新建
快速开始 ABACUS｜自洽 能带 态密度 结构优化
MileAway
ABACUS 兴趣小组
推荐镜像 :ABACUS:3.2.0-toolkit-notebook
推荐机型 :c32_m128_cpu
赞 19
27
目录
快速开始 ABACUS｜自洽 能带 态密度 结构优化
代码
文本
代码
文本
©️ Copyright 2023 @ Authors
作者：赵天琦 📨
日期：2023-04-01
共享协议：本作品采用知识共享署名-非商业性使用-相同方式共享 4.0 国际许可协议进行许可。
代码
文本
🎯 本教程旨在快速掌握使用 ABACUS 进行第一性原理材料模拟计算。
一键运行，你可以快速在实践中检验你的想法。
丰富完善的注释，对于入门者友好。
代码
文本
在 Bohrium Notebook 界面，你可以点击界面上方蓝色按钮 开始连接，选择 abacus:3.2.0-toolkit-notebook 镜像及任何一款节点配置，稍等片刻即可运行。
代码
文本
目标
上手使用ABACUS做第一性原理材料模拟计算
在学习本教程后，你将能够：
- 使用ABACUS做自洽计算
- 计算得到电子能带结构图
- 计算得到电子态密度图
- 使用ABACUS做结构优化计算
阅读该教程【最多】约需 15 分钟，让我们开始吧！
代码
文本
代码
文本
代码
文本
自洽，是自洽场计算(SCF, self-consistent field calculation)的简称，它是密度泛函理论（DFT）算法的基础，也是DFT被称为第一性原理方法的原因。
从自洽场计算中我们可以得到一个体系的基态结构和基态能量。
基态结构使得我们可以得到体系的结构参数，例如晶胞参数，化学键键长键角等，而基态能量使得我们可以进一步得到原子受力和晶体材料的应力-应变性质等。
因此自洽场计算是开始做DFT计算的基础。
代码
文本
这里我们在gitee上为大家准备好了快速上手的案例，大家可以直接使用git命令来下载。
代码
文本
[1]
! git clone https://gitee.com/deepmodeling/colombo-ABACUS.git
Cloning into 'colombo-ABACUS'... remote: Enumerating objects: 300, done. remote: Counting objects: 100% (118/118), done. remote: Compressing objects: 100% (81/81), done. remote: Total 300 (delta 29), reused 98 (delta 19), pack-reused 182 Receiving objects: 100% (300/300), 3.79 MiB | 821.00 KiB/s, done. Resolving deltas: 100% (77/77), done. Updating files: 100% (160/160), done.
代码
文本
注意：该临时文件将在本案例末执行删除操作。
代码
文本
代码
文本
[2]
cd colombo-ABACUS/ABACUS/MgO_LCAO/SCF
/data/bohrium-notebook/Proem/colombo-ABACUS/ABACUS/MgO_LCAO/SCF
代码
文本
[3]
cat INPUT
INPUT_PARAMETERS suffix MgO ntype 2 pseudo_dir ../../PP_ORB orbital_dir ../../PP_ORB ecutwfc 100 scf_thr 1e-6 basis_type lcao calculation scf
代码
文本
[4]
cat KPT
K_POINTS 0 Gamma 4 4 4 0 0 0
代码
文本
[5]
cat STRU
ATOMIC_SPECIES Mg 24.305 Mg_ONCV_PBE-1.0.upf O 15.999 O_ONCV_PBE-1.0.upf NUMERICAL_ORBITAL Mg_gga_8au_100Ry_4s2p1d.orb O_gga_8au_100Ry_2s2p1d.orb LATTICE_CONSTANT 1.8897259886 # 1.8897259886 Bohr = 1.0 Angstrom LATTICE_VECTORS 4.27957 0.00000 0.00000 0.00000 4.27957 0.00000 0.00000 0.00000 4.27957 ATOMIC_POSITIONS Direct Mg 0.0 4 0.0 0.0 0.0 0 0 0 0.0 0.5 0.5 0 0 0 0.5 0.0 0.5 0 0 0 0.5 0.5 0.0 0 0 0 O 0.0 4 0.5 0.0 0.0 0 0 0 0.5 0.5 0.5 0 0 0 0.0 0.0 0.5 0 0 0 0.0 0.5 0.0 0 0 0
代码
文本
[6]
!. /opt/intel/oneapi/setvars.sh && OMP_NUM_THREADS=1 abacus
:: initializing oneAPI environment ...
dash: SH_VERSION = unknown
args: Using "$@" for setvars.sh arguments:
:: compiler -- latest
:: debugger -- latest
:: dev-utilities -- latest
:: mkl -- latest
:: mpi -- latest
:: tbb -- latest
:: vtune -- latest
:: oneAPI environment initialized ::
WARNING: Total thread number on this node mismatches with hardware availability. This may cause poor performance.
