Jump to [Info. NCPS95] in Japanese
Here is supported for NCPS95(97)(2K)(How to get CD-R, How to use
NCPS95(97)(2K), New information about NCPS95(97)(2K), etc.).
CD-R distribution is now available.
(2/29, 2016)Pseudopotential data of Mn (using 1.00 0.50 0.50 for Non-PCC version) is "MNKBB.DAT". MNKBA.DAT (using "1.75 0.75 0.50" [No Warranty]) is for comparison with MNKBB.DAT.
(2/16, 2016)We have found the problem of data files in NCPS as follows. As for Ar,Ar_pcc,Ba_pcc,Ca,Ca_3p,Kr,Kr_3d,Mg_pcc,Ne,Rn_5d,Rn_pcc,S,S_TM,Sr_4p,Xe,Xe_4d(Name of directory), these pseudopotential data are s, p, d(non-local).
For example, although there is NETMD.DAT in the Ne directory, this is not used for the electronic structure calculation.
In other cases, CUD.DAT (an example of local pseudopotential data) as zero-byte data is not used (dummy data) because non-local pseudopotential data of Cu consist of s, p and d parts (not used CUD.DAT).
(5/19, 2015)Pseudo-wave functions (s, p and d) for Al, As, B, Co, Cu, Mg, N, O, Sb, Si, Te and Ti. [psw_051815.tar.gz](tar+gzip, 201.4 kb) [No Warranty].
(6/25, 2015)We have constructed pseudo-wave functions of C and Y (still not public).
(7/22, 2015)We have constructed pseudo-wave functions of Mn (still not public).
(12/17,2015)Pseudo-wave functions (s, p and d) for Al, As, B, Co, C,
Cu, Fe, Mg, Mn, N, O, Sb, Si, Te, Ti and Y. [psw_121715.tar.gz](tar+gzip,
716.8 kb) [No Warranty].
(3/12, 2015)Modified Mg pseudopotential(pcc, TM) has been constructed. The lattice mismatch of the TiN(001)/MgO(001) interface is improved from 3.7 % to 1.4 % by using this pseudopotential. It is possible to download new Mg pseudopotential [No Warranty] at NCPS2K,NCPS2K+ [page](www.nims.go.jp/cmsc).
(7/1, 2013)P pseudopotential(TM) has been constructed. It is possible to download new P pseudopotential [No Warranty] at NCPS2K,NCPS2K+ [page](www.nims.go.jp/cmsc).
(2/20, 2013)Mg pseudopotential(TM) with PCC has a ghost band at about 15 eV above the Fermi level. An influence on calculational results using this pseudopotential is small because the ghost band is at more than 1 Ry above the Fermi level. At present, we try to construct new Mg pseudopotential without ghost bannds.
(7/3, 2012)W pseudopotential(TM) with PCC has been modified in order to avoid a ghost band above at about 10 eV above the Fermi level. We succeed to construct modified W pseudopotential with PCC and without the ghost band.
I wish to acknowledge Dr. Suehara's suggestion about the ghost band.
(9/21, 2010)Li pseudopotential(TM) has been constructed. It is possible to download new Li pseudopotential [No Warranty] at NCPS2K,NCPS2K+ [page](www.nims.go.jp/cmsc).
(10/29, 2007)Modified Re pseudopotential has been constructed. It is possible to download modified Os and Re pseudopotentials (*New PS)[No Warranty] at NCPS2K,NCPS2K+ [page](www.nims.go.jp/cmsc).
(10/26, 2007)Re pseudopotential had a problem. The author is constructing updated and modified Re pseudopotential (with PCC) at present. The old and problematic Re pseudopotential will be replaced by new Re pseudopotential soon. Therefore, Please do not use old problematic Re pseudopotential.
Hg pseudopotential is problematic at present.
(4/26, 2007)It is possible to construct pseudopotentials with various conditions (cutoff radius, LDA types, with/without PCC, etc.). However, it is impossible to translate the data format of NCPS pseudopotentials into those of other packages at present.
(11/29, 2006)The distribution of NCPS2K CD-R is available at present.
(11/28, 2006)The distribution of NCPS2K CD-R will start soon.
(10/30, 2006)At present, it is impossible to send CD-R until the end of November, 2006 (August -> october -> november: postponed, 10/30, 2006).
(8/4, 2006)Please pay attention to treat compounds which use both s,p non-local and s,p,d non-local pseudopotentials. There is a possibility of some bugs in electronic structure calculation codes because a treatment of local pseudopotential with s,p non-local PS is different from that with s,p,d non-local PS.
