[Pw_forum] 答复: How to set the ensemble when doing molecule

Tue Jun 9 10:17:09 CEST 2009

It is possible to do MD with initial T=0K? it seems the tempw can not be set
to zero.

-----邮件原件-----
发件人: Lu Yunhao [mailto:phylyh at nus.edu.sg] 
发送时间: 2009年6月9日 16:14
收件人: 'pw_forum at pwscf.org'
主题: Re: [Pw_forum] How to set the ensemble when doing molecule

Yes, you are quite right. Now, I just try to compare results with different
parameter. 

If I set ion_temperature='not_controlled', the nraise=1, and tolp=100, are
still used or have some other meaning. When I do NVE MD in E-field, the
following setting is right?

    ion_dynamics='verlet',
    ion_temperature= 'rescale-v'  ,
    tempw=300.D0,
    nraise=1,
    tolp=100,

Thanks a lot 

Best wishes,
Yunhao
NUS

Message: 2
Date: Mon, 8 Jun 2009 22:34:55 -0400
From: Axel Kohlmeyer <akohlmey at cmm.chem.upenn.edu>
Subject: Re: [Pw_forum] How to set the ensemble when doing molecule
To: PWSCF Forum <pw_forum at pwscf.org>
Message-ID:
	<7b6913e90906081934p534fa40v2660e90a90b6e492 at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

On Mon, Jun 8, 2009 at 9:53 PM, Lu Yunhao<phylyh at nus.edu.sg> wrote:
> If I set ion_temperature= 'not_controlled', the tempw is still used and
> meaningful?

yes. it will set the initial temperature.

> I attach part of my input file as follows. Is it right? By the way, I add
> electric field along Z direction.

if you add an electric field, you _don't_ do an NVE ensemble
by not controlling the temperature. the external field will create
energy, i.e. you'll have a non-equilibrium system.

>
> &CONTROL
> ? ? ? ? ? ? ? ? title = 'MD-contT' ,
> ? ? ? ? ? ? ? ? calculation = 'md' ,
> ? ? ? ? ? ? ? ?prefix='MD-contT',
> ? ? ? ? ? ? ? ?restart_mode = 'from_scratch' ,
> ? ? ? ? ? ? ? ? tefield=.true.,
> ? ? ? ? ? ? ? ? ?outdir = '/home/phylyh/Au_gr-check/MD-contT/tmp' ,
> ? ? ? ? ? ? ? ? ?pseudo_dir = '/home/phylyh/pseudo_dir' ,
> ? ? ? ? ? ? ? ? dt=40,
> ? ? ? ? ? ? ? ? nstep=500,
> &IONS
> ? ?ion_dynamics='verlet',
> ? ?ion_temperature='rescale-v',
> ? ?tempw=300.D0,
> ? ?nraise=1,
> ? ?tolp=100,

if you use rescale and tolp, you are not in a defined ensemble
either. since you remove or add energy to your system in an
arbitrary way.

cheers,
   axel.

> ? ?pot_extrapolation='second-order',
> ? ?wfc_extrapolation='second-order'
> /
>
>
>
> Message: 5
> Date: Sat, 06 Jun 2009 10:37:16 -0400
> From: Axel Kohlmeyer <akohlmey at cmm.chem.upenn.edu>
> Subject: Re: [Pw_forum] How to set the ensemble when doing molecule
> ? ? ? ?dynamics calculation with pw.x
> To: PWSCF Forum <pw_forum at pwscf.org>
> Message-ID: <1244299036.3912.9870.camel at zero>
> Content-Type: text/plain
>
> On Sat, 2009-06-06 at 20:13 +0800, Lu Yunhao wrote:
>> Dear All
>
>> Can the pw.x code do molecule dynamic calculation with NVE ensemble?
>
> yes, of course!
>
>> I search the Manu and can not find any descriptions about ensemble.
>
> well, all you have to do is to a) have a constant number of atoms,
> b) a constant volume, and c) _not_ control the temperature.
>
> a) is trivial (you cannot add or remove atoms anyways)
> b) is simple, too (this is the default)
> c) only requires a look at the documentation of the &ions
> ? namelist. check out "tempw"
>
>
> it cannot get much simpler.
> ? ?axel.
>>
>>
>> Best wishes,
>>
>> Yunhao
>>
>> Resear Fellow
>>
>> NUS
>>
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>
> --
> =======================================================================
> Axel Kohlmeyer ? akohlmey at cmm.chem.upenn.edu ? http://www.cmm.upenn.edu
> ? Center for Molecular Modeling ? -- ? University of Pennsylvania
> Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
> tel: 1-215-898-1582, ?fax: 1-215-573-6233, ?office-tel: 1-215-898-5425
> =======================================================================
> If you make something idiot-proof, the universe creates a better idiot.
>
>
>
> ------------------------------
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.