Info: Local MPI proc number: 1,OpenMP thread number: 1,Total thread number: 1,Local thread limit: 52
ABACUS v3.1
Atomic-orbital Based Ab-initio Computation at UStc
Website: http://abacus.ustc.edu.cn/
Documentation: https://abacus.deepmodeling.com/
Repository: https://github.com/abacusmodeling/abacus-develop
https://github.com/deepmodeling/abacus-develop
Mon Apr 3 15:12:21 2023
MAKE THE DIR : OUT.MgO/
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Warning: number valence electrons > 2 for Mg: [Ne] 3s2
Please make sure the pseudopotential file is what you need
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
UNIFORM GRID DIM : 54 * 54 * 54
UNIFORM GRID DIM(BIG): 18 * 18 * 18
DONE(0.083602 SEC) : SETUP UNITCELL
DONE(0.0944471 SEC) : INIT K-POINTS
---------------------------------------------------------
Self-consistent calculations for electrons
---------------------------------------------------------
SPIN KPOINTS PROCESSORS NBASE
1 36 1 112
---------------------------------------------------------
Use Systematically Improvable Atomic bases
---------------------------------------------------------
ELEMENT ORBITALS NBASE NATOM XC
Mg 4s2p1d-8au 15 4
O 2s2p1d-8au 13 4
---------------------------------------------------------
Initial plane wave basis and FFT box
---------------------------------------------------------
-------------------------------------------
SELF-CONSISTENT :
-------------------------------------------
START CHARGE : atomic
DONE(2.59466 SEC) : INIT SCF
ITER ETOT(eV) EDIFF(eV) DRHO TIME(s)
GE1 -7.653810e+03 0.000000e+00 1.172e-01 9.182e+00
GE2 -7.660698e+03 -6.887227e+00 9.635e-02 8.233e+00
GE3 -7.663893e+03 -3.195615e+00 1.673e-02 8.213e+00
GE4 -7.663908e+03 -1.521795e-02 2.058e-03 8.213e+00
GE5 -7.663910e+03 -1.464449e-03 2.765e-04 8.192e+00
GE6 -7.663910e+03 -1.052038e-05 2.365e-05 8.226e+00
GE7 -7.663910e+03 -6.776327e-08 4.643e-07 8.209e+00
|CLASS_NAME---------|NAME---------------|TIME(Sec)-----|CALLS----|AVG------|PER%-------
total 61.091 7 8.7 1e+02 %
Driver driver_line 61.056 1 61 1e+02 %
ORB_control set_orb_tables 2.2134 1 2.2 3.6 %
ORB_gen_tables gen_tables 2.2134 1 2.2 3.6 %
ORB_table_phi init_Table 1.528 1 1.5 2.5 %
ORB_table_phi cal_ST_Phi12_R 1.5131 278 0.0054 2.5 %
ORB_table_beta init_Table_Beta 0.48746 1 0.49 0.8 %
ORB_table_beta VNL_PhiBeta_R 0.48358 120 0.004 0.79 %
ESolver_KS_LCAO Run 58.629 1 59 96 %
ESolver_KS_LCAO beforescf 0.15308 1 0.15 0.25 %
Potential update_from_charge 0.61232 8 0.077 1 %
Potential cal_v_eff 0.60904 8 0.076 1 %
PW_Basis real2recip 0.1081 62 0.0017 0.18 %
PotXC cal_v_eff 0.55798 8 0.07 0.91 %
XC_Functional v_xc 0.55683 8 0.07 0.91 %
HSolverLCAO solve 57.812 7 8.3 95 %
HamiltLCAO updateHk 30.389 252 0.12 50 %
OperatorLCAO init 27.198 504 0.054 45 %
Veff contributeHk 27.193 252 0.11 45 %
Gint_interface cal_gint 49.622 14 3.5 81 %
Gint_interface cal_gint_vlocal 26.86 7 3.8 44 %
Gint_Tools cal_psir_ylm 6.6024 81648 8.1e-05 11 %
Gint_k folding_vl_k 0.33298 252 0.0013 0.55 %
Gint_k Distri 0.10942 252 0.00043 0.18 %
Overlap contributeHR 0.10519 1 0.11 0.17 %
LCAO_gen_fixedH calculate_S_no 0.10518 1 0.11 0.17 %
Ekin<LCAO> contributeHR 0.10553 1 0.11 0.17 %
Nonlocal<LCAO> contributeHR 0.65638 1 0.66 1.1 %
LCAO_gen_fixedH b_NL_mu_new 0.65617 1 0.66 1.1 %
OperatorLCAO folding_fixed 2.3242 252 0.0092 3.8 %
LCAO_nnr folding_fixedH 2.3144 252 0.0092 3.8 %
HSolverLCAO hamiltSolvePsiK 1.512 252 0.006 2.5 %
DiagoElpa elpa_solve 1.4488 252 0.0057 2.4 %
ElecStateLCAO psiToRho 25.909 7 3.7 42 %
LCAO_Charge cal_dk_k 3.0993 7 0.44 5.1 %
Gint_interface cal_gint_rho 22.761 7 3.3 37 %
----------------------------------------------------------------------------------------
START Time : Mon Apr 3 15:12:21 2023
FINISH Time : Mon Apr 3 15:13:22 2023
TOTAL Time : 61
SEE INFORMATION IN : OUT.MgO/
代码
文本
[7]
! grep ETOT OUT.MgO/running_scf.log
!FINAL_ETOT_IS -7663.909829526411 eV
代码
文本
代码
文本
这里的STRU结构为MgO的原胞(primitive cell)。
代码
文本
[8]
cd ../../MgO_LCAO/band_structure
/data/bohrium-notebook/Proem/colombo-ABACUS/ABACUS/MgO_LCAO/band_structure
代码
文本
[9]
cat STRU
ATOMIC_SPECIES Mg 24.305 Mg_ONCV_PBE-1.0.upf upf201 O 15.999 O_ONCV_PBE-1.0.upf upf201 NUMERICAL_ORBITAL Mg_gga_8au_100Ry_4s2p1d.orb O_gga_8au_100Ry_2s2p1d.orb LATTICE_CONSTANT 8.087214649032902 ATOMIC_POSITIONS Direct Mg #label 0 #magnetism 1 #number of atoms 0 0 0 m 0 0 0 O #label 0 #magnetism 1 #number of atoms 0.5 0.5 0.5 m 0 0 0
代码
文本
[10]
cp INPUT_scf INPUT && cp KPT_scf KPT
代码
文本
[11]
cat INPUT
INPUT_PARAMETERS suffix MgO ntype 2 pseudo_dir ../../PP_ORB orbital_dir ../../PP_ORB ecutwfc 100 scf_thr 1e-6 basis_type lcao calculation scf out_chg 1 symmetry 1 latname fcc
代码
文本
[12]
!. /opt/intel/oneapi/setvars.sh && OMP_NUM_THREADS=1 abacus
:: initializing oneAPI environment ...
