QE

Quantum ESPRESSO


  • 执行计算
    pw.x -in qe.in 1> qe.out 2> qe.err
    pw.x < qe.in > qe.log
  • nat (number of atoms in the unit cell)
    用来指明体系的原胞中原子的总共数目,赋值为1表示所计算的原胞中只有一个原子
  • ntyp (number of types of atoms/ATOMIC_SPECIES in the unit cell)  
    用来指明体系中原子类型的数目,赋值为1表示所计算的体系只有一种类型的原子
  • charged systems
    • tot_charge
    • starting_charge(i), i=1,ntyp. starting charge on atomic type 'i',指定原子类型'i'的初始化电荷量,默认为0.0,即电中性.
      starting_charge(i) = 0.0, for i in {1..ntyp}
      对于extrinsic defect,可以按照电荷类型指定带电量,但是对于intrinsic defect(如氧空位)则无法通过该方式指定具体某个缺陷的电荷量. 此时,只能指定tol_charge,并引入背景电荷来补偿.
  • atomic units (a.u)
    • Hartree atomic units (Ha a.u.)
      • ħ=me=e=1
      • 1 Hartree (Ha) = 2 Rydberg (Ry) ≈ 27.211 eV
      • QE中能量习惯用 Ry = 0.5 Ha ≈ 13.6
      • electric field unit in the Hartree a.u: 1 a.u. = 51.4220632 V/Å
    • Rydberg atomic units (Ry a.u.)
      • ħ=2me=e2/2=1
      • electric field unit in the Rydberg a.u: 1 a.u. = 36.3609 V/Å = 51.4220632/Sqrt(2) V/Å 
  • Scientific notation
    • D notation comes from Fortran, where it means DOUBLE PRECISION
    • Note that the D notation is identical to the E notation
  • 查看力和能量的收敛情况(直接在log中查找keywords)
    grep "Gradient error" log  # or grep Grad log, 查看力
    grep "estimated scf accuracy" log # or grep scf log, 查看能量
  • FCP (Fictitious Charge Particle) 
  • QE原子坐标形式
    • 直角坐标
      • bohr:以Bohr半径为单位
      • angstrom:以Å为单位
      • alat:alat是晶格矢量v1(lattice constant a)的长度,即celldm(1) or A
        • 注:晶格参数由celldm(1)=a,celldm(2)=b/a,celldm(3)=c/a及夹角的余弦值elldm(4:6)或A、B、C及cosAB、cosAC、cosBC来设置
    • 分数坐标
      • crystal:以晶格矢量为基矢量的相对坐标
  • Toolchains
    • BURAI
      • 可以在windows上运行QE(v6.1)
      •  Input文件手动修改后,需要点击↑,GUI修改后需要点击刷新
        • 手动添加参数后,会重新按照字母顺序排序!
        • 参数需要在对应的任务中修改!
      • 从Result中查看结果,Reload更新,左下←返回上级
      • Linux上运行BURAI的JRE(JRE、JVM及JDK的关系如下)建议到官网下载.tar.gz解压包或.rpm包安装!

    •  PWgui
      • 产生输入文件(包括计算及后处理)和了解相关参数
        • 从Lattice&Atoms中编辑和导入结构信息(load atomic coordinates from file...)
        • View中查看输入文件,若已安装XCrySDen,可以显示三维结构 
    •  XCrySDen(类似VESTA)
      • 编译很恶心(很难安装!),特别是在CentOS下!参考Sobereva(卢天老师)针对CentOS打包的预编译版
      • 也可在Ubuntu上安装bin版本使用,或者直接使用包管理器sudo apt-get install xcrysden安装旧版本
        sudo apt-get install xcrysden
  • 价电子读取
    • 从赝势中读取价电子数(z_valence)
    • 从log中读取:Zval
  • QE安装
    • 安装前,首先部署好MPI并行环境(建议Intel oneAPI套装,oneAPI Base+HPC toolkit,参考)./configure会默认检测MPI并行环境,如果未检测到,则安装的是串行版
    • 下载qe-7.x安装目录并解压到指定目录
      ./configure # configure如有Error,先执行 make depend, or ./install/makedeps.sh
      make -jN all #指定N核编译,N为具体核数,建议4个,多核(如10个)编译似乎容易失败
                   #若编译时长时间停顿可能是外网问题导致无法下载依赖包,可手动下载后
                   #放在archive/下
      #make install 如果有root权限,可以把编译好的可执行文件添加至/usr/local/bin
      #测试
      cd  test-suite
      make run-tests-XX-parallel #run tests for XX={pw,cp,ph,epw,hp,tddfpt kcw} only
  • 输入文件格式
    • &Keywords
      /
    • It is alternative to preserve the comma-separation of lines in sections such as the CONTROLSYSTEM, and ELECTRONS cards. 
    • Comments符号用!
  • gap体系Occupation建议使用fixed
  • 加电场方法
    • tefield
      &CONTROL
        tefield  = .true.
        dipfield = .true.
      /
      &SYSTEM
        edir     = 3     ! The direction of the electric field or dipole correction
        eamp     = 0.001 ! Amplitude of the electric field(unit: a.u.,1 a.u. = 51.4220632*10^10 V/m = 51.4220632 V/A)
        emaxpos  = 0.99  ! 0 < eopreg < 1, Position of the maximum of the saw-like potential along crystal axis edir
        eopreg   = 0.01  ! 0 < eopreg < 1, Zone in the unit cell where the saw-like potential decreases
      /

       

      • emaxpos, eopreg and edir are the essential variables needed to define position and width of electric double layer giving the dipole correction.
      • The saw-like potential increases with slope eamp in the region from (emaxpos+eopreg-1) to (emaxpos), then decreases to 0 until (emaxpos+eopreg).
      • It is important that the change of slope, that is emaxpos, lies in the vacuum region and so must emaxpos+eopreg (the opposite charge) – otherwise the correction will not work.
      • It is better to have your system in the middle of the super-cell, and the vacuum region should be large enough so that you can make sure that your system lies between (emaxpos+eopreg-1) and (emaxpos+eopreg)
      • The position of the maximum of the sawlike potential within the unit cell ( emaxpos) should be in the middle of the vacuum region.
      • 在真空层中加入一个电偶极子,也就是引入一个沿着z方向(z是垂直表面方向,由edir设置)的锯齿形势能,这个势能在slab处上升,上升的梯度为电场强度eamp,在真空一个很小的宽度(eopreg)内下降,偶极矩的位置在edir方向电势下降区域 [QE实践详解]
      • eamp的正负号对应电场的正反方向,positive/negative eamp 的saw potential对比

    • lefield
    • esm_efield
posted @ 2022-07-31 17:51  Deven_xu  阅读(231)  评论(0)    收藏  举报