Section LS_SCF

• Specifies the parameters of the linear scaling SCF routines

• Section path: CP2K_INPUT / FORCE_EVAL / DFT / LS_SCF

• This section cannot be repeated.

• This section cites the following reference: [VandeVondele2012]

• Subsections

  • CHEBYSHEV
  • CURVY_STEPS
  • PAO
  • PEXSI
  • RHO_MIXING

• Keywords

  • CHECK_S_INV
  • DYNAMIC_THRESHOLD
  • EPS_DIIS
  • EPS_FILTER
  • EPS_LANCZOS
  • EPS_SCF
  • EXTRAPOLATION_ORDER
  • FIXED_MU
  • INI_DIIS
  • LS_DIIS
  • MATRIX_CLUSTER_TYPE
  • MAX_DIIS
  • MAX_ITER_LANCZOS
  • MAX_SCF
  • MIXING_FRACTION
  • MU
  • NMIXING
  • NON_MONOTONIC
  • PERFORM_MU_SCAN
  • PURIFICATION_METHOD
  • REPORT_ALL_SPARSITIES
  • RESTART_READ
  • RESTART_WRITE
  • SIGN_METHOD
  • SIGN_ORDER
  • SIGN_SYMMETRIC
  • SINGLE_PRECISION_MATRICES
  • SUBMATRIX_SIGN_METHOD
  • S_INVERSION
  • S_PRECONDITIONER
  • S_SQRT_METHOD
  • S_SQRT_ORDER

• Keyword descriptions

  CHECK_S_INV
  	CHECK_S_INV {Logical}
  	Perform an accuracy check on the inverse/sqrt of the s matrix.
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .FALSE.

  DYNAMIC_THRESHOLD
  	DYNAMIC_THRESHOLD {Logical}
  	Should the threshold for the purification be chosen dynamically
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .FALSE.

  EPS_DIIS
  	EPS_DIIS {Real}
  	Threshold on the convergence to start using DIIS
  	This keyword cannot be repeated and it expects precisely one real.
  	Default value: 1.00000000E-001

  EPS_FILTER
  	EPS_FILTER {Real}
  	Threshold used for filtering matrix operations.
  	This keyword cannot be repeated and it expects precisely one real.
  	Default value: 1.00000000E-006

  EPS_LANCZOS
  	EPS_LANCZOS {Real}
  	Threshold used for lanczos estimates.
  	This keyword cannot be repeated and it expects precisely one real.
  	Default value: 1.00000000E-003

  EPS_SCF
  	EPS_SCF {Real}
  	Target accuracy for the SCF convergence in terms of change of the total energy per electron.
  	This keyword cannot be repeated and it expects precisely one real.
  	Default value: 1.00000000E-007

  EXTRAPOLATION_ORDER
  	EXTRAPOLATION_ORDER {Integer}
  	Number of previous matrices used for the ASPC extrapolation of the initial guess. 0 implies that an atomic guess is used at each step. low (1-2) will result in a drift of the constant of motion during MD. high (>5) might be somewhat unstable, leading to more SCF iterations.
  	This keyword cannot be repeated and it expects precisely one integer.
  	Default value: 4

  FIXED_MU
  	FIXED_MU {Logical}
  	Should the calculation be performed at fixed chemical potential, or should it be found fixing the number of electrons
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .FALSE.

  INI_DIIS
  	INI_DIIS {Integer}
  	Iteration cycle to start DIIS Kohn-Sham matrix update
  	This keyword cannot be repeated and it expects precisely one integer.
  	Default value: 2

  LS_DIIS
  	LS_DIIS {Logical}
  	Perform DIIS within linear scaling SCF
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .FALSE.

  MATRIX_CLUSTER_TYPE
  	MATRIX_CLUSTER_TYPE {Keyword}
  	Specify how atomic blocks should be clustered in the used matrices, in order to improve flop rate, and possibly speedup the matrix multiply. Note that the atomic s_preconditioner can not be used.Furthermore, since screening is on matrix blocks, slightly more accurate results can be expected with molecular.
  	This keyword cannot be repeated and it expects precisely one keyword.
  	Default value: ATOMIC
  	List of valid keywords: ATOMIC Using atomic blocks MOLECULAR Using molecular blocks.

  MAX_DIIS
  	MAX_DIIS {Integer}
  	Size of LS_DIIS buffer
  	This keyword cannot be repeated and it expects precisely one integer.
  	Default value: 4

  MAX_ITER_LANCZOS
  	MAX_ITER_LANCZOS {Integer}
  	Maximum number of lanczos iterations.
  	This keyword cannot be repeated and it expects precisely one integer.
  	Default value: 128

  MAX_SCF
  	MAX_SCF {Integer}
  	Maximum number of SCF iteration to be performed for one optimization
  	This keyword cannot be repeated and it expects precisely one integer.
  	Default value: 20

  MIXING_FRACTION
  	MIXING_FRACTION {Real}
  	Mixing density matrices uses the specified fraction in the SCF procedure.
  	This keyword cannot be repeated and it expects precisely one real.
  	Default value: 4.50000000E-001

  MU
  	MU {Real}
  	Value (or initial guess) for the chemical potential, i.e. some suitable energy between HOMO and LUMO energy.
  	This keyword cannot be repeated and it expects precisely one real.
  	Default value: -1.00000000E-001

  NMIXING
  	NMIXING {Integer}
  	Minimal number of density mixing before start DIIS
  	This keyword cannot be repeated and it expects precisely one integer.
  	Default value: 2

  NON_MONOTONIC
  	NON_MONOTONIC {Logical}
  	Should the purification be performed non-monotonically. Relevant for TC2 only.
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .TRUE.