(7/5, 2006)Hg and Re pseudopotentials had a problem. Updated and modified Hg and Re pseudopotentials (with PCC) will be constructed. The old and problematic Hg and Re pseudopotentials will be replaced by new Hg and Re pseudopotentials. Therefore, Please do not use these problematic Hg and Re pseudopotentials.
(2/21, 2006)As a result of recalculations, it is confirmed that the new pseudopotential of S (TM type) is correct.
(2/14, 2006)A new pseudopotential of S (TM type) is constructed.(Now recalculating)
(1/29, 2006)A new pseudopotential of Cl (TM type) is constructed.
(4/11, 2005)Periodic Table(png image, 181 kB) of NCPS2K+ is updated.
(2/18, 2005)New B pseudopotential is constructed as the cutoff radius is 1.25 a.u., and LDA type of von-Barth and Hedin. In old version, the cutoff radius is 1.75 a.u., and LDA type of Wigner. There are no discripancies of electronic and lattice properties using both pseudopotentials in test calculations.
(1/14, 2005)Os pseudopotential had a problem. Updated and modified Os pseudopotential (with PCC) has been constructed. The old and problematic Os pseudopotential is replaced by new Os pseudopotential. Therefore, Please do not use old problematic Os pseudopotential.
(8/20, 2003)[asse] is available to use NCPS2K.
"asse" is the package of the electronic structure calculation using norm-conserving pseudopotentials (available for NCPS2K, 8/26, 2003).
If someone wants CD-R recorded data of NCPS95(97)(2K), Please see
[below] and Please send me E-mail(kobayashi.kazuaki-@-nims.go.jp[Please
replace "-@-" with "@" for against spam-mail at 13. June, 2003.]).
It is possible to distribute the NCPS97(2K) database by [CD-R] media.
How to get CD-R [Head]
- It is impossible to download
NCPS95(97)(2K) data at present.
- Please E-mail me (kobayashi.kazuaki-@-nims.go.jp[Please
replace "-@-" with "@" for against spam-mail at 13. June, 2003.]) and
fill out in your E-mail as follows.
- At present, it is possible to send CD-R.
- Your Name.
- Your E-mail address.
- Your Address to send CD-R.
- Please add a sentence of "Please send me
CD-R." in your E-mail.
- If possible, Please answer me "What is your study?".
- If possible, Any coments or suggestions for NCPS95(97)(2K) or this WWW site.
Simple description of CD-R [Head]
- Why NCPS2K?, now is 2001. In Japan, the financial period of
year 2000(=2K) is from 4/1(2000) to 3/31(2001). A Japanese version has been only
distributed in this period. [NCPS2K is a trial
- Why NCPS97?, now is 1998. In Japan, the financial period of
year 1997 is from 4/1(1997) to 3/31(1998).
- There are no archived files on CD-R.
- These data recorded at CD-R are only numerical
data of norm-conserving pseudopotentials.
There are no any codes for the electronic
structure calculation at this CD-R.
- Because of compatibility for lengths of file names between UNIX
and Windows 95(MS-DOS), thses data are recorded by "joliet
format"(available for only Windows95,98,ME,NT[ver4.0],2000,XP).
(It may be possible to access and read data and
files on CD-R on new versions of linux[PC-UNIX].)
- In detail, please see a readme file "README.TXT" in a directory
of "/ps" on CD-R.
- (4/27,1998)A distributed CD-R media is labeled by NEAT CD
LABELER KIT [Microtech].
- Please take care using the labeled CD-R on high speed (x8, x12,
x24, x36, x40, etc.) CD drives or CD drives in a Note-type PC(10/19,
1999). Please see [No Warranty].
- (12/21, 2000)A distributed CD-R is directly printed at
present(Not use a label paper).
- How to use NCPS95(97)(2K) in detail is [here].
- If the sended CD-R is broken, impossible to access or has some
troubles, Please send me E-mail(kobayashi.kazuaki-@-nims.go.jp[Please
replace "-@-" with "@" for against spam-mail at 13. June, 2003.]) with a
detailed report of "What is wrong?".
- The author send you a new CD-R for NCPS95(97)(2K) as soon as
- A difference of recent version of NCPS95
and new version of [NCPS97 and 2K] are not so large with the exception of very old
version of NCPS95.
How to use NCPS95(97)(2K) (Simple description) ........[Detailed
- This CD-R is available only on Windows95,98,ME,NT(ver4.0),Windows2000,XP.
