Processing math: 100%

CP2K Open Source Molecular Dynamics

User Tools

Site Tools

Trace:
howto:gw

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
howto:gw [2021/04/21 20:51] – jwilhelmhowto:gw [2024/01/14 12:15] (current) – oschuett
Line 1: Line 1:
-====== GW method for computing electronic levels ====== +This page has been moved to: https://manual.cp2k.org/trunk/methods/properties/bandstructure_gw.html
- +
-The purpose of this section is to explain how to compute the energy of molecular orbitals/bands from GW for molecules/condensed phase systems with CP2K. In DFT, the energy of a molecular orbital corresponds to an eigenvalue of the Kohn-Sham matrix. In GW, the procedure for getting the level energies is to first perform a DFT calculation (commonly with the PBE or PBE0 functional) to get the molecular orbital wavefunctions and then compute a new GW energy for the molecular orbitals of interest. For an introduction into the concept of GW, please have a look at [[doi>10.3389/fchem.2019.00377]] or read Sec. II and the introduction to Sec. III in [[doi>10.1103/PhysRevB.87.235132]]. +
- +
-The GW implementation in CP2K is based on the developments described in [[doi>10.1021/acs.jctc.6b00380]]. The computational cost of GW is comparable to RPA and MP2 total energy calculations and therefore high. The computational cost of a canonical GW implementation grows as N4 with the system size N, while the memory scales as N3 with the system size. The basis set convergence of GW is slow and therefore has to be carefully examined. +
- +
-Since the calculations are rather small, please use a single MPI rank for the calculation: +
- +
-<code> +
-mpirun -n 1 cp2k.popt H2O_GW100.inp | tee cp2k.out +
-</code> +
- +
-===== 1. Reproducing values from the GW100 set ===== +
-See below the input for a G0W0@PBE calculation of the water molecule in a def2-QZVP basis: A PBE calculation is used for computing the molecular orbitals which can be seen from the keyword "XC_FUNCTIONAL PBE". The input parameters for G0W0 are commented below. While the calculation is running, you can look up the G0W0@PBE value for the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) in [[doi>10.1021/acs.jctc.5b00453]] [Table 2 and 3 (column AIMS-P16/TM-no RI, molecule index 76) which should be -11.97 eV and 2.37 eV for HOMO and LUMO, respectively]. CP2K should be able to exactly reproduce these values. In the output of CP2K, the G0W0@PBE results are listed after the SCF after the headline //GW quasiparticle energies//.   +
- +
-For checking the basis set convergence, we refer to a detailed analysis in [[doi>10.1021/acs.jctc.6b00380]] in Fig. 2 for benzene. An extensive table of basis set extrapolated GW levels can be found in [[doi>10.1021/acs.jctc.5b00453]] (column EXTRA). +
-<code - H2O_GW100.inp> +
-&FORCE_EVAL +
-  METHOD Quickstep +
-  &DFT +
-    BASIS_SET_FILE_NAME BASIS_def2_QZVP_RI_ALL +
-    POTENTIAL_FILE_NAME POTENTIAL +
-    &MGRID +
-      CUTOFF 400 +
-      REL_CUTOFF 50 +
-    &END MGRID +
-    &QS +
-      ! all electron calculation since GW100 is all-electron test +
-      METHOD GAPW +
-    &END QS +
-    &POISSON +
-      PERIODIC NONE +
-      PSOLVER MT +
-    &END +
-    &SCF +
-      EPS_SCF 1.0E-6 +
-      SCF_GUESS ATOMIC +
-      MAX_SCF 200 +
-    &END SCF +
-    &XC +
-      &XC_FUNCTIONAL PBE +
-      &END XC_FUNCTIONAL +
-      ! GW is part of the WF_CORRELATION section +
