[Wien] spin and orbital moments
angle (M,z) and angle (M,x) deg are THETA and PHI, respectively. Here is how the code calculates the Projection of M (for your crystal system). Your lattice constants a = b = c = 13.6697120 angstrom Your crystal angles alpha = beta = gamma = 60 deg = 1.04719755119660 rad M|| XMS1 = 1.000 XMS2 = 1.000 XMS3 = -1.000 XA=XMS1*a*sin(gamma) XB=XMS1*a*cos(gamma)+b*XMS2 XC=c*XMS3 XX=sqrt(XA**2+XB**2+XC**2) theta=acos(XC/XX) XX=sqrt(XA^2+XB^2) if XX < 1e-5 phi=0; else phi=acos(XA/XX) if abs(XB) > 1e-5 phi=phi*XB/abs(XB) end end M = sin(theta)*(cos(phi)*x+sin(phi)*y)+cos(theta)*z (equation from line 168 of code in $WIENROOT/SRC_lapwdm/output.f) Example for :ORB005: ORBITAL MOMENT: -0.03637 -0.06090 0.04160 PROJECTION ON M -0.08224 XA=1*13.6697120*sin(1.04719755119660) = 11.8383179 XB=1*13.6697120*cos(1.04719755119660)+13.6697120*1 = 20.504568 XC=13.6697120*-1 = -13.669712 XX=sqrt(11.8383179**2+20.504568**2+(-13.669712)**2) = 27.339424 theta=acos(-13.669712/27.339424) = 2.0943951 rad =*120 deg* XX=sqrt(11.8383179**2+20.504568**2) = 23.6766357 phi=acos(XA/XX) = acos(11.8383179/23.6766357) = 1.04719755 phi=phi*XB/abs(XB) = 1.04719755*20.504568/abs(20.504568) = 1.04719755 rad = *60 deg* M = *sin(2.0943951)*(cos(1.04719755)*-0.03637+sin(1.04719755)*-0.06090)+cos(2.0943951)*0.04160* = *-0.82223672* (has slight but acceptable round off error) Now you can confirm for yourself that all ORBxxx and SPIxxx are satisfied. On 7/1/2012 4:29 AM, foyevtsova at th.physik.uni-frankfurt.de wrote: > Dear Gavin, > > in case.outputdmup, for instance, I find only this information on angles: > > 120.0 60.0 angle (M,z), angle (M,x) deg > > Here below is a passage where this line comes from: > > SUBSTANCE= blebleble > s-o calc. M|| 1.00 1.00 -1.00 > > LATTICE = P > LATTICE CONSTANTS ARE= 13.6697120 13.6697120 13.6697120 > NUMBER OF ATOMS IN UNITCELL = 15 > MODE OF CALCULATION IS = RELA >BR1, BR2 > 0.56295 -0.18765 -0.18765 0.56295 -0.18765 -0.18765 > 0.0 0.53075 -0.26537 0.0 0.53075 -0.26537 > 0.0 0.0 0.45964 0.0 0.0 0.45964 >alpha test 1.047197551196601.04719755119660 > 1.04719755119660 > SO= T > Spin-polarized + s-o calculation, M|| 1.000 1.000 -1.000 >alpha test 1.047197551196601.04719755119660 > 1.04719755119660 > LATTICE:P >alpha test 1.047197551196601.04719755119660 > 1.04719755119660 > 120.0 60.0 angle (M,z), angle (M,x) deg > SYMMETRY OPERATIONS IN SPIN COORD. SYSTEM > > There is no information on THETA and PHI. > >> Do you have a case.outputdm, case.outputdmup, or case.outputdmdn file? >> Can you see if the THETA and PHI is different from that in case.outsymso? >> >> How to explain the 1st iteration ORB005, since sqrt((-0.08361)**2 + >> (-0.01872)**2 + (0.02851)**2) = +0.0903 != -0.06454 > Sorry, this is my mistake: what you see is the last iteration. The true > first iteration is > :ORB005: ORBITAL MOMENT: -0.03637 -0.06090 0.04160 PROJECTION ON M -0.08224 > > For these values, sqrt(x**2 + y**2 + z**2) indeed holds. Then, in the > converged solution the orbital moment deviates from M. > > Could