dash: SH_VERSION = unknown
args: Using "$@" for setvars.sh arguments:
:: compiler -- latest
:: debugger -- latest
:: dev-utilities -- latest
:: mkl -- latest
:: mpi -- latest
:: tbb -- latest
:: vtune -- latest
:: oneAPI environment initialized ::
WARNING: Total thread number on this node mismatches with hardware availability. This may cause poor performance.
Info: Local MPI proc number: 1,OpenMP thread number: 1,Total thread number: 1,Local thread limit: 52
ABACUS v3.1
Atomic-orbital Based Ab-initio Computation at UStc
Website: http://abacus.ustc.edu.cn/
Documentation: https://abacus.deepmodeling.com/
Repository: https://github.com/abacusmodeling/abacus-develop
https://github.com/deepmodeling/abacus-develop
Mon Apr 3 15:13:25 2023
MAKE THE DIR : OUT.MgO/
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Warning: number valence electrons > 2 for Mg: [Ne] 3s2
Please make sure the pseudopotential file is what you need
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
UNIFORM GRID DIM : 40 * 40 * 40
UNIFORM GRID DIM(BIG): 10 * 10 * 10
DONE(0.0639761 SEC) : SETUP UNITCELL
DONE(0.0997776 SEC) : SYMMETRY
DONE(0.173368 SEC) : INIT K-POINTS
---------------------------------------------------------
Self-consistent calculations for electrons
---------------------------------------------------------
SPIN KPOINTS PROCESSORS NBASE
1 29 1 28
---------------------------------------------------------
Use Systematically Improvable Atomic bases
---------------------------------------------------------
ELEMENT ORBITALS NBASE NATOM XC
Mg 4s2p1d-8au 15 1
O 2s2p1d-8au 13 1
---------------------------------------------------------
Initial plane wave basis and FFT box
---------------------------------------------------------
-------------------------------------------
SELF-CONSISTENT :
-------------------------------------------
START CHARGE : atomic
DONE(2.52608 SEC) : INIT SCF
ITER ETOT(eV) EDIFF(eV) DRHO TIME(s)
GE1 -1.913453e+03 0.000000e+00 1.172e-01 3.067e+00
GE2 -1.915174e+03 -1.721656e+00 9.638e-02 2.854e+00
GE3 -1.915974e+03 -7.993561e-01 1.673e-02 2.857e+00
GE4 -1.915977e+03 -3.794312e-03 2.061e-03 2.856e+00
GE5 -1.915978e+03 -3.672082e-04 2.767e-04 2.848e+00
GE6 -1.915978e+03 -2.633162e-06 2.365e-05 2.853e+00
GE7 -1.915978e+03 -1.705373e-08 4.561e-07 2.848e+00
|CLASS_NAME---------|NAME---------------|TIME(Sec)-----|CALLS----|AVG------|PER%-------
total 22.823 9 2.5 1e+02 %
Driver driver_line 22.788 1 23 1e+02 %
ORB_control set_orb_tables 2.1871 1 2.2 9.6 %
ORB_gen_tables gen_tables 2.1871 1 2.2 9.6 %
ORB_table_phi init_Table 1.5112 1 1.5 6.6 %
ORB_table_phi cal_ST_Phi12_R 1.4965 278 0.0054 6.6 %
ORB_table_beta init_Table_Beta 0.48066 1 0.48 2.1 %
ORB_table_beta VNL_PhiBeta_R 0.47688 120 0.004 2.1 %
ESolver_KS_LCAO Run 20.345 1 20 89 %
Potential update_from_charge 0.22907 8 0.029 1 %
Potential cal_v_eff 0.22801 8 0.029 1 %
PotXC cal_v_eff 0.21081 8 0.026 0.92 %
XC_Functional v_xc 0.21048 8 0.026 0.92 %
HSolverLCAO solve 19.402 7 2.8 85 %
HamiltLCAO updateHk 11.516 203 0.057 50 %
OperatorLCAO init 10.811 406 0.027 47 %
Veff contributeHk 10.81 203 0.053 47 %
Gint_interface cal_gint 18.225 14 1.3 80 %
Gint_interface cal_gint_vlocal 10.745 7 1.5 47 %
Gint_Tools cal_psir_ylm 2.4918 14000 0.00018 11 %
Nonlocal<LCAO> contributeHR 0.16289 1 0.16 0.71 %
LCAO_gen_fixedH b_NL_mu_new 0.16283 1 0.16 0.71 %
OperatorLCAO folding_fixed 0.48838 203 0.0024 2.1 %
LCAO_nnr folding_fixedH 0.48693 203 0.0024 2.1 %
HSolverLCAO hamiltSolvePsiK 0.17813 203 0.00088 0.78 %
DiagoElpa elpa_solve 0.12734 203 0.00063 0.56 %
ElecStateLCAO psiToRho 7.7075 7 1.1 34 %
LCAO_Charge cal_dk_k 0.22205 7 0.032 0.97 %
Gint_interface cal_gint_rho 7.4794 7 1.1 33 %
----------------------------------------------------------------------------------------
START Time : Mon Apr 3 15:13:25 2023
FINISH Time : Mon Apr 3 15:13:48 2023
TOTAL Time : 23
SEE INFORMATION IN : OUT.MgO/
代码
文本
运行非自洽计算需在INPUT中将calculation设置为nscf，为了输出能带结构文件，还需设置out_band为true或者1。高对称k点路径在KPT文件中设置，选用Line模式。
代码
文本
[13]
cp INPUT_nscf INPUT && cp KPT_nscf KPT
代码
文本
[14]
cat INPUT
INPUT_PARAMETERS suffix MgO ntype 2 pseudo_dir ../../PP_ORB orbital_dir ../../PP_ORB ecutwfc 100 scf_thr 1e-6 basis_type lcao calculation nscf init_chg file out_band 1 out_dos 1 latname fcc symmetry 0
代码
文本
[15]
cat KPT
K_POINTS 7 Line 0.375 0.375 0.750 20 # K 0.000 0.000 0.000 20 # G 0.500 0.000 0.500 20 # X 0.625 0.250 0.625 20 # U 0.500 0.500 0.500 20 # L 0.000 0.000 0.000 20 # G 0.500 0.250 0.750 1 # W
代码
文本
[16]
!. /opt/intel/oneapi/setvars.sh && OMP_NUM_THREADS=1 abacus
:: initializing oneAPI environment ...