-      &WF_CORRELATION +
-        &RI_RPA +
-          ! use 100 points to perform the frequency integration in GW +
-          QUADRATURE_POINTS 100 +
-          ! SIZE_FREQ_INTEG_GROUP is a group size for parallelization and +
-          ! should be increased for large calculations to prevent out of memory. +
-          ! maximum for SIZE_FREQ_INTEG_GROUP is the number of MPI tasks +
-          &GW +
-           ! compute the G0W0@PBE energy of HOMO-9, +
-           ! HOMO-8, ... , HOMO-1, HOMO +
-           CORR_OCC   10 +
-           ! compute the G0W0@PBE energy of LUMO, +
-           ! LUMO+1, ... , LUMO+20 +
-           CORR_VIRT  20 +
-           ! use the RI approximation for the exchange part of the self-energy +
-           RI_SIGMA_X +
-          &END GW +
-        &END RI_RPA +
-      &END +
-    &END XC +
-  &END DFT +
-  &SUBSYS +
-    &CELL +
-      ABC 10.0 10.0 10.0 +
-      PERIODIC NONE +
-    &END CELL +
-    &COORD +
-      O  0.0000 0.0000 0.0000 +
-      H  0.7571 0.0000 0.5861 +
-      H -0.7571 0.0000 0.5861 +
-    &END COORD +
-    &TOPOLOGY +
-      &CENTER_COORDINATES +
-      &END +
-    &END TOPOLOGY +
-    &KIND H +
-      ! def2-QZVP is the basis which has been used in the GW100 paper +
-      BASIS_SET        def2-QZVP +
-      ! just use a very large RI basis to ensure excellent +
-      ! convergence with respect to the RI basis +
-      BASIS_SET RI_AUX RI-5Z +
-      POTENTIAL ALL +
-    &END KIND +
-    &KIND O +
-      BASIS_SET        def2-QZVP +
-      BASIS_SET RI_AUX RI-5Z +
-      POTENTIAL ALL +
-    &END KIND +
-  &END SUBSYS +
-&END FORCE_EVAL +
-&GLOBAL +
-  RUN_TYPE     ENERGY +
-  PROJECT      ALL_ELEC +
-  PRINT_LEVEL  MEDIUM +
-&END GLOBAL +
- +
-</code> +
- +
-===== 2. Basis set extrapolation ===== +
- +
-In this section, the slow basis set convergence of GW calculations is examined.  +
-We compute the G0W0@PBE HOMO and LUMO level of the water molecule with Dunning's cc-DZVP, cc-TZVP, cc-QZVP and cc-5ZVP all-electron basis sets and extrapolate these values to the complete basis set limit.  +
-To do so, download the cc basis sets {{exercises:2017_uzh_cp2k-tutorial:cc_basis_h2o.tar ?direct&400 |}} which has been taken from the EMSL basis set database. Run the input from Sec. 1 using the cc-DZVP to cc-5ZVP basis set (in total four calculations) by replacing the basis sets: +
-<code> +
-BASIS_SET_FILE_NAME BASIS_def2_QZVP_RI_ALL +
-BASIS_SET_FILE_NAME ./BASIS_H2O +
-</code> +
-<code> +
-&KIND H +
-  BASIS_SET        cc-DZVP-all +
-  BASIS_SET RI_AUX RI-5Z +
-  POTENTIAL ALL +
-&END KIND +
-&KIND O +
-  BASIS_SET        cc-DZVP-all +
-  BASIS_SET RI_AUX RI-5Z +
-  POTENTIAL ALL +
-&END KIND +
-</code> +
- +
-Employ the RI-5Z basis set as RI-basis which ensures excellent convergence for the RI basis.  +
-In practice, smaller RI basis sets can be used from the EMSL database (just check the convergence with respect to the RI basis by using smaller and larger RI basis sets).  +
- +
-The results for the G0W0@PBE HOMO and LUMO from CP2K should be as follows: +
- +
-<code> +
-Basis set   G0W0@PBE HOMO (eV)   G0W0@PBE LUMO (eV)   N_basis   N_card +
-cc-DZVP         -12.480              4.770               23        2 +
-cc-TZVP         -12.417              3.424               57        3 +
-cc-QZVP         -12.180              2.773              114        4 +
-cc-5ZVP         -12.108              2.088              200        5 +
- +
