[Pw_forum] Tb09-MetaGGA convergence problem
Éric Germaneau
germaneau at sjtu.edu.cn
Thu Apr 9 09:29:57 CEST 2015
Some PP will work some won't.
Do everything go well using other GGA or LDA functionals?
What do you want to use TB09 for ? and Why?
It's very tricky to use it you know.
On 04/09/2015 02:46 PM, plgong wrote:
> Dear Éric Germaneau,
> I used your input file (Si.pbe-rrkj.UPF) to do a test, but it reported error and stopped lastly.
>
> k = 0.4583 0.4583 0.4583 band energies (ev):
>
> ************************************************************************
> ************************************************************************
> ************************************************************************
> ******************************************************
>
> highest occupied, lowest unoccupied level (ev): -1104.4202-1136.6919
>
> ....
> MKL ERROR: Parameter 4 was incorrect on entry to DLASCL.
> { 0, 0}: On entry to
> DSTEQR parameter number -14 had an illegal value
> { 1, 1}: On entry to
> DSTEQR parameter number -14 had an illegal value
>
> What is wrong?
>
>
> And I used the Si.pbe-n-nc.UPF and it still deverge (Error in routine c_bands (1):
> too many bands are not converged). In this test, I just replace pp file without changing
> other paremeters. I have done too much tests on Ecutoff and K_points, but the problem still holds.
>
>
>
> Please help me sovle it, thanks in anvance!
>
>
>
>
>
>
>
>
>> -----原始邮件-----
>> 发件人:"Éric Germaneau" <germaneau at sjtu.edu.cn>
>> 发送时间:2015年4月9日 星期四
>> 收件人:"PWSCF Forum" <pw_forum at pwscf.org>
>> 抄送:
>> 主题:Re: [Pw_forum] Tb09-MetaGGA convergence problem
>>
>> well, I don't think your input file is suitable.
>> First, IONS part is useless for scf calculation, second your ecutwfc is likely too small.
>> What people usually do is to perform runs with different value of ecutwfc and ecutrho and see when the calculation get converged.
>> I don't think you have read well enough the tutorials available online.
>> Before using TB09 functional you have to make sure everything goes well with a regular one.
>> You definitely have to read more about what you are doing.
>>
>>
>> On 04/08/2015 01:53 PM, plgong wrote:
>>
>>> Dear Éric Germaneau,
>>> Thanks for you reply.
>>>
>>> My inputfile is,
>>> &CONTROL
>>> calculation = 'scf' ,
>>> restart_mode = 'from_scratch' ,
>>> outdir = './tmp' ,
>>> pseudo_dir = './' ,
>>> verbosity='high',
>>> prefix = 'graphene' ,
>>> nstep=200
>>> /
>>> &SYSTEM
>>> ibrav =0,
>>> nat = 2
>>> ntyp = 1,
>>> celldm(1)=1.89,
>>> ecutwfc= 30, input_dft='tb09'
>>> nosym = .false. ,
>>> tot_charge = 0.000000,
>>> occupations = 'fixed', nbnd=24
>>> /
>>> &ELECTRONS
>>> conv_thr = 1.D-5 ,
>>> mixing_mode = 'plain' ,
>>> mixing_beta = 2.0D0 , mixing_ndim=12
>>> diagonalization = 'cg' ,
>>> /
>>> &IONS
>>> ion_dynamics = 'bfgs' ,
>>> pot_extrapolation = 'second_order' ,
>>> wfc_extrapolation = 'second_order' ,
>>> /
>>> ATOMIC_SPECIES
>>> Si 12.01000 Si.pz-vbc.UPF
>>> K_POINTS automatic
>>> 12 12 12 1 1 1
>>> CELL_PARAMETERS
>>> 0.000000000000000 2.715350000000000 2.715350000000000
>>> 2.715350000000000 -0.000000000000001 2.715350000000000
>>> 2.715350000000000 2.715349999999999 -0.000000000000000
>>> ATOMIC_POSITIONS (crystal)
>>> Si 0.0000000000000000 0.0000000000000000 0.0000000000000000
>>> Si 0.2500000000000000 1.2500000000000002 0.2500000000000001
>>>
>>> My outfile is,
>>> Program PWSCF v.5.1.2 starts on 8Apr2015 at 13:55:52
>>>
>>> This program is part of the open-source Quantum ESPRESSO suite
>>> for quantum simulation of materials; please cite
>>> "P. Giannozzi et al., J. Phys.:Condens. Matter 21 395502 (2009);
>>> URL http://www.quantum-espresso.org",
>>> in publications or presentations arising from this work. More details at
>>> http://www.quantum-espresso.org/quote
>>>
>>> Parallel version (MPI), running on 1 processors
>>> Waiting for input...