------------------------------

Message: 3
Date: Mon, 8 Jun 2009 20:03:17 -0700 (PDT)
From: loc duong ding <mambom1902 at yahoo.com>
Subject: Re: [Pw_forum] Problem with parallel running by mpich2
To: pw_forum at pwscf.org
Message-ID: <338657.51336.qm at web38804.mail.mud.yahoo.com>
Content-Type: text/plain; charset=iso-8859-1

> doesn't it even print the first dozen of lines? If it does, it may just
be? 
> the usual problem with input redirection: try feeding the input file by? 
> using the syntax
>?? mpich2 [...] pw.x -in input.in [...]
> which is more parallel safe.
> 
> regards
> 
> 
> -- 
> Lorenzo Paulatto

It is solved.?Thank you very much, Prof. Lorenzo Paullatto.

Sincerely,?-----------------------------------------------
Loc Duong Dinh
Ms-Ph.D Student
Sungkyunkwan Advanced Institute of Nanotechnology,
Sungkyunkwan University,
Suwon, 440-746, Korea 
Email: mambom1902 at yahoo.com 

------------------------------

Message: 4
Date: Mon, 8 Jun 2009 23:03:40 -0700 (PDT)
From: loc duong ding <mambom1902 at yahoo.com>
Subject: [Pw_forum] Problem when compile Quantum espresso (CVS)
To: PWscf forum <pw_forum at pwscf.org>
Message-ID: <700031.57815.qm at web38803.mail.mud.yahoo.com>
Content-Type: text/plain; charset=iso-8859-1

?Dear developers and users,
?
I complie QE (CVS version) with my PC with Ferdora linux and Cywin without
any problem.?
However, when I try to complie with my cluster (Centos, i686-pc-linux-gnu,?
architecture ia32), I have a problem. The error message is:
?

In file casino2upf.f90:98?
????? REAL*8, ALLOCATABLE :: wavefunc(:)
?????????????????????? 1
?Error: Attribute at (1) is not allowed in a TYPE definition
??In file casino2upf.f90:304
???????? IF ( ALLOCATED(mtail%wavefunc) ) THEN
???????????????????????????????????? 1
?Error: 'wavefunc' at (1) is not a member of the 'wavfun_list' structure
??In file casino2upf.f90:308
??????????? ALLOCATE( mtail%wavefunc(mesh_) )
?????????????????????????????????? 1
?Error: 'wavefunc' at (1) is not a member of the 'wavfun_list' structure
??In file casino2upf.f90:309
???????? ELSE
??????????? 1
?Error: Unexpected ELSE statement at (1)
??In file casino2upf.f90:310
??????????? ALLOCATE( mtail%wavefunc(mesh_) )
?????????????????????????????????? 1
?Error: 'wavefunc' at (1) is not a member of the 'wavfun_list' structure
??In file casino2upf.f90:311
???????? END IF
?????????? 1
?Error: Expecting END DO statement at (1)
??In file casino2upf.f90:317
???????? READ(j, *, err=300) (mtail%wavefunc(ir),ir=1,mesh_)
?????????????????????????? 1
?Error: Expected variable in READ statement at (1)
??In file casino2upf.f90:352
???????? chi_(ir:,i) = mptr%wavefunc(ir)
?????????????????????????????????? 1
?Error: 'wavefunc' at (1) is not a member of the 'wavfun_list' structure
??In file casino2upf.f90:354
????? deallocate( mptr%wavefunc )
????????????????????????????? 1
?Error: 'wavefunc' at (1) is not a member of the 'wavfun_list' structure
?make[1]: *** [casino2upf.o] Error 1
?make[1]: Leaving directory `/home/loc/espresso/upftools'
?make: *** [upf] Error 2
?
I don't know where this error come from. Could you give me some instructions
for?
solving this error? I am running QE4.0 in this cluster.
?
Sincerely,?

Loc Duong Dinh
Ms-Ph.D Student
Sungkyunkwan Advanced Institute of Nanotechnology,
Sungkyunkwan University,
Suwon, 440-746, Korea 
Email: mambom1902 at yahoo.com ?

------------------------------

Message: 5
Date: Tue, 9 Jun 2009 08:18:11 +0200
From: Paolo Giannozzi <giannozz at democritos.it>
Subject: Re: [Pw_forum] Problem when compile Quantum espresso (CVS)
To: PWSCF Forum <pw_forum at pwscf.org>
Message-ID: <C8AFF3B6-6380-4360-AC53-73334A50B2BC at democritos.it>
Content-Type: text/plain; charset=US-ASCII; format=flowed

On Jun 9, 2009, at 8:03 , loc duong ding wrote:

>       REAL*8, ALLOCATABLE :: wavefunc(:)
>                        1
>  Error: Attribute at (1) is not allowed in a TYPE definition

allocatable arrays in derived types are not allowed in
fortran-90 standard but only in fortran-95 standard
(e.g. old versions of gfortran do not like them). You
need a fortran-95 compiler to complile recent
versions of q-e.

Paolo
---
Paolo Giannozzi, Dept of Physics, University of Udine
via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222

------------------------------

_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org
http://www.democritos.it/mailman/listinfo/pw_forum

End of Pw_forum Digest, Vol 24, Issue 12
****************************************