-Extrapolation using cc-TZVP to cc-5ZVP +
-with 1/N_card^3  -12.02 +/- 0.01       1.90 +/- 0.29 +
-with 1/N_basis   -11.97 +/- 0.02       1.71 +/- 0.29 +
-GW100            -12.05                2.01 +
-</code> +
- +
-For the extrapolation, two schemes have been used as described in the GW100 paper and its supporting information [[doi>10.1021/acs.jctc.5b00453]].  +
-The first scheme employs a linear fit on the HOMO or LUMO values when they are plotted against the inverse cardinal number Ncard of the basis set while the second scheme extrapolates versus the inverse number of basis functions Nbasis which can be computed as sum of the number of occupied orbitals and the number of virtual orbitals as printed in RI_INFO in the output.  +
-You can check the extrapolation from the table above with your tool of choice. +
- +
-The basis set extrapolated values from the table above deviate from the values reported in the GW100 paper [[doi>10.1021/acs.jctc.5b00453]], probably because only two basis sets (def2-TZVP, def2-QZVP) have been used in [[doi>10.1021/acs.jctc.5b00453]] for the extrapolation. The extrapolation for the LUMO is not working well because one would need much more diffuse functions to represent unbound electronic levels (with positive energy).  +
- +
-Often, the HOMO-LUMO gap is of interest. In this case, augmented basis sets (e.g. from the EMSL database) can offer an alternative for very fast basis set convergence, see also Fig. 2b in [[doi>10.1021/acs.jctc.6b00380]]. +
-===== 3. Input for large-scale calculations ===== +
-An exemplary input for a parallel calculation can be found in the supporting information of [[doi>10.1021/acs.jctc.6b00380]] [2]. The emphasis is on the parameters SIZE_FREQ_INTEG_GROUP and NUMBER_PROC which should be increased for larger calculations. In case of a too small number, the code will crash due to out of memory while a too large number results in slow speed. Typically, one starts for large-scale calculations from a small molecule. When increasing the system size, the parameters SIZE_FREQ_INTEG_GROUP and NUMBER_PROC should be both increased to avoid a crash due to out of memory. The maximum number for both parameters is the number of MPI tasks. Also, the number of nodes should be increased with N3atoms to account for the scaling of the memory of GW.  +
- +
-===== 4. Self-consistent GW calculations and DFT starting point ===== +
- +
-The G0W0@PBE HOMO value of the H2O molecule (~ -12.0 eV) is not in good agreement with the experimental ionization potential (12.62 eV). Benchmarks on molecules and solids indicate that self-consistency of eigenvalues in the Green's function G improves the agreement between the GW calculation and experiment. This scheme is called GW0@PBE and is our favorite GW flavor so far (experience based on nano-sized aromatic molecules). You can also check [[https://www.youtube.com/watch?v=1vUuethWhbs&t=5563s|this video]] or [[doi>10.3389/fchem.2019.00377]] on which DFT starting functional and which self-consistency scheme could be good for your system. +
- +
-You can run GW0 calculations in CP2K by putting +
-<code> +
-&GW +
-  SC_GW0_ITER  10 +
-  CORR_OCC     10 +
-  CORR_VIRT    20 +
-  RI_SIGMA_X +
-&END GW +
-</code> +
-"SC_GW0_ITER 10" means that at most ten iterations in the eigenvalues of G are performed. In case convergence is found, the code terminates earlier. "SC_GW0_ITER 1" corresponds to G0W0. +
- +
