers give overlaps between: s-type(atom 9)/s-type(atom 2), s-type(atom 9)/p-type(atom 2), s-type(atom 9)/d-type(atom 2), p-type(atom 9)/s-type(atom 2), p-type(atom 9)/p-type(atom 2), p-type(atom 9)/d-type(atom 2), d-type(atom 9)/s-type(atom 2), d-type(atom 9)/p-type(atom 2), d-type(atom 9)/d-type(atom 2).
QM_pcds¶
This is a binary file that is required in the data directory to restart a QXMD simulation.
qm_pds.d¶
This file gives the different angular momemta contributions to each energy band. Sample output for a water molecule is given below.
# Mulliken analysis : s, p, & d contribution to each band
0 2 10 -1
2 0 1
1 0
1 0.8239 0.1159 0.0602 1.0000
2 -0.0000 0.7977 0.2023 1.0000
3 0.1248 0.8752 0.0000 1.0000
4 0.0000 1.0000 0.0000 1.0000
5 0.0130 -0.0058 0.4393 0.4465
6 -0.0000 -0.0071 0.2595 0.2524
7 -0.0103 0.0017 0.1011 0.0925
8 0.0011 -0.0010 0.0401 0.0402
9 0.0099 -0.0033 0.0676 0.0742
10 0.0000 0.0306 0.0655 0.0961
After the comment line, the next line gives the step number, the total number of atomic species in the system (‘2’ for hydrogen and oxygen), and the total number energy bands, in this case ‘10’. The ‘-1’ refers to ***. The next two lines give the number of diffrent angular momenta (orbital types) for each atomic species. In this case, the species corresponding to keyword ‘1’, oxygen, has s-type ‘0’ and p-type ‘1’ orbitals, while the species corresponding to keyword ‘2’, hydrogen, has only s-type orbitals. The next 10 lines give the different angular momemta contributions to each of the energy bands. Looking at the 4th line, for example, the first number, ‘1’, represents the energy band index, the next three numbers give the contributions from oxygen s-type and p-type orbitals and hydrogen s-type orbitals. The final number is a sum of these three numbers. In this case, the sum is ‘1’ since the first energy band is comprised entirely of oxygen s-type and p-type orbitals and hydrogen s-type orbitals. However, higher energy bands may have some, even majority contribution from higher angular momenta, in which case the last number in the line will not be ‘1’.
QM_peig¶
This is a binary file that is required in the data directory to restart a QXMD simulation.
qm_str.d¶
This file outputs components of the stress tensor [GPa] of the system, computed quantum mechanically, at given time steps intervals (the interval is defined in IN.PARAM). This file will only be written if dumping stress data is set to true in IN.PARAM. Below is sample output.
# Stress in [GPa]
# Pxx Pyy Pzz Pyz Pzx Pxy
0 1.33128802E+01 1.31392351E+01 5.29606443E+00 -3.16079663E-03 5.31442065E-04 -9.35079164E-01
5 1.36386034E+01 1.33425150E+01 6.22908157E+00 -7.05568167E-03 1.18151950E-01 -1.04635711E+00
10 1.51764215E+01 1.56807679E+01 8.35762389E+00 2.34764126E-02 2.40917207E-01 -8.65032053E-01
15 1.68410502E+01 1.59107220E+01 1.10723229E+01 1.08462786E-01 4.08898975E-01 7.83659509E-01
20 1.78493502E+01 1.62776114E+01 1.32497640E+01 1.19138247E-01 4.21580509E-01 1.08288678E+00
After the first two comment lines, the following lines list the step number followed by the components of the stress tensor for that time step. Here, stress data is dumped every 5 time steps. Note that qm_str.d varies slightly from md_str.d due since md_str.d includes contributions from the kinetic energy of the ions, while qm_str.d does not.
QM_tddftfssh¶
This is a binary file that is required in the data directory to restart a Non-Adiabtic QXMD simulation, containing electronic transition probability data for the system at the last step of the previous run.
qm_td_eig.d¶
This file gives the enery eigenvalues and the electron occupation numbers for each energy band for each time step in a Non-Adiabtic QMD simulation. As such, this file is only written when TD-DFT is set to true. Sample output for a water molecule with two electrons excited from the highest occupied molecular orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO) is shown below.
# Eigenvalues of GS & occupations of Excited States
0 19 10
1 -2.12398E+00 2.000
2 -1.11576E+00 2.000
3 -6.58981E-01 2.000
4 -5.43674E-01 0.000
5 -5.55956E-02 2.000
6 1.62120E-01 0.000
7 1.63390E-01 0.000
8 1.87276E-01 0.000
9 2.13296E-01 0.000
10 3.57972E-01 0.000
After the comment line, the second line gives the step number, the cumulative number of SCF iterations performed up to that step number, and the number of energy bands in the system. The following lines give the energy band index number, the energy eigenvalue for that band, and the electronic occupation number for that band. In this case, spin polarization was not used, however, if it is turned on, there will be one column for spin-up electrons and one column for spin-down electrons. As can be seen, the lowest three energy bands are fully occupied, the fourth band had two electrons removed and placed in the fifth band, simulation an electronic excitation in a NAQMD simulation.
qm_zan.d¶
This file gives energy differences per electron and residuals for each time step. Note that the word “zansa” means residual in Japanese.
# Difference of tot E/el. E(HF)-E(KS) & maximum & average residuals
# difene difene2 zansa1 zansa2 bfzansa1 bfzansa2
0 19 1.4700E-07 3.0082E-05 2.0454E-05 2.1813E-08 7.7861E-06 1.7108E-08
1 32 8.4920E-07 5.7235E-05 5.8779E-05 3.3700E-08 5.3446E-05 1.5236E-08
2 42 9.8762E-07 2.2899E-04 2.5317E-05 4.9784E-08 3.4752E-06 1.0213E-08
3 52 8.5924E-07 1.1135E-04 3.0278E-05 2.1025E-08 2.3697E-05 1.2596E-08
After the two comment lines, each line gives the time step number, the total number of SCF iterations completed up to that time step, the difference in energy between the current and previous iteration upon convergence, the difference between the Harris-Foulkes and Kohn-Sham energies, the maximum (zansa1) and average (zansa2) residuals before the the Kohn-Sham equations are appoximately solved, and the maximum (bfzansa1) and average (bfzansa2) residuals after the the Kohn-Sham equations are appoximately solved.