>>> Reading input from standard input
>>> Warning: card &IONS ignored
>>> Warning: card ION_DYNAMICS = 'BFGS' , ignored
>>> Warning: card POT_EXTRAPOLATION = 'SECOND_ORDER' , ignored
>>> Warning: card WFC_EXTRAPOLATION = 'SECOND_ORDER' , ignored
>>> Warning: card / ignored
>>> Message from routine read_cards :
>>> DEPRECATED: no units specified in CELL_PARAMETERS card
>>>
>>> Current dimensions of program PWSCF are:
>>> Max number of different atomic species (ntypx) = 10
>>> Max number of k-points (npk) = 40000
>>> Max angular momentum in pseudopotentials (lmaxx) = 3
>>>
>>> IMPORTANT: XC functional enforced from input :
>>> Exchange-correlation = TB09 ( 0 0 0 0 0 3)
>>> Any further DFT definition will be discarded
>>> Please, verify this is what you really want
>>>
>>>
>>> Subspace diagonalization in iterative solution of the eigenvalue problem:
>>> a serial algorithm will be used
>>>
>>>
>>> G-vector sticks info
>>> --------------------
>>> sticks: dense smooth PW G-vecs: dense smooth PW
>>> Sum 433 433 151 5985 5985 1243
>>>
>>>
>>>
>>> bravais-lattice index = 0
>>> lattice parameter (alat) = 1.8900 a.u.
>>> unit-cell volume = 270.3291 (a.u.)^3
>>> number of atoms/cell = 2
>>>
>>>
>>> Best wishes
>>> P L Gong
>>>
>>>
>>>> -----原始邮件-----
>>>> 发件人:"Éric Germaneau" <germaneau at sjtu.edu.cn>
>>>> 发送时间:2015年4月8日 星期三
>>>> 收件人:"PWSCF Forum" <pw_forum at pwscf.org>
>>>> 抄送:
>>>> 主题:Re: [Pw_forum] Tb09-MetaGGA convergence problem
>>>>
>>>> i'm not sure whether you are using the TB09, check the beginning of the outfile.
>>>> As soon as I've time, I'll make the test myself.
>>>> Can you share your input file ?
>>>> I think you are using TPSS pseudopotential not TB09.
>>>>
>>>>
>>>> On 04/08/2015 09:05 AM, plgong wrote:
>>>>
>>>>
>>>>> Dear Éric Germaneau,
>>>>> I did a test on gap of Silicon using tb09. This time I used a another pseudopotential (Si.tpss-mt.UPF)
>>>>> and it converged lastly. But the gap converged to 1.75 eV, larrger than experimental results （1.17eV）.
>>>>> I saw a literature on the gap calculaion with MBJ (tb09) for silicon, and it showed that theoretical results coincide with that in experiment. I an confused.
>>>>> A few days ago, I tried to installed the libxc linked to QE, and my procedure was the following,
>>>>> 1. intall libxc;
>>>>> 2. install QE and modify make.sys,
>>>>> (DFLAGS = -D__LIBXC -D__INTEL -D__FFTW -D__MPI -D__PARA -D__SCALAPACK $(MANUAL_DFLAGS
>>>>> LD_LIBS = -L/home/plgong/software/QE512/libxc-2.2.2/lib -lxcf90 -lxc)
>>>>> The above make QE linkig libxc sucessfully.
>>>>> But, I have a question: I do not understand the tips
>>>>> 1. move to libxc, getting rid of all duplicated functionals, keeping only
>>>>> those that are not available in libxc;
>>>>> 2. add the possibility to use any functional from libxc (currently only a few
>>>>> can be used). This may require serious restructuring work in our XC
>>>>> functional (il-)logic, that has become clumsy, obscure and redundant.
>>>>>
>>>>> Whether the above installation leads to larger gap in my test?
>>>>>
>>>>>
>>>>> Best wishes
>>>>> P L Gong
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>> -----原始邮件-----
>>>>>> 发件人:"Éric Germaneau" <germaneau at sjtu.edu.cn>
>>>>>> 发送时间:2015年4月8日 星期三
>>>>>> 收件人:"PWSCF Forum" <pw_forum at pwscf.org>
>>>>>> 抄送:
>>>>>> 主题:Re: [Pw_forum] Tb09-MetaGGA convergence problem
>>>>>>
>>>>>> I'll install it and make the test.
>>>>>> Just give me some little time please.
>>>>>>
>>>>>>
>>>>>> On 04/07/2015 08:44 PM, plgong wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>> Dear pwscf_user,
>>>>>>> Recently, I used the tb09 (Meta-GGA) in the version of QE5.1.2, but it aways did not converge.
>>>>>>> The error is like,
>>>>>>> 'Error in routine cdiaghg (23):
>>>>>>> eigenvectors failed to converge'
>>>>>>>
>>>>>>> I have performed too much tests on Ecutoff, k_points, degauss, as well as reduce mixing. What should I do?