-===== 5. Periodic GW calculations ===== +
-For periodic GW calculations, a special correction scheme is necessary. A similar problem is appearing in Hartree-Fock calculations. In HF, an easy way out is given by the truncation of the Coulomb operator which works due to the convenient form of the HF equations. GW does not exhibit this convenient form and therefore, this truncation does not work for GW calculations. The theory why a correction is necessary and the correction scheme is described in [[doi>10.1103/PhysRevB.95.235123]]. +
- +
-The basis can be found in {{exercises:2017_uzh_cp2k-tutorial:LiH_BASIS_RI.tar ?direct&400 |}}. Then run the calculation as listed below. The computed gap of LiH with the input from below should be 6.05 eV which is still significantly off from the converged result of 4.7 eV from [[doi>10.1103/PhysRevB.95.235123]]. Here, a basis set extrapolation (as shown above) and a larger supercell are necessary to get closer to the result of 4.7 eV. Please redo the calculation without the flag PERIODIC in the GW section and see that the resulting gap of 11.25 eV is much more off than the gap with correction.  +
- +
-<code - LiH_periodic.inp> +
-&FORCE_EVAL +
-  METHOD Quickstep +
-  &DFT +
-    BASIS_SET_FILE_NAME ./LiH_BASIS_RI +
-    POTENTIAL_FILE_NAME POTENTIAL +
-    &MGRID +
-      CUTOFF 600 +
-      REL_CUTOFF 60 +
-    &END MGRID +
-    &QS +
-      METHOD GPW +
-      EPS_DEFAULT 1.0E-15 +
-      EPS_PGF_ORB 1.0E-20 +
-      EPS_FILTER_MATRIX 0.0e0 +
-    &END QS +
-    &SCF +
-      EPS_SCF 1.0E-6 +
-      MAX_SCF 100 +
-    &END SCF +
-    &XC +
-      &XC_FUNCTIONAL PBE +
-      &END XC_FUNCTIONAL +
-      &WF_CORRELATION +
-        &RI_RPA +
-          QUADRATURE_POINTS  100 +
-          &GW +
-           CORR_OCC  5 +
-           CORR_VIRT 5 +
-           ! activate the periodic correction +
-           PERIODIC +
-          &END GW +
-          ! HF calculation for the exchange part of the self-energy +
-          ! Here, the truncation of the Coulomb operator works  +
-          &HF +
-            &SCREENING +
-              ! for other materials, a smaller EPS_SCHWARZ might be necessary +
-              EPS_SCHWARZ 1.0E-6 +
-              SCREEN_ON_INITIAL_P TRUE +
-            &END +
-            &INTERACTION_POTENTIAL +
-              POTENTIAL_TYPE TRUNCATED +
-              ! the truncation radius is half the cell size +
-              CUTOFF_RADIUS  2.00  +
-              T_C_G_DATA t_c_g.dat +
-            &END +
-            &MEMORY +
-              MAX_MEMORY  0 +
-            &END +
-          &END +
-        &END RI_RPA +
-        NUMBER_PROC  1 +
-      &END +
-    &END XC +
-  &END DFT +
-  &SUBSYS +
-    &CELL +
-      ABC  4.084 4.084 4.084 +
-    &END CELL +
-    &COORD +
-      Li 0 0 0 +
-      Li 2.042 2.042 0 +
-      Li 2.042 0 2.042 +
-      Li 0 2.042 2.042 +
-      H 0 2.042 0 +
-      H 0 0 2.042 +
-      H 2.042 0 0 +
-      H 2.042 2.042 2.042 +
-    &END COORD +
-    &KIND H +
-      BASIS_SET cc-DZVP-GTH +
-      BASIS_SET RI_AUX RI_DZ_opt_basis +
-      POTENTIAL GTH-PBE-q1 +
-    &END KIND +
-    &KIND Li +
-      BASIS_SET cc-DZVP-GTH +
-      BASIS_SET RI_AUX RI_DZ_opt_basis +
-      POTENTIAL GTH-PBE-q3 +
-    &END KIND +
-  &END SUBSYS +
-&END FORCE_EVAL +
-&GLOBAL +
-  PROJECT     LiH_bulk_2x2x2_DZVP +
-  PRINT_LEVEL MEDIUM +
-  RUN_TYPE ENERGY +
-&END GLOBAL +
-</code>+
howto/gw.1619038261.txt.gz · Last modified: by jwilhelm