it be that something is wrong in the code? > > > > >> For those angles, I also get 0.927 for SPI005 and -0.06356 for ORB005. >> If THETA and PHI in case.outputdm are slightly different, then both >> calculations could work out. >> >> Kind Regards >> >> On 6/29/2012 7:36 AM, Kateryna Foyevtsova wrote: >>> Dear Gavin, >>> >>> that's the point: sqrt(x**2 + y**2 + z**2) works! I indeed get 1.075 >>> when I insert my x, y and z into this equation! >>> >>> >From case.outsymso: >>> >>> THETA, PHI 1.57079632679490 0.955316618124509 >>> >>> and using your formula I get 0.927. >>> >>> Bests >>> >>> On 29/06/12 14:49, Gavin Abo wrote: That should be because the equation is not sqrt(x**2 + y**2 + z**2). The equation that it seems to use is sin(theta)*(cos(phi)*x+sin(phi)*y)+cos(theta)*z for both ORBxxx and SPIxxx. So, sin(theta)*(cos(phi)*0.46560+sin(phi)*0.80642)+cos(theta)*0.53749 = 1.075 (projection on the M axis). What are the values of phi and theta? I believe they are given in case.outputdm(up/dn). Hopefully the values satisfy the equation, else I must have overlooked something. On 6/29/2012 1:54 AM, Kateryna Foyevtsova wrote: > Dear Gavin, > > thanks a lot for your detailed answer and the very useful links! > > If ORBxxx and SPIxxx are in CCS, how to explain the fact that for, eg, > SPI005 in the first iteration > > sqrt(0.46560**2 + 0.80642**2 + 0.53749**2) = 1.075 > > ie, exactly the projection on the M axis. I would not expect that if > 0.46560, 0.80642 and 0.53749 were projections on the non-orthogonal > axes. That is for me the hardest thing to unde
[Wien] spin and orbital moments
Dear Gavin,
in case.outputdmup, for instance, I find only this information on angles:
120.0 60.0 angle (M,z), angle (M,x) deg
Here below is a passage where this line comes from:
SUBSTANCE= blebleble
s-o calc. M|| 1.00 1.00 -1.00
LATTICE = P
LATTICE CONSTANTS ARE= 13.6697120 13.6697120 13.6697120
NUMBER OF ATOMS IN UNITCELL = 15
MODE OF CALCULATION IS = RELA
BR1, BR2
0.56295 -0.18765 -0.18765 0.56295 -0.18765 -0.18765
0.0 0.53075 -0.26537 0.0 0.53075 -0.26537
0.0 0.0 0.45964 0.0 0.0 0.45964
alpha test 1.047197551196601.04719755119660
1.04719755119660
SO= T
Spin-polarized + s-o calculation, M|| 1.000 1.000 -1.000
alpha test 1.047197551196601.04719755119660
1.04719755119660
LATTICE:P
alpha test 1.047197551196601.04719755119660
1.04719755119660
120.0 60.0 angle (M,z), angle (M,x) deg
SYMMETRY OPERATIONS IN SPIN COORD. SYSTEM
There is no information on THETA and PHI.
> Do you have a case.outputdm, case.outputdmup, or case.outputdmdn file?
> Can you see if the THETA and PHI is different from that in case.outsymso?
>
> How to explain the 1st iteration ORB005, since sqrt((-0.08361)**2 +
> (-0.01872)**2 + (0.02851)**2) = +0.0903 != -0.06454
Sorry, this is my mistake: what you see is the last iteration. The true
first iteration is
:ORB005: ORBITAL MOMENT: -0.03637 -0.06090 0.04160 PROJECTION ON M -0.08224
For these values, sqrt(x**2 + y**2 + z**2) indeed holds. Then, in the
converged solution the orbital moment deviates from M.
