The ABINIT Users Mailing List ( CLOSED )

["Zhenhua Zeng" <myid520@163.com>]

Dear Masayoshi,
    Thank you for your suggestion. I have recalculated the bond energy of oxygen based on your suggestion. But those parameters have little influence on the result. Especially when I follow Test_v1/t09 and set occopt and occ to
occopt 2
occ 1.0  1.0  1.0  1.0  1.0  1.0  1.0  0.0
     1.0  1.0  1.0  1.0  1.0  0.0  0.0  0.0
the bond energy becomes larger. The Pseufopotentials I used are GGA (PBE) pseudopotential files from FHI code and other pseudopotentials downloaded from web site.
    In the ABINIT bibliographical database of web site, I find there are a lot of articles which deal with oxides and oxygenous compounds. The authors of these articles should be in the ABINIT users’ mailing list. I want to kwon whether they have tested oxygen before using it for predictions, which situations had been tested, and what about the result. The purpose of testing oxygen is that I want to predict the adsorption performance of oxygen on transition metals, such as adsorption energy. So I think that some performance of oxygen with respect to energy should be tested before predictions. I hope someone could help me. Thank you.
Best regards

Zhenhua Zeng
Hunan University, PR. china



> Hello,
> 
> I have little time, but wanted to point out several things:
> 1) Please read textbooks, about O2 molecule; 
>    O2 should be treated as 
>    nspol 2
>    spinat 0.0 0.0 1.0 
>           0.0 0.0 1.0
>    see Test_v1/t09.in for example to see how to set "occ"/"occopt"
>    (test cases including "oxygen" can be found by typing
>     grep -i oxygen Test*/*.in
>    Then, the Test_v1/t09 is noticed as such a example...)
>    (It will be a good idea to try Test_* in this occasion.)
> 
> 2) Which type of O pseudopotential(PP) is adopted ?
>    Sufficient ecut may depends on the type of the PP...
> 
> 3) How about adopting the special k-point (0.25, 0.25, 0.25) 
>    for simple cubic supercel ?
> 
> 4) Kindly read the netiquette 
>      http://www.abinit.org/community/?text=netiquette
>    and try to avoid HTML e-mails; in Plain Text, please...
> 
> Good luck,
> Masayoshi
> 
> On Sat, 4 Dec 2004 10:37:53 +0800 (CST)
> "Zhenhua Zeng" <myid520@163.com> wrote:
> 
> > Dear abinit users, 
> >     Now, I want to predict using oxygen. So I test some well-known situations, such as bond length and bond energy (atomisation energy) of oxygen. The error of bond length is less than 1%. But the error of bond energy is very large. The calculated result using every pseudo potential downloaded from the web cite is between 8.5ev and 9.7ev. But The experimental data is 5.12ev. The result calculated using VASP and CASTEP are between 5.5ev and 6.5ev. The input file is given at below which based on Tutorial\t21.in. In the input file, the acell and ecut have been tested for the convergence of calculation. I want to know why it differs so much from experimental data and other calculations Does there have something wrong in my input file, the method used in tutorial to calculate bond energy (atomisation energy) can$B!G(Bt be used here, Or there has something wrong else? In addition, Can I use oxygen for prediction only depends on the good tested result of bond length? Your help !
>  will be greatly appreciated!
> > Best regards.
> > Zhenhua Zeng
> > Hunan University, PR. china
> > 
> > ####  input files   ##############
> > #calculate oxygen molecular and atom
> > #common parameters
> > ndtset 2
> > ixc 1 or ixc 3 or ixc 7 or ixc 11 for different pseudo potential
> > acell 3*12  Bohr 
> > ecut  30.0  Hatree
> > nstep 100
> > ngkpt 2 2 2
> > kptopt 1
> > ntypat 1
> > znucl  8
> > diemac 1.0   
> > diemix 0.5   
> > 
> > #the parameters for oxygen atoms
> > nsppol1 2             ##SPin POLarization calculation of oxygen atom
> > spinat1 0.0 0.0 1.0
> > natom1 1           
> > typat1 1
> > xcart1 3*0.0        
> > toldfe1 1.0d-6
> > 
> > #The parameters for oxygen molecular
> > natom2 2   
> > typat2 1 1        
> > xcart2   -1.145 0.0 0.0    #the bond length of oxygen molecular is 
> >           1.145 0.0 0.0    #about 2.29 Bohr
> > ionmov2  3  
> > ntime2  100  
> > tolmxf2  5.0d-5 
> > toldff2  5.0d-6
> 
> 
> -- 
> ----------------------------------------------
> Masayoshi Mikami (Yokohama, Japan)
>   home:    mmikami@yk.rim.or.jp
>   webmail: mmikami_yk@yahoo.co.jp
> ----------------------------------------------
>

・AOC(冠捷)六大随心技,开启"随心视"代http://design.163.com/aocmonitor/index.html

・1.5G 全国最大空间――网易126免费邮箱 http://www.126.com 火热抢注中>>>

  体验30兆超大附件发送，享受100％卡巴斯基专业杀毒……更多超G感受等您来体验！