计算化学公社
标题: 做scf计算一直不收敛 [打印本页]
作者Author: 茶味果冻 时间: 2014-11-15 10:51
标题: 做scf计算一直不收敛
模型建立的为二氧化钛锐钛矿001面，1x1的超晶胞，在QE中先做的scf计算，结果很奇怪，迭代100次都没有收敛，下面附上in文件，及结果，求大家帮看看问题出在哪:)
&CONTROL
calculation= "scf",
restart_mode='from_scratch'
prefix="tio2",
pseudo_dir= "/vol6/home/zhang00/pseudo1",
outdir= "/",
wf_collect=.TRUE
/
&SYSTEM
ibrav=0,celldm(1)=3.776000,nat=12,ntyp=2,
ecutwfc=25,ecutrho=300,
occupations ='smearing',
smearing ='gaussian',
degauss =0.002,
/
&ELECTRONS
conv_thr = 1.D-5
mixing_beta = 0.7D0
/
ATOMIC_SPECIES
O 16.00 O.pbe-n-rrkjus_psl.0.1.UPF
Ti 48.00 Ti.pbe-sp-van_ak.UPF
ATOMIC_POSITIONS crystal
O 0.0000000.000000 -0.393602
O 0.000000 0.000000-0.103370
O 0.5000000.500000 -0.145116
O 0.5000000.500000 -0.351856
O 0.0000000.500000 -0.269359
O 0.0000000.500000 -0.476099
O 0.5000000.000000 -0.020873
O 0.5000000.000000 -0.227613
Ti 0.000000 0.000000 0.000000
Ti 0.500000 0.500000 -0.248486
Ti 0.000000 0.500000 -0.372729
Ti 0.500000 0.000000 -0.124243
K_POINTS automatic
2 2 4 0 0 0
CELL_PARAMETERS angstrom
3.7760000 0.0000000 0.0000000
0.0000000 3.7760000 0.0000000
0.0000000 0.0000000 19.0875880
这是in文件，下面是计算结果：
Self-consistentCalculation
iteration # 1 ecut= 25.00 Ry beta=0.70
Davidson diagonalization with overlap
ethr = 1.00E-02, avg # of iterations = 2.0
negative rho (up, down): 0.163E-01 0.000E+00
total cpu time spent up to nowis 63.4 secs
total energy = -397.12256856 Ry
Harris-Foulkes estimate = -736.27759390 Ry
estimated scf accuracy < 378.46031089 Ry
iteration # 2 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
c_bands: 3 eigenvalues not converged
c_bands: 5 eigenvalues not converged
c_bands: 2 eigenvalues not converged
c_bands: 2 eigenvalues not converged
c_bands: 1 eigenvalues not converged
c_bands: 4 eigenvalues not converged
c_bands: 4 eigenvalues not converged
ethr = 1.00E-02, avg # of iterations = 22.6
negative rho (up, down): 0.314E+00 0.000E+00
total cpu time spent up to nowis 228.6 secs
total energy = -767.86613127 Ry
Harris-Foulkes estimate = -1895.37992668 Ry
estimated scf accuracy < 3919.46724766 Ry
iteration # 3 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 15.2
negative rho (up, down): 0.286E+00 0.000E+00
total cpu time spent up to nowis 347.5 secs
total energy = -775.65518344 Ry
Harris-Foulkes estimate = -1038.96639788 Ry
estimated scf accuracy < 1721.30345809 Ry
iteration # 4 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 13.5
negative rho (up, down): 0.280E+00 0.000E+00
total cpu time spent up to nowis 418.8 secs
total energy = -755.80317573 Ry
Harris-Foulkes estimate = -833.23419122 Ry
estimated scf accuracy < 1500.67222812 Ry
iteration # 5 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 6.3
negative rho (up, down): 0.240E+00 0.000E+00
total cpu time spent up to nowis 466.8 secs
total energy = -707.58129894 Ry
Harris-Foulkes estimate = -769.94919724 Ry
estimated scf accuracy < 1060.31353961 Ry
iteration # 6 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 13.6
negative rho (up, down): 0.132E+00 0.000E+00
total cpu time spent up to nowis 540.7 secs
total energy = -740.24716235 Ry
Harris-Foulkes estimate = -756.86916599 Ry
estimated scf accuracy < 84.62104022 Ry
iteration # 7 ecut= 25.00 Ry beta=0.70
Davidson diagonalization with overlap
ethr = 1.00E-02, avg # of iterations = 3.2
negative rho (up, down): 0.133E+00 0.000E+00
total cpu time spent up to nowis 580.1 secs