dash: SH_VERSION = unknown
args: Using "$@" for setvars.sh arguments:
:: compiler -- latest
:: debugger -- latest
:: dev-utilities -- latest
:: mkl -- latest
:: mpi -- latest
:: tbb -- latest
:: vtune -- latest
:: oneAPI environment initialized ::
WARNING: Total thread number on this node mismatches with hardware availability. This may cause poor performance.
Info: Local MPI proc number: 1,OpenMP thread number: 1,Total thread number: 1,Local thread limit: 52
ABACUS v3.1
Atomic-orbital Based Ab-initio Computation at UStc
Website: http://abacus.ustc.edu.cn/
Documentation: https://abacus.deepmodeling.com/
Repository: https://github.com/abacusmodeling/abacus-develop
https://github.com/deepmodeling/abacus-develop
Mon Apr 3 15:13:51 2023
MAKE THE DIR : OUT.MgO/
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Warning: number valence electrons > 2 for Mg: [Ne] 3s2
Please make sure the pseudopotential file is what you need
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
UNIFORM GRID DIM : 40 * 40 * 40
UNIFORM GRID DIM(BIG): 10 * 10 * 10
DONE(0.0518589 SEC) : SETUP UNITCELL
DONE(0.0620326 SEC) : INIT K-POINTS
---------------------------------------------------------
---------------------------------------------------------
SPIN KPOINTS PROCESSORS NBASE
1 121 1 28
---------------------------------------------------------
Use Systematically Improvable Atomic bases
---------------------------------------------------------
ELEMENT ORBITALS NBASE NATOM XC
Mg 4s2p1d-8au 15 1
O 2s2p1d-8au 13 1
---------------------------------------------------------
Initial plane wave basis and FFT box
---------------------------------------------------------
-------------------------------------------
NONSELF-CONSISTENT :
-------------------------------------------
START CHARGE : file
NON-SELF CONSISTENT CALCULATIONS
|CLASS_NAME---------|NAME---------------|TIME(Sec)-----|CALLS----|AVG------|PER%-------
total 8.4843 5 1.7 1e+02 %
Driver driver_line 8.4595 1 8.5 1e+02 %
ORB_control set_orb_tables 2.178 1 2.2 26 %
ORB_gen_tables gen_tables 2.178 1 2.2 26 %
ORB_table_phi init_Table 1.5037 1 1.5 18 %
ORB_table_phi cal_ST_Phi12_R 1.4886 278 0.0054 18 %
ORB_table_beta init_Table_Beta 0.47611 1 0.48 5.6 %
ORB_table_beta VNL_PhiBeta_R 0.47223 120 0.0039 5.6 %
ESolver_KS_LCAO othercalculation 2.2761 1 2.3 27 %
HSolverLCAO solve 2.1905 1 2.2 26 %
HamiltLCAO updateHk 2.0828 121 0.017 25 %
OperatorLCAO init 1.5669 242 0.0065 18 %
Veff contributeHk 1.5661 121 0.013 18 %
Gint_interface cal_gint 1.5275 1 1.5 18 %
Gint_interface cal_gint_vlocal 1.5275 1 1.5 18 %
Gint_Tools cal_psir_ylm 0.1782 1000 0.00018 2.1 %
Nonlocal<LCAO> contributeHR 0.16534 1 0.17 1.9 %
LCAO_gen_fixedH b_NL_mu_new 0.16529 1 0.17 1.9 %
OperatorLCAO folding_fixed 0.29656 121 0.0025 3.5 %
LCAO_nnr folding_fixedH 0.59353 242 0.0025 7 %
HSolverLCAO hamiltSolvePsiK 0.10727 121 0.00089 1.3 %
ModuleIO nscf_band 0.51157 1 0.51 6 %
----------------------------------------------------------------------------------------
START Time : Mon Apr 3 15:13:51 2023
FINISH Time : Mon Apr 3 15:14:00 2023
TOTAL Time : 9
SEE INFORMATION IN : OUT.MgO/
代码
文本
[17]
cd OUT.MgO
/data/bohrium-notebook/Proem/colombo-ABACUS/ABACUS/MgO_LCAO/band_structure/OUT.MgO
代码
文本
[18]
cp ../KPT ../config.json .