- If you have got CD-R, copy pseudopotential data from directories
of "/ps/buff" or "/ps/ps_database5" to your adequate directory on
PC(IBM PC-AT compatible machine).
- In a case of s, p non-local, it is necessary to copy two files.
One file is for non-local parts(KB separable pseudopotentials, a file
name indicated by a keyword "KB"), the other
is for local.
- Please pay attention to treat
the above local potential.
- In a case of s, p and d non-local, it is necessary to copy only
one files (KB separable pseudopotentials, a file name also indicated
by a keyword "KB").
- In this case, a local potential part is the same of
Vcore(r) in BHS paper.
- Parameters of Vcore(r) are here
or Information about NCPS95.
- Please pay attention to treat
the above local potential.
- In a case of considering PCC(partial core correction), a
PCC data file is added above each case.
- A list of file names of NCPS95 is [here].
- In some directories of elements, dummy or not important files
are recorded on the directory.
- More detailed information is [here].
- Calculated results by using NCPS95(97)(2K) are [here].
- Forbid to redistribute this CD-R
and all data (NCPS95(97)(2K) on CD-R to
other sites(Laboratory, Group, Institute, etc).
- Please see Copyright.
As for media of CD-R which are recorded
, they are FREE of charge on the distribution.
Some WWW data and documents on distributed CR-R are incorrect and
old. New and more correct information are always in WWW pages.
To construct new version pseudopotentials
considered a partial core correction(PCC) for 3d, 4d and 5d transition
New pseudopotentials(11/6, 1996, Update 3/26,
At present(3/26, 1998), the author finds that some pseudopotentials
of transition metals have ghost bands above the Fermi energy.
If a position of the ghost band is enough higher than that of the
Fermi energy over about 1 Ry, it is possible to neglect an effect of
For a case of less than 1 Ry, it is necessary to reconstruct and
test new pseudopotentials.
More detailed imformations will be uploaded here as soon as
Already(6/30,1998), most pseudopotentials of transition metals have
been modified and avoid ghost bands, and to check lattice properties
Most pseudopotentials of transition and noble metals with PCC are
modified to remove ghost bands at present(12/8, 1998), and add new
parameters to the below list.
For NCPS95,97,2K users, if you have a plan to use the
pseudopotentials of transition or noble metals, Please contact me (kobayashi.kazuaki-@-nims.go.jp[Please replace
"-@-" with "@" for against spam-mail at 13. June, 2003.]).
Pseudopotentials for F(8/4, 1997),Cl,Br,Au,Ir,Po,Pt,Ta,W are
Only Cl is constructed as a BHS type. The other is constructed as a
TM type. Pseudopotentials for d are treated as local for F,Cl and Br.
In the other pseudopotentials, Vcore(r) is a local pseudopotential(s,p
and d are non-local).
Al(TM type) is constructed at 3/23(1998).
Parameters and pseudopotential data of Bi,Cs,Hf,Hg,Os,Pb,Re and Tl
are appended(3/25, 1997).
Parameters of Vcore(r) for 5d transition metals with PCC are
tabulated as follows(7/29, 1998).
Parameters of Vcore(r) for F are the same of BHS.
As for F,Cl and Br, d parts including Vcore parts are local.
Parameters for Vcore(r) are as follows.