>>>>>>>
>>>>>>>
>>>>>>> My input file is,
>>>>>>>
>>>>>>> &CONTROL
>>>>>>> calculation = 'scf' ,
>>>>>>> restart_mode = 'from_scratch' ,
>>>>>>> outdir = './tmp' ,
>>>>>>> pseudo_dir = './' ,
>>>>>>> verbosity='high',
>>>>>>> prefix = 'graphene' ,
>>>>>>> /
>>>>>>> &SYSTEM
>>>>>>> ibrav =0,
>>>>>>> nat = 2
>>>>>>> ntyp = 1,
>>>>>>> celldm(1)=1.89,
>>>>>>> ecutwfc= 60
>>>>>>> input_dft='tb09'
>>>>>>> nosym = .false. ,
>>>>>>> tot_charge = 0.000000,
>>>>>>> occupations = 'fixed', nbnd=24
>>>>>>> /
>>>>>>> &ELECTRONS
>>>>>>> conv_thr = 1.D-7 ,
>>>>>>> mixing_mode = 'plain' ,
>>>>>>> mixing_beta = 0.1D0 , mixing_ndim=12
>>>>>>> diagonalization = 'cg' ,
>>>>>>> /
>>>>>>> &IONS
>>>>>>> ion_dynamics = 'bfgs' ,
>>>>>>> pot_extrapolation = 'second_order' ,
>>>>>>> wfc_extrapolation = 'second_order' ,
>>>>>>> /
>>>>>>> ATOMIC_SPECIES
>>>>>>> Si 12.01000 Si.pz-vbc.UPF
>>>>>>> K_POINTS automatic
>>>>>>> 12 12 12 0 0 0
>>>>>>> CELL_PARAMETERS
>>>>>>> 0.000000000000000 2.715350000000000 2.715350000000000
>>>>>>> 2.715350000000000 -0.000000000000001 2.715350000000000
>>>>>>> 2.715350000000000 2.715349999999999 -0.000000000000000
>>>>>>> ATOMIC_POSITIONS (crystal)
>>>>>>> Si 0.0000000000000000 0.0000000000000000 0.0000000000000000
>>>>>>> Si 0.2500000000000000 1.2500000000000002 0.2500000000000001
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> PL Gong
>>>>>>> --
>>>>>>>
>>>>>>> ====================================================
>>>>>>> Addr: Institute of Solid State Physics, Chinese Academy of
>>>>>>> Sciences, Hefei, Anhui 230031, China
>>>>>>> Tel: +86-551-65591591(office), 18756086113(cell phone)
>>>>>>> Email: plgong at theory.issp.ac.cn
>>>>>>> ========================================================================
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Pw_forum mailing list
>>>>>>> Pw_forum at pwscf.org
>>>>>>> http://pwscf.org/mailman/listinfo/pw_forum
>>>>>>> --
>>>>>>> Éric Germaneau (艾海克), Specialist
>>>>>> Center for High Performance Computing
>>>>>> Shanghai Jiao Tong University
>>>>>> Room 205 Network Center, 800 Dongchuan Road, Shanghai 200240 China
>>>>>> Email:germaneau at sjtu.edu.cnMobi:+86-136-4161-6480 http://hpc.sjtu.edu.cn
>>>>>> --
>>>>>> ====================================================
>>>>> Addr: Institute of Solid State Physics, Chinese Academy of
>>>>> Sciences, Hefei, Anhui 230031, China
>>>>> Tel: +86-551-65591591(office), 18756086113(cell phone)
>>>>> Email: plgong at theory.issp.ac.cn
>>>>> ========================================================================
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Pw_forum mailing list
>>>>> Pw_forum at pwscf.org
>>>>> http://pwscf.org/mailman/listinfo/pw_forum
>>>>> --
>>>> Éric Germaneau (艾海克), Specialist
>>>> Center for High Performance Computing
>>>> Shanghai Jiao Tong University
>>>> Room 205 Network Center, 800 Dongchuan Road, Shanghai 200240 China
>>>> M:germaneau at sjtu.edu.cnP:+86-136-4161-6480 W:http://hpc.sjtu.edu.cn
>>>>
>>>> --
>>> ====================================================
>>> Addr: Institute of Solid State Physics, Chinese Academy of
>>> Sciences, Hefei, Anhui 230031, China
>>> Tel: +86-551-65591591(office), 18756086113(cell phone)
>>> Email: plgong at theory.issp.ac.cn
>>> ========================================================================
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Pw_forum mailing list
>>> Pw_forum at pwscf.org
>>> http://pwscf.org/mailman/listinfo/pw_forum
>> --
>> Éric Germaneau (艾海克), Specialist
>> Center for High Performance Computing
>> Shanghai Jiao Tong University
>> Room 205 Network Center, 800 Dongchuan Road, Shanghai 200240 China
>> M:germaneau at sjtu.edu.cnP:+86-136-4161-6480 W:http://hpc.sjtu.edu.cn
> --
>
> ====================================================
> Addr: Institute of Solid State Physics, Chinese Academy of
> Sciences, Hefei, Anhui 230031, China
> Tel: +86-551-65591591(office), 18756086113(cell phone)
> Email: plgong at theory.issp.ac.cn
> ========================================================================
>
>
>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
--
Éric Germaneau (艾海克), Specialist
Center for High Performance Computing
Shanghai Jiao Tong University
Room 205 Network Center, 800 Dongchuan Road, Shanghai 200240 China
M:germaneau at sjtu.edu.cn P:+86-136-4161-6480 W:http://hpc.sjtu.edu.cn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150409/a9e66528/attachment.html>
More information about the users
mailing list