代码
文本
在running_scf.log里面搜索E_Fermi，找到费米能级的值
代码
文本
[19]
cat config.json
{
"bandfile" : "BANDS_1.dat",
"efermi" : 8.27,
"energy_range" : [-4, 9],
"kptfile" : "KPT"
}
代码
文本
[20]
!/opt/miniconda3/bin/abacus-plot -b
代码
文本
[21]
import matplotlib.pyplot as plt
from PIL import Image
plt.axis('off')
plt.imshow(Image.open('band.png'))
plt.show()
代码
文本
态密度(Density of states)是另外一个用于分析材料性质的重要性质。
它本质上描述的是在不同能量上电子态的密度，利用态密度可以得到材料的光学性质等。
通过将波函数投影到每个原子的不同轨道，我们还可以分析不同原子轨道对电子态密度的贡献，用于从原子和电子角度分析和设计材料性质。
代码
文本
与能带计算的自洽计算相同。
代码
文本
[22]
cd ../../../MgO_LCAO/dos
/data/bohrium-notebook/Proem/colombo-ABACUS/ABACUS/MgO_LCAO/dos
代码
文本
[23]
cp INPUT_scf INPUT && cp KPT_scf KPT
代码
文本
[24]
cat KPT
K_POINTS 0 Gamma 8 8 8 0 0 0
代码
文本
[25]
!. /opt/intel/oneapi/setvars.sh && OMP_NUM_THREADS=1 abacus
:: initializing oneAPI environment ...
dash: SH_VERSION = unknown
args: Using "$@" for setvars.sh arguments:
:: compiler -- latest
:: debugger -- latest
:: dev-utilities -- latest
:: mkl -- latest
:: mpi -- latest
:: tbb -- latest
:: vtune -- latest
:: oneAPI environment initialized ::
WARNING: Total thread number on this node mismatches with hardware availability. This may cause poor performance.
Info: Local MPI proc number: 1,OpenMP thread number: 1,Total thread number: 1,Local thread limit: 52
ABACUS v3.1
Atomic-orbital Based Ab-initio Computation at UStc
Website: http://abacus.ustc.edu.cn/
Documentation: https://abacus.deepmodeling.com/
Repository: https://github.com/abacusmodeling/abacus-develop
https://github.com/deepmodeling/abacus-develop
Mon Apr 3 15:14:08 2023
MAKE THE DIR : OUT.MgO/
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Warning: number valence electrons > 2 for Mg: [Ne] 3s2
Please make sure the pseudopotential file is what you need
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
UNIFORM GRID DIM : 40 * 40 * 40
UNIFORM GRID DIM(BIG): 10 * 10 * 10
DONE(0.0685985 SEC) : SETUP UNITCELL
DONE(0.104701 SEC) : SYMMETRY
DONE(0.178032 SEC) : INIT K-POINTS
---------------------------------------------------------
Self-consistent calculations for electrons
---------------------------------------------------------
SPIN KPOINTS PROCESSORS NBASE
1 29 1 28
---------------------------------------------------------
Use Systematically Improvable Atomic bases
---------------------------------------------------------
ELEMENT ORBITALS NBASE NATOM XC
Mg 4s2p1d-8au 15 1
O 2s2p1d-8au 13 1
---------------------------------------------------------
Initial plane wave basis and FFT box
---------------------------------------------------------
-------------------------------------------
SELF-CONSISTENT :
-------------------------------------------
START CHARGE : atomic
DONE(2.51577 SEC) : INIT SCF
ITER ETOT(eV) EDIFF(eV) DRHO TIME(s)
GE1 -1.913453e+03 0.000000e+00 1.172e-01 3.064e+00
GE2 -1.915174e+03 -1.721656e+00 9.638e-02 2.841e+00
GE3 -1.915974e+03 -7.993561e-01 1.673e-02 2.842e+00
GE4 -1.915977e+03 -3.794312e-03 2.061e-03 2.842e+00
GE5 -1.915978e+03 -3.672082e-04 2.767e-04 2.829e+00
GE6 -1.915978e+03 -2.633162e-06 2.365e-05 2.848e+00
GE7 -1.915978e+03 -1.705373e-08 4.561e-07 2.845e+00
|CLASS_NAME---------|NAME---------------|TIME(Sec)-----|CALLS----|AVG------|PER%-------
total 22.745 9 2.5 1e+02 %
Driver driver_line 22.709 1 23 1e+02 %
ORB_control set_orb_tables 2.1734 1 2.2 9.6 %
ORB_gen_tables gen_tables 2.1734 1 2.2 9.6 %
ORB_table_phi init_Table 1.506 1 1.5 6.6 %
ORB_table_phi cal_ST_Phi12_R 1.4913 278 0.0054 6.6 %
ORB_table_beta init_Table_Beta 0.47219 1 0.47 2.1 %
ORB_table_beta VNL_PhiBeta_R 0.46835 120 0.0039 2.1 %
ESolver_KS_LCAO Run 20.271 1 20 89 %
Potential update_from_charge 0.22668 8 0.028 1 %
Potential cal_v_eff 0.22561 8 0.028 0.99 %
PotXC cal_v_eff 0.20855 8 0.026 0.92 %
XC_Functional v_xc 0.20823 8 0.026 0.92 %
HSolverLCAO solve 19.347 7 2.8 85 %
HamiltLCAO updateHk 11.549 203 0.057 51 %
OperatorLCAO init 10.84 406 0.027 48 %
Veff contributeHk 10.838 203 0.053 48 %
Gint_interface cal_gint 18.179 14 1.3 80 %
Gint_interface cal_gint_vlocal 10.774 7 1.5 47 %
Gint_Tools cal_psir_ylm 2.5668 14000 0.00018 11 %
Nonlocal<LCAO> contributeHR 0.16355 1 0.16 0.72 %
LCAO_gen_fixedH b_NL_mu_new 0.16349 1 0.16 0.72 %
OperatorLCAO folding_fixed 0.49287 203 0.0024 2.2 %
LCAO_nnr folding_fixedH 0.4914 203 0.0024 2.2 %
HSolverLCAO hamiltSolvePsiK 0.16938 203 0.00083 0.74 %
DiagoElpa elpa_solve 0.12265 203 0.0006 0.54 %
ElecStateLCAO psiToRho 7.6272 7 1.1 34 %
LCAO_Charge cal_dk_k 0.2167 7 0.031 0.95 %
Gint_interface cal_gint_rho 7.4048 7 1.1 33 %
----------------------------------------------------------------------------------------
START Time : Mon Apr 3 15:14:08 2023
FINISH Time : Mon Apr 3 15:14:31 2023
TOTAL Time : 23
SEE INFORMATION IN : OUT.MgO/
代码
文本
做态密度计算中的非自洽计算需设置更加密的k点网格。为了输出能带结构文件，还需设置out_dos为true或者1。
代码
文本
[26]
cp INPUT_nscf INPUT && cp KPT_nscf KPT
代码
文本
[27]
cat KPT
K_POINTS 0 Gamma 10 10 10 0 0 0
代码
文本
[28]
!. /opt/intel/oneapi/setvars.sh && OMP_NUM_THREADS=1 abacus
:: initializing oneAPI environment ...