total energy = -742.97109663 Ry
Harris-Foulkes estimate = -744.51709411Ry
estimated scf accuracy < 122.58044785 Ry
iteration # 8 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 1.2
negative rho (up, down): 0.127E+00 0.000E+00
total cpu time spent up to nowis 613.5 secs
total energy = -744.39121638 Ry
Harris-Foulkes estimate = -744.38140910 Ry
estimated scf accuracy < 185.57825644 Ry
iteration # 9 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 2.1
negative rho (up, down): 0.148E+00 0.000E+00
total cpu time spent up to nowis 647.7 secs
total energy = -742.92517405 Ry
Harris-Foulkes estimate = -744.57245777 Ry
estimated scf accuracy < 201.56723332 Ry
iteration # 10 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 1.2
negative rho (up, down): 0.253E+01 0.000E+00
total cpu time spent up to nowis 681.1 secs
total energy = -714.19687735 Ry
Harris-Foulkes estimate = -743.17939759 Ry
estimated scf accuracy < 166.56686994 Ry
iteration # 11 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 12.8
negative rho (up, down): 0.119E+01 0.000E+00
total cpu time spent up to nowis 768.0 secs
total energy = -720.12214182 Ry
Harris-Foulkes estimate = -771.65212427 Ry
estimated scf accuracy < 648.43840803 Ry
iteration # 12 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 7.3
negative rho (up, down): 0.623E+00 0.000E+00
total cpu time spent up to nowis 827.8 secs
total energy = -720.38448364 Ry
Harris-Foulkes estimate = -747.22863447 Ry
estimated scf accuracy < 304.23289007 Ry
iteration # 13 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 10.3
negative rho (up, down): 0.191E+00 0.000E+00
total cpu time spent up to nowis 893.4 secs
total energy = -729.65948254 Ry
Harris-Foulkes estimate = -740.42400166 Ry
estimated scf accuracy < 87.94588865 Ry
iteration # 14 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 5.8
negative rho (up, down): 0.530E-01 0.000E+00
total cpu time spent up to nowis 939.3 secs
total energy = -732.61535430 Ry
Harris-Foulkes estimate = -740.72404564 Ry
estimated scf accuracy < 55.97266718 Ry
iteration # 15 ecut= 25.00 Ry beta=0.70
Davidson diagonalization withoverlap
ethr = 1.00E-02, avg # of iterations = 3.3
后面省略。。。。。
End ofself-consistent calculation
convergence NOT achieved after 100iterations: stopping
Writing output data filetio2.save
作者Author: 卡开发发 时间: 2014-11-15 12:27
PWscf我不是很会用，但是从参数上看，有下面几个问题。表面体系为啥k取2x2x4?对于1x1的表面差不多4x4x1才是合理的。晶格参数来看你的真空层也许只有10A，最好还能大一些，上下表面本身不对称，层间其实就会有微弱的作用，可以的话考虑work function的校正，截断能原则上也要进行测试。如果上述都没问题，那么应该尝试减小mixing_beta这个参数，默认是0.7还是很大的，不妨减到0.3以下试试。
作者Author: 茶味果冻 时间: 2014-11-15 14:13
刚才调整了一下参数，还是不行，
作者Author: 卡开发发 时间: 2014-11-16 00:35
您的in文件写的有问题，我还特地看了一下PW的资料：
&SYSTEM
ibrav=0,celldm(1)=3.776000,nat=12,ntyp=2,
上面的部分celldm(1)=3.776000修改为celldm(1)=1.89，其中1.89这个取值的由来是1/0.529。这是因为从您的Lattice vector来看您采用的单位是Angstrom，但是我按照这个输入报错了，似乎不支持在CELL_PARAMETERS后面加上单位。而默认的单位是Bohr，我们在这个位置进行换算，并且TiO2锐钛矿的001面的a和b应该是3.776Angstrom。
如果按照您原先的设置，实际计算采用的a和b会变为
a=3.776000*3.776000 b=3.776000*3.776000 c=19.0875880*3.776000，这里单位也会是Bohr，这样不收敛也非常正常。
celldm(1)这个参数相当于一个整体晶格的标度因子
celldm*CELL_PARAMETERS才是实际的晶格参数。
另外，我测试的时候我采用mixing_beta = 0.3D0算收敛的，不过0.7D0也许也能收敛。
作者Author: 茶味果冻 时间: 2014-11-17 09:29
非常感谢您！问题解决了，之前一直看的5.1版本的手册，而计算中心的居然是4.3版本的，现在把4.3版本的手册和in文件说明都下载下来，看了看发现一些设置都不一样，发现自己好不认真，这次在这里学到好多，真心的感谢老师您，祝好！:)
| 欢迎光临 计算化学公社 (http://bbs.keinsci.com/) |
Powered by Discuz! X3.3 |