Problem of calculating melt structure with cp2k ab initio calculation method

Sun Yiwei

unread,

Aug 28, 2021, 3:09:04 AM8/28/21

to cp2k

Hello everyone,

At present, based on the ab initio calculation of the cp2k running silicate melt structure, the atoms in the initial structure are randomly distributed, and there are a total of 266 atoms. Through the xyz file, it is found that the system continues to expand during the running process, and the distance between the atoms becomes very large. The structure is obviously unreasonable, as shown below. Please help to see where the problem is. Thank you all in advance!

@SET SYSNAME CaSiOB

@SET CELL 16.3258

&GLOBAL

PRINT_LEVEL MEDIUM

PROJECT_NAME ${SYSNAME}

RUN_TYPE MD

&END GLOBAL

&FORCE_EVAL

METHOD QS

STRESS_TENSOR DIAGONAL_ANALYTICAL

&DFT

BASIS_SET_FILE_NAME BASIS_MOLOPT

POTENTIAL_FILE_NAME GTH_POTENTIALS

&SCF

MAX_SCF 200

EPS_SCF 1.0e-5

SCF_GUESS ATOMIC

&MIXING T

METHOD PULAY_MIXING

ALPHA 0.5

NBUFFER 5

&END MIXING

&PRINT

&RESTART OFF

&END RESTART

&END PRINT

&OT ON

PRECONDITIONER FULL_SINGLE_INVERSE

MINIMIZER DIIS

&END OT

&END SCF

&QS

EPS_DEFAULT 1e-10

&END QS

&MGRID

NGRIDS 4

CUTOFF 400

REL_CUTOFF 40

&END MGRID

&XC

&XC_FUNCTIONAL PBE

&END XC_FUNCTIONAL

DENSITY_CUTOFF 1e-10

GRADIENT_CUTOFF 1e-10

TAU_CUTOFF 1e-10

&VDW_POTENTIAL

POTENTIAL_TYPE PAIR_POTENTIAL

&PAIR_POTENTIAL

R_CUTOFF 9

TYPE DFTD3

PARAMETER_FILE_NAME dftd3.dat

REFERENCE_FUNCTIONAL PBE

LONG_RANGE_CORRECTION T

&END PAIR_POTENTIAL

&END VDW_POTENTIAL

&END XC

&END DFT

&SUBSYS

&TOPOLOGY

COORD_FILE ${SYSNAME}.xyz

COORD_FILE_FORMAT XYZ

CONN_FILE_FORMAT OFF

&END TOPOLOGY

&CELL

ABC ${CELL} ${CELL} ${CELL}

ALPHA_BETA_GAMMA 90.0 90.0 90.0

MULTIPLE_UNIT_CELL 1 1 1

PERIODIC XYZ

&END CELL

&KIND B

BASIS_SET DZVP-MOLOPT-SR-GTH

POTENTIAL GTH-PBE-q3

&END KIND

&KIND O

BASIS_SET DZVP-MOLOPT-SR-GTH

POTENTIAL GTH-PBE-q6

&END KIND

&KIND Si

BASIS_SET DZVP-MOLOPT-SR-GTH

POTENTIAL GTH-PBE-q4

&END KIND

&KIND Ca

BASIS_SET DZVP-MOLOPT-SR-GTH

POTENTIAL GTH-PBE-q10

&END KIND

&END SUBSYS

&END FORCE_EVAL

&MOTION

&MD

ENSEMBLE NVT

&THERMOSTAT

TYPE CSVR

&CSVR

TIMECON 1000

&END CSVR

&END THERMOSTAT

STEPS 5000

TEMPERATURE 1823

TIMESTEP 2.0

&END MD

&PRINT

&RESTART

FILENAME =${SYSNAME}.rst

&EACH

MD 10

&END EACH

&END RESTART

&STRESS

FILENAME =${SYSNAME}.stress

&EACH

MD 10

&END EACH

&END STRESS

&TRAJECTORY

FILENAME =${SYSNAME}.pos.xyz

FORMAT XYZ

UNIT angstrom

&EACH

MD 10

&END EACH

&END TRAJECTORY

&VELOCITIES

FILENAME =${SYSNAME}.vel.xyz

FORMAT XYZ

UNIT angstrom/fs

&EACH

MD 10

&END EACH

&END VELOCITIES

&FORCES

FILENAME =${SYSNAME}.frc.xyz

FORMAT XYZ

UNIT amu*angstrom/fs^2

&EACH

MD 10

&END EACH

&END FORCES

&CELL

FILENAME =${SYSNAME}.cell

&EACH

MD 10

&END EACH

&END CELL

&END PRINT

&END MOTION

Lucas Lodeiro

unread,

Aug 28, 2021, 2:23:40 PM8/28/21

to cp...@googlegroups.com

are you using a NVT ensemble and your cell volume is changing? Maybe your "cell" variable which defines the a, b and c length vectors is so big... I mean, you define a cubic cell of 16.3258 A, and the smaller cell image is not cubic... probably your are not visualizing the real shape of the box...

regards

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/bf2626ca-f339-42c3-b1f1-437995c5663bn%40googlegroups.com.

hut...@chem.uzh.ch

unread,

Aug 30, 2021, 4:57:29 AM8/30/21

to cp...@googlegroups.com

Hi

I don't know about your structure, but I would also make the
following changes to your input file:

Stress tensor : ANALYTICAL

&PAIR_POTENTIAL --> use a larger CUTOFF (e.g.15 or 18)
disable LONG_RANGE_CORRECTION (see manual)
disable CALCULATE_C9_TERM

Use a higher REL_CUTOFF e.g. 60
Use a &REF_CELL in &CELL to define your target volume. This is important
to assure no degradation in XC energy integration.

regards

Juerg Hutter
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie C FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: "cp2k" <cp...@googlegroups.com>
From: "Sun Yiwei"
Sent by: cp...@googlegroups.com
Date: 08/28/2021 09:09AM
Subject: [CP2K:15799] Problem of calculating melt structure with cp2k ab initio calculation method

Hello everyone,
At present, based on the ab initio calculation of the cp2k running silicate melt structure, the atoms in the initial structure are randomly distributed, and there are a total of 266 atoms. Through the xyz file, it is found that the system continues to expand during the running process, and the distance between the atoms becomes very large. The structure is obviously unreasonable, as shown below. Please help to see where the problem is. Thank you all in advance!

Sun Yiwei

unread,

Sep 7, 2021, 4:10:29 AM9/7/21

to cp2k

Thank you very much for your reply. I modified my inp file according to Hutter's suggestion, and the balanced structure obtained by simulation is greatly improved compared to before. However, there are still some unreasonable places in the obtained structure, such as the existence of free Si atoms and O atoms. I described my problem in a new post（The website is https://groups.google.com/g/cp2k/c/DSMHbzqAUTc）, I hope you can give some suggestions. Thank you all!

Reply all

Reply to author

Forward