dash: SH_VERSION = unknown
args: Using "$@" for setvars.sh arguments:
:: compiler -- latest
:: debugger -- latest
:: dev-utilities -- latest
:: mkl -- latest
:: mpi -- latest
:: tbb -- latest
:: vtune -- latest
:: oneAPI environment initialized ::
WARNING: Total thread number on this node mismatches with hardware availability. This may cause poor performance.
Info: Local MPI proc number: 1,OpenMP thread number: 1,Total thread number: 1,Local thread limit: 52
ABACUS v3.1
Atomic-orbital Based Ab-initio Computation at UStc
Website: http://abacus.ustc.edu.cn/
Documentation: https://abacus.deepmodeling.com/
Repository: https://github.com/abacusmodeling/abacus-develop
https://github.com/deepmodeling/abacus-develop
Mon Apr 3 15:14:34 2023
MAKE THE DIR : OUT.MgO/
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Warning: number valence electrons > 2 for Mg: [Ne] 3s2
Please make sure the pseudopotential file is what you need
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
UNIFORM GRID DIM : 40 * 40 * 40
UNIFORM GRID DIM(BIG): 10 * 10 * 10
DONE(0.0614797 SEC) : SETUP UNITCELL
DONE(0.0851821 SEC) : INIT K-POINTS
---------------------------------------------------------
---------------------------------------------------------
SPIN KPOINTS PROCESSORS NBASE
1 504 1 28
---------------------------------------------------------
Use Systematically Improvable Atomic bases
---------------------------------------------------------
ELEMENT ORBITALS NBASE NATOM XC
Mg 4s2p1d-8au 15 1
O 2s2p1d-8au 13 1
---------------------------------------------------------
Initial plane wave basis and FFT box
---------------------------------------------------------
-------------------------------------------
NONSELF-CONSISTENT :
-------------------------------------------
START CHARGE : file
NON-SELF CONSISTENT CALCULATIONS
|CLASS_NAME---------|NAME---------------|TIME(Sec)-----|CALLS----|AVG------|PER%-------
total 12.769 5 2.6 1e+02 %
Driver driver_line 12.741 1 13 1e+02 %
ORB_control set_orb_tables 2.253 1 2.3 18 %
ORB_gen_tables gen_tables 2.253 1 2.3 18 %
ORB_table_phi init_Table 1.5323 1 1.5 12 %
ORB_table_phi cal_ST_Phi12_R 1.5173 278 0.0055 12 %
ORB_table_beta init_Table_Beta 0.50964 1 0.51 4 %
ORB_table_beta VNL_PhiBeta_R 0.50553 120 0.0042 4 %
ESolver_KS_LCAO othercalculation 3.6888 1 3.7 29 %
HSolverLCAO solve 3.582 1 3.6 28 %
HamiltLCAO updateHk 3.1582 504 0.0063 25 %
OperatorLCAO init 1.7253 1008 0.0017 14 %
Veff contributeHk 1.722 504 0.0034 13 %
Gint_interface cal_gint 1.5624 1 1.6 12 %
Gint_interface cal_gint_vlocal 1.5624 1 1.6 12 %
Gint_Tools cal_psir_ylm 0.17806 1000 0.00018 1.4 %
Gint_k folding_vl_k 0.1592 504 0.00032 1.2 %
Nonlocal<LCAO> contributeHR 0.16364 1 0.16 1.3 %
LCAO_gen_fixedH b_NL_mu_new 0.16358 1 0.16 1.3 %
OperatorLCAO folding_fixed 1.2155 504 0.0024 9.5 %
LCAO_nnr folding_fixedH 2.4493 1008 0.0024 19 %
HSolverLCAO hamiltSolvePsiK 0.42219 504 0.00084 3.3 %
DiagoElpa elpa_solve 0.31119 504 0.00062 2.4 %
----------------------------------------------------------------------------------------
START Time : Mon Apr 3 15:14:34 2023
FINISH Time : Mon Apr 3 15:14:46 2023
TOTAL Time : 12
SEE INFORMATION IN : OUT.MgO/
代码
文本
总态密度(Total density of states)是指在一个能量范围内电子在每个能量点上总的态密度值。
这里的“总”是相对下面的投影态密度(projected density of states)讲的，而投影态密度指的是在一个能量范围内将电子波函数投影到每个原子的每个轨道得到的态密度值。
代码
文本
[29]
cd OUT.MgO
/data/bohrium-notebook/Proem/colombo-ABACUS/ABACUS/MgO_LCAO/dos/OUT.MgO
代码
文本
[30]
cp ../config_tdos.json ./config.json
代码
文本
[31]
cat config.json
{
"tdosfile": "TDOS",
