24小时热门版块排行榜

返回列表

wqg1806

木虫 (著名写手)

科研农民工

应助: 14 (小学生)
金币: 1998.3
散金: 3307
红花: 11
沙发: 1
帖子: 2275
在线: 643.6小时
虫号: 588353
注册: 2008-08-22
性别: GG
专业: 凝聚态物性 II ：电子结构

[交流] 【求助】盼高手指点，NEB计算

我在计算过渡态的时候插入了4个点，计算收敛之后处理数据的时候发现找到的不是能量的鞍点。在这种情况下应该怎么办？是不是继续优化，找到鞍点？或者提高计算精度，去寻找鞍点？我把计算的精度设置以及得到的数据贴一下：SYSTEM =Nb O
ENCUT =400

Ionic relaxation
EDIFF = 1E-5  !  low accuracy
EDIFFG =-0.01  !force
#PREC    = high
#NELM    = 100    ! maximum electronic steps
#NELMIN = 4       !  minimum electronic steps
#NELMDL = -8    ! update of charge for 3 steps       (for continuation: 0)
#IBRION =  -1       !  use CG algorithm for ions
#IBRION = 2       !  use CG algorithm for ions
NSW    =1550    ! No. of steps for ion
#NSW       =0    ! No. of steps for ion
# new added lines:
AMIX = 0.05
#AMIX = 0.1
BMIX = 0.0001 ! almost zero, but 0 will crash some versions
AMIX_MAG = 0.3
#AMIX_MAG = 0.5
BMIX_MAG = 0.0001 ! almost zero, but 0 will crash some versions
#LPARD=.TRUE.
#NBMOD=-3
#EINT=-1 0
#ISIF=3
#ISYM=1
POTIM = 0.3

#add for NEB
IMAGES=4
SPRING=-5.0
LCLIMB=.TRUE.
LTANGENTOLD=.FALSE.
LDNEB=.FALSE.
IBRION=3

#IALGO = 48    ! RMM-DIIS algorithm for electrons
ALGO = Fast
LREAL =.TRUE.    ! evaluate projection operators in real space

#NPAR = 1
NSIM = 4    ! blocked algorithm update, four bands a time
GGA    = PE    ! PBE

DOS related values
LVTOT =  .TRUE.
计算收敛之后处理的数据：3    1.171351       -0.650848    1.218286
3.05 1.1930247    -0.677556873901334    1.24581005302295
3.1    1.2146984    -0.704827288335876    1.27010368993822
3.15 1.2363721    -0.732589228234031    1.29116691074581
3.2    1.2580458    -0.760772678526208    1.30899971544572
3.25 1.2797195    -0.789307624142813    1.32360210403796
3.3    1.3013932    -0.818124050014252    1.33497407652251
3.35 1.3230669    -0.847151941070934    1.34311563289939
3.4    1.3447406    -0.876321282243264    1.34802677316859
3.45 1.3664143    -0.90556205846165    1.34970749733011
3.5    1.388088       -0.9348042546565       1.34815780538395
3.55 1.4097617    -0.963977855758219    1.34337769733011
3.6    1.4314354    -0.993012846697216    1.33536717316859
3.65 1.4531091    -1.0218392124039       1.32412623289939
3.7    1.4747828    -1.05038693780867    1.30965487652251
3.75 1.4964565    -1.07858600784194    1.29195310403796
3.8    1.5181302    -1.10636640743411    1.27102091544572
3.85 1.5398039    -1.1336581215156       1.24685831074581
3.9    1.5614776    -1.1603911350168       1.21946528993822
3.95 1.5831513    -1.18649543286814    1.18884185302295
4    1.604825       -1.211901    1.154988

所以，3不是鞍点？CNEB不是可以自动找到能量鞍点的吗？

回复此楼