  PERFORM_MU_SCAN
  	PERFORM_MU_SCAN {Logical}
  	Do a scan of the chemical potential after the SCF
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .FALSE.

  PURIFICATION_METHOD
  	PURIFICATION_METHOD {Keyword}
  	Scheme used to purify the Kohn-Sham matrix into the density matrix.
  	This keyword cannot be repeated and it expects precisely one keyword.
  	Default value: SIGN
  	List of valid keywords: PEXSI PEXSI method SIGN Sign matrix iteration. TC2 Trace conserving 2nd order scheme TRS4 Trace resetting 4th order scheme
  	This keyword cites the following references: [Niklasson2003] [VandeVondele2012]

  REPORT_ALL_SPARSITIES
  	REPORT_ALL_SPARSITIES {Logical}
  	Run the sparsity report at the end of the SCF
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .TRUE.

  RESTART_READ
  	RESTART_READ {Logical}
  	Read the density matrix before the (first) SCF.
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .FALSE.

  RESTART_WRITE
  	RESTART_WRITE {Logical}
  	Write the density matrix at the end of the SCF (currently requires EXTRAPOLATION_ORDER>0). Files might be rather large.
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .FALSE.

  SIGN_METHOD
  	SIGN_METHOD {Keyword}
  	Method used for the computation of the sign matrix.
  	This keyword cannot be repeated and it expects precisely one keyword.
  	Default value: NEWTONSCHULZ
  	List of valid keywords: NEWTONSCHULZ Newton-Schulz iteration. PROOT p-th order root iteration SUBMATRIX Submatrix method
  	This keyword cites the following references: [Niklasson2003] [VandeVondele2012]

  SIGN_ORDER
  	SIGN_ORDER {Integer}
  	Order of the method used for the computation of the sign matrix.
  	This keyword cannot be repeated and it expects precisely one integer.
  	Default value: 2

  SIGN_SYMMETRIC
  	SIGN_SYMMETRIC {Logical}
  	Use symmetric orthogonalization when generating the input for the sign function.
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .FALSE.

  SINGLE_PRECISION_MATRICES
  	SINGLE_PRECISION_MATRICES {Logical}
  	Matrices used within the LS code can be either double or single precision.
  	This keyword cannot be repeated and it expects precisely one logical.
  	The lone keyword behaves as a switch to .TRUE.
  	Default value: .FALSE.

  SUBMATRIX_SIGN_METHOD
  	SUBMATRIX_SIGN_METHOD {Keyword}
  	Method used for the computation of the sign matrix of all submatrices.
  	This keyword cannot be repeated and it expects precisely one keyword.
  	Default value: NEWTONSCHULZ
  	List of valid keywords: DIRECT Direct method calculating all eigenvalues. DIRECT_MUADJ Direct method with internal adjustment of mu DIRECT_MUADJ_LOWMEM Direct method with internal adjustment of mu, using two passes to save memory NEWTONSCHULZ Newton-Schulz iteration.

  S_INVERSION
  	S_INVERSION {Keyword}
  	Method used to compute the inverse of S.
  	This keyword cannot be repeated and it expects precisely one keyword.
  	Default value: SIGN_SQRT
  	List of valid keywords: HOTELLING Using the Hotellign iteration. SIGN_SQRT Using the inverse sqrt as obtained from sign function iterations.

  S_PRECONDITIONER
  	S_PRECONDITIONER {Keyword}
  	Preconditions S with some appropriate form.
  	This keyword cannot be repeated and it expects precisely one keyword.
  	Default value: ATOMIC
  	List of valid keywords: ATOMIC Using atomic blocks MOLECULAR Using molecular sub-blocks. Recommended if molecules are defined and not too large. NONE No preconditioner

  S_SQRT_METHOD
  	S_SQRT_METHOD {Keyword}
  	Method for the caclulation of the sqrt of S.
  	This keyword cannot be repeated and it expects precisely one keyword.
  	Default value: NEWTONSCHULZ
  	List of valid keywords: NEWTONSCHULZ Using a Newton-Schulz-like iteration PROOT Using the p-th root method.

  S_SQRT_ORDER
  	S_SQRT_ORDER {Integer}
  	Order of the iteration method for the calculation of the sqrt of S.
  	This keyword cannot be repeated and it expects precisely one integer.
  	Default value: 3
  	Alias names for this keyword: SIGN_SQRT_ORDER