"efermi": 8.27771540465,
"energy_range": [-5,7],
"dos_range": [0,5],
"figsize":[14,10],
"tdosfig": "tdos.png"
}
代码
文本
[32]
!/opt/miniconda3/bin/abacus-plot -d
代码
文本
[33]
import matplotlib.pyplot as plt
from PIL import Image
plt.axis('off')
plt.imshow(Image.open('tdos.png'))
plt.show()
代码
文本
[34]
cp ../config_pdos.json ./config.json
代码
文本
[35]
cat config.json
{
"pdosfile": "PDOS",
"efermi": 8.27771540465,
"energy_range": [-5,7],
"dos_range": [0,5],
"figsize":[14,10],
"species":{"Mg":[0,1,2],"O":[0,1,2]},
"pdosfig": "pdos.png"
}
代码
文本
[36]
!/opt/miniconda3/bin/abacus-plot -d -p -o
代码
文本
[37]
import matplotlib.pyplot as plt
from PIL import Image
plt.axis('off')
plt.imshow(Image.open('pdos.png'))
plt.show()
代码
文本
结构优化是从一个一般的结构出发，利用优化算法得到基态结构的过程。对于周期性体系，优化计算可以根据需要针对体系的不同自由度来进行优化。
在ABACUS中，如果只优化原子位置，而晶胞固定不变，则需在INPUT中将calculation设置为relax;
如果需要同时做原子位置优化和晶胞优化，则需要将calculation设置为cell-relax；
此外，需要注意的参数还有原子受力的收敛标准force_thr_ev和应力收敛标准stress_thr的设置；
更多的优化设置，请阅读ABACUS Documentation。关于结构优化，可参考文档中对这部分的详细说明。
代码
文本
[38]
cd ../../../MgO_LCAO/optimization
/data/bohrium-notebook/Proem/colombo-ABACUS/ABACUS/MgO_LCAO/optimization
代码
文本
[39]
cat INPUT
INPUT_PARAMETERS suffix MgO ntype 2 pseudo_dir ../../PP_ORB orbital_dir ../../PP_ORB ecutwfc 100 scf_thr 1e-6 basis_type lcao calculation cell-relax force_thr_ev 0.01 stress_thr 5 relax_nmax 100 out_stru 1
代码
文本
[40]
!. /opt/intel/oneapi/setvars.sh && OMP_NUM_THREADS=1 abacus
:: initializing oneAPI environment ...
dash: SH_VERSION = unknown
args: Using "$@" for setvars.sh arguments:
:: compiler -- latest
:: debugger -- latest
:: dev-utilities -- latest
:: mkl -- latest
:: mpi -- latest
:: tbb -- latest
:: vtune -- latest
:: oneAPI environment initialized ::
WARNING: Total thread number on this node mismatches with hardware availability. This may cause poor performance.
Info: Local MPI proc number: 1,OpenMP thread number: 1,Total thread number: 1,Local thread limit: 52
ABACUS v3.1
Atomic-orbital Based Ab-initio Computation at UStc
Website: http://abacus.ustc.edu.cn/
Documentation: https://abacus.deepmodeling.com/
Repository: https://github.com/abacusmodeling/abacus-develop
https://github.com/deepmodeling/abacus-develop
Mon Apr 3 15:15:01 2023
MAKE THE DIR : OUT.MgO/
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Warning: number valence electrons > 2 for Mg: [Ne] 3s2
Please make sure the pseudopotential file is what you need
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
UNIFORM GRID DIM : 54 * 54 * 54
UNIFORM GRID DIM(BIG): 18 * 18 * 18
DONE(0.0830077 SEC) : SETUP UNITCELL
DONE(0.0940724 SEC) : INIT K-POINTS
---------------------------------------------------------
Cell relaxation calculations
---------------------------------------------------------
SPIN KPOINTS PROCESSORS NBASE
1 36 1 112
---------------------------------------------------------
Use Systematically Improvable Atomic bases
---------------------------------------------------------
ELEMENT ORBITALS NBASE NATOM XC
Mg 4s2p1d-8au 15 4
O 2s2p1d-8au 13 4
---------------------------------------------------------
Initial plane wave basis and FFT box
---------------------------------------------------------
-------------------------------------------
STEP OF RELAXATION : 1
-------------------------------------------
START CHARGE : atomic
DONE(2.56792 SEC) : INIT SCF
ITER ETOT(eV) EDIFF(eV) DRHO TIME(s)
GE1 -7.653810e+03 0.000000e+00 1.172e-01 9.129e+00
GE2 -7.660698e+03 -6.887227e+00 9.635e-02 8.247e+00
GE3 -7.663893e+03 -3.195615e+00 1.673e-02 8.224e+00