a1 a2 c1
Au 1.00 0.50 0.50
Bi 1.00 0.50 0.50
Br 1.75 0.75 0.50
Ca 1.00 0.50 0.50 (Addition at 7/14, 2008)
Cs 0.20 0.20 0.50
Hf 1.00 0.50 0.50
Hg 1.00 0.50 0.50
Ir 1.00 0.50 0.50
Os 1.00 0.50 0.50
Pb 1.00 0.50 0.50
Po 1.75 0.75 0.50
Pt 1.00 0.50 0.50
Re 1.00 0.50 0.50
Ta 1.00 0.50 0.50
Tl 1.00 0.50 0.50
W 1.00 0.50 0.50
K(3p) 0.50 0.50 0.50
Ti(3p) 1.00 0.50 0.50
Al(TM) 1.00 0.50 0.50
Ag(pcc) 1.00 0.50 0.50
Au(pcc) 1.00 0.50 0.50
Cd(pcc) 1.00 0.50 0.50
Cr(pcc) 1.75 0.75 0.50
Cu(pcc) 1.75 0.75 0.50
Hf(pcc) 0.50 0.50 0.50
Hg(pcc) 1.00 0.50 0.50
In(4d) 1.75 0.75 0.50 (4d, Addition at 7/9, 2012)
Ir(pcc) 0.50 0.50 0.50
Li(pcc) 1.00 0.50 0.50 (TM, Addition at 9/21, 2010)
Mn 1.00 0.50 0.50 <-- MNKBB.DAT (2/26, 2016)(for Non-PCC version, Revised, 2/29, 2016)
Mn(pcc) 1.00 0.50 0.50
Mo(pcc) 0.50 0.50 0.50
Nb(pcc) 0.50 0.50 0.50
Os(pcc) 0.50 0.50 0.50 <-- (Modified) 1.00 0.50 0.50 (*New PS)
Pd(pcc) 1.00 0.50 0.50
Pt(pcc) 0.50 0.50 0.50
Re(pcc) 0.50 0.50 0.50 <-- (Modified) 1.00 0.50 0.50 (*New PS)
Rh(pcc) 0.50 0.50 0.50
Ru(pcc) 1.00 0.50 0.50
Sc(pcc) 0.50 0.50 0.50
Ta(pcc) 0.50 0.50 0.50
Tc(pcc) 0.50 0.50 0.50
V(pcc) 1.00 0.50 0.50
W(pcc) 0.50 0.50 0.50
Y(pcc) 0.50 0.50 0.50
Zn(pcc) 1.75 0.75 0.50
Zr(pcc) 0.50 0.50 0.50
- (4/16, 2003)[Announce]NCPS2K --> NCPS2K+ (Version up by 12/31,
2003 <-- Give up! and postponed.[1/28, 2004])
- (1/31, 2003)[Summary of 1/24, 2003 problem]
- (No Problem when Ecut = 25 Ry)
Rh, Sc, Ti, Zr, Hg, Y, Hf, Mo, Nb, Ag, V, Cd, Ru, Tc, Po, Ta, Ir,
Au, Re, Pt, Pb(10 eV > Ef), Tl(10 eV > Ef), Ru, W, Os, He(?),
Ne(?), "(?)": Checking at present.
- (Problematics when Ecut = 25 Ry)
Cu, Ni(49 Ry > ), Zn, Cr(Slightly), Co(49 Ry > ), Fe(49 Ry
> ), Mn(Slightly), Pd(Slightly, Gamma point), Bi(10 eV >,
"10 eV > Ef": There is a problem for the electronic band
structure at about 10 eV above the Fermi level.
"Slightly": Problem is not serious.
"49 Ry > ": There is no problem when Ecut = 49 Ry.
- (1/24, 2003)At present, pseudopotentials of Rh, Ti, Sc (NCPS2K)
are investigated in 25 Ry and 89 k-points as fcc and there are no
anomaly of band structures (Continue investigations. <-- This may
not be a serious problem.).
- (1/23, 2003)In NCPS2K pseudopotentials for 3d transition metals
(include Cu), it should be noted that it is necessary to be more than
60 - 70 Ry of the energy cutoff (Ecut) in order to give correct band
structures and electronic propertties, etc.
It is found that the electronic band structure of Cu when Ecut = 25
Ry is not correct (Continue investigations at present).
- (11/2, 2001)NCPS2K is available for a trial version at present.
- (6/27, 2001)NCPS2K is still preparing at present. More detailed
information will be announced here.
- (5/26, 2000)The problem of (5/15, 2000) has been solved
- (5/15, 2000)The author found the data on NCPS95,97 CD-R have a
The problem is that a local potential is dummy in a case of s, p, d
non-local case for NCPS. In this case, a d local potential is dummy (No
data) file and Vcore(r) is only the local potential.
But in some cases, the d local potential is not dummy in spite of s,
p, d non-local. Using this local potential data may lead to incorrect
electronic structure calculational results.
Please delete the local potentials on the following elements of
- (1/4, 2000)NCPS2K is not prepared.
- (11/17, 1999)NCPS97 will be updated to NCPS2K(tentative name,
Year 2000 version) by 3/31, 2001(the end of the financial period of
year 2000 in Japan).
- (11/9, 1999)Any program packages (codes) are not recorded in
- (10/20, 1999)It is found that there is an error of input
parameter(cutoff radius) for hydrogen. A web [page]
as for this input parameter has already been modified.
- At present(10/1, 1999), pseudopotentials for Br, Pd, Ti and Zn
have a problem as for constructing local parts of them. The author is
checking whether this problem is serious or not in the electronic
structure calculation using above pseudopotentials(Br, Pd, Ti and
- (3/3,2000)As for Ti, there are no problems to use it.