GE4 -7.663908e+03 -1.521795e-02 2.058e-03 8.142e+00
GE5 -7.663910e+03 -1.464449e-03 2.765e-04 8.164e+00
GE6 -7.663910e+03 -1.052038e-05 2.365e-05 8.149e+00
GE7 -7.663910e+03 -6.776327e-08 4.643e-07 8.150e+00
><><><><><><><><><><><><><><><><><><><><><><
TOTAL-STRESS (KBAR):
><><><><><><><><><><><><><><><><><><><><><><
1.202e+00 4.377e-12 1.201e-12
4.377e-12 1.202e+00 1.541e-12
1.201e-12 1.541e-12 1.202e+00
TOTAL-PRESSURE: 1.202e+00 KBAR
LARGEST GRAD (eV/A) : 2.106e-11
|CLASS_NAME---------|NAME---------------|TIME(Sec)-----|CALLS----|AVG------|PER%-------
total 79.81 7 11 1e+02 %
Driver driver_line 79.772 1 80 1e+02 %
ORB_control set_orb_tables 2.1614 1 2.2 2.7 %
ORB_gen_tables gen_tables 2.1614 1 2.2 2.7 %
ORB_table_phi init_Table 1.4899 1 1.5 1.9 %
ORB_table_phi cal_ST_Phi12_R 1.4749 278 0.0053 1.8 %
ORB_table_beta init_Table_Beta 0.476 1 0.48 0.6 %
ORB_table_beta VNL_PhiBeta_R 0.47219 120 0.0039 0.59 %
ESolver_KS_LCAO Run 58.369 1 58 73 %
ESolver_KS_LCAO beforescf 0.15329 1 0.15 0.19 %
Potential update_from_charge 0.61303 8 0.077 0.77 %
Potential cal_v_eff 0.6098 8 0.076 0.76 %
PW_Basis real2recip 0.11709 68 0.0017 0.15 %
PotXC cal_v_eff 0.55894 8 0.07 0.7 %
XC_Functional v_xc 0.55778 8 0.07 0.7 %
HSolverLCAO solve 57.551 7 8.2 72 %
HamiltLCAO updateHk 30.066 252 0.12 38 %
OperatorLCAO init 26.874 504 0.053 34 %
Veff contributeHk 26.87 252 0.11 34 %
Gint_interface cal_gint 61.396 15 4.1 77 %
Gint_interface cal_gint_vlocal 26.539 7 3.8 33 %
Gint_Tools cal_psir_ylm 6.5462 81648 8e-05 8.2 %
Gint_k folding_vl_k 0.33013 252 0.0013 0.41 %
Gint_k Distri 0.10859 252 0.00043 0.14 %
Overlap contributeHR 0.10385 1 0.1 0.13 %
LCAO_gen_fixedH calculate_S_no 0.10385 1 0.1 0.13 %
Ekin<LCAO> contributeHR 0.105 1 0.11 0.13 %
Nonlocal<LCAO> contributeHR 0.66882 1 0.67 0.84 %
LCAO_gen_fixedH b_NL_mu_new 2.1443 2 1.1 2.7 %
OperatorLCAO folding_fixed 2.3136 252 0.0092 2.9 %
LCAO_nnr folding_fixedH 2.3033 252 0.0091 2.9 %
HSolverLCAO hamiltSolvePsiK 1.521 252 0.006 1.9 %
DiagoElpa elpa_solve 1.4572 252 0.0058 1.8 %
ElecStateLCAO psiToRho 25.963 7 3.7 33 %
LCAO_Charge cal_dk_k 3.1115 7 0.44 3.9 %
Gint_interface cal_gint_rho 22.803 7 3.3 29 %
Force_Stress_LCAO getForceStress 19.022 1 19 24 %
Force_LCAO_k ftable_k 18.865 1 19 24 %
Force_LCAO_k allocate_k 2.3928 1 2.4 3 %
Force_LCAO_k cal_foverlap_k 0.11431 1 0.11 0.14 %
Force_LCAO_k cal_edm_2d 0.10043 1 0.1 0.13 %
Local_Orbital_Chargecal_dm_R 0.18659 2 0.093 0.23 %
Force_LCAO_k cal_fvl_dphi_k 12.054 1 12 15 %
Gint_interface cal_gint_force 12.054 1 12 15 %
Gint_Tools cal_dpsir_ylm 4.9742 5832 0.00085 6.2 %
Gint_Tools cal_dpsirr_ylm 0.69358 5832 0.00012 0.87 %
Force_LCAO_k cal_fvnl_dbeta_k_new4.1995 1 4.2 5.3 %
----------------------------------------------------------------------------------------
START Time : Mon Apr 3 15:15:01 2023
FINISH Time : Mon Apr 3 15:16:21 2023
TOTAL Time : 80
SEE INFORMATION IN : OUT.MgO/
代码
文本
[41]
# 在案例完成后，删除该案例中下载的 colombo-ABACUS 文件夹
import os
# 获取当前工作目录的绝对路径
current_dir = os.getcwd()
if current_dir.endswith("colombo-ABACUS/ABACUS/MgO_LCAO/optimization"):
# 获取 colombo-ABACUS 上级目录的绝对路径
parent_dir = os.path.abspath(os.path.join(current_dir, '../../../..'))
os.chdir(parent_dir)
! if [ -d "colombo-ABACUS" ]; then rm -r colombo-ABACUS; echo "删除成功"; fi
删除成功
代码
文本
代码
文本
已赞19
本文被以下合集收录
ABACUS
bohr9d1500
更新于 2024-05-18
14 篇5 人关注
测试合集文章列表100篇
xingyanshi@dp.tech很长的名字xingyanshi@dp.tech很长的名字很长的
更新于 2024-08-04
104 篇2 人关注
推荐阅读
公开
未命名Esplendo
发布于 2024-05-16
公开
快速开始 TensorFlow2｜面向初学者的快速入门MileAway
发布于 2023-06-10
3 转存文件
评论
wanglichao@dp.tech