Could it be that something is wrong in the code?
>
> For those angles, I also get 0.927 for SPI005 and -0.06356 for ORB005.
> If THETA and PHI in case.outputdm are slightly different, then both
> calculations could work out.
>
> Kind Regards
>
> On 6/29/2012 7:36 AM, Kateryna Foyevtsova wrote:
>> Dear Gavin,
>>
>> that's the point: sqrt(x**2 + y**2 + z**2) works! I indeed get 1.075
>> when I insert my x, y and z into this equation!
>>
>> >From case.outsymso:
>>
>> THETA, PHI 1.57079632679490 0.955316618124509
>>
>> and using your formula I get 0.927.
>>
>> Bests
>>
>> On 29/06/12 14:49, Gavin Abo wrote:
>>> That should be because the equation is not sqrt(x**2 + y**2 + z**2).
>>>
>>> The equation that it seems to use is
>>> sin(theta)*(cos(phi)*x+sin(phi)*y)+cos(theta)*z for both ORBxxx and
>>> SPIxxx.
>>>
>>> So, sin(theta)*(cos(phi)*0.46560+sin(phi)*0.80642)+cos(theta)*0.53749 =
>>> 1.075 (projection on the M axis).
>>>
>>> What are the values of phi and theta? I believe they are given in
>>> case.outputdm(up/dn). Hopefully the values satisfy the equation, else
>>> I
>>> must have overlooked something.
>>>
>>> On 6/29/2012 1:54 AM, Kateryna Foyevtsova wrote:
Dear Gavin,
thanks a lot for your detailed answer and the very useful links!
If ORBxxx and SPIxxx are in CCS, how to explain the fact that for, eg,
SPI005 in the first iteration
sqrt(0.46560**2 + 0.80642**2 + 0.53749**2) = 1.075
ie, exactly the projection on the M axis. I would not expect that if
0.46560, 0.80642 and 0.53749 were projections on the non-orthogonal
axes. That is for me the hardest thing to understand.
Best regards,
Kateryna
On 29/06/12 04:49, Gavin Abo wrote:
> 1) In which coordinate system are SPI005 and ORB005 given?
>
> In Appendix C (http://www.wien2k.at/reg_user/textbooks/) of "New
> notes
> about Hyperfinefield calculations (ps)", it mentions that the
> subroutine
> /couplx/ (of lapwdm) now calculates matrices of all components of
> spin
> and orbital momentum in the "crystal coordinate system
> (sx,sy,sz,lx,ly,lz)". Therefore, *I believe the x, y, and z values of
> SPIxxx and ORBxxx are also in the crystal coordinate system (CCS),
> while
> the M values ("PROJECTION ON M" values) are parallel to the
> magnetization. *
>
> If your good with reading fortan, you can look into the code. I don't
> full understand what is going on in the code, but I believe the
> "direction to M" (in your case: 1 1 -1) specified in case.inso is
> read
> in SRC_lapwdm/lapwdm.f. Then, the angles theta and phi between the
> "direction to M" and CCS are calculated in SRC_lapwdm/angle.f. Next,
> the
> x, y, and z values of SPIxxx and ORBxxx are calculated in the CCS.
> The
> x, y, and z values are written to case.outputdm(up/dn) and
> case.scfdm(up/dn), while a Cartesian to spherical equation [r =
> sin(theta)*(cos(phi)*x+sin(phi)y)+cos(theta)*z] is used to calculate
> the
> radius (M) using the x, y, and z, theta, and phi values before
> writing
> to the same output files as performed by SRC_lapwdm/output.f.
>
> 2) Why for the first iteration MMI005 is not even roughly eq
[Wien] spin and orbital moments
Dear Gavin, that's the point: sqrt(x**2 + y**2 + z**2) works! I indeed get 1.075 when I insert my x, y and z into this equation!
[Wien] spin and orbital moments
Dear Gavin,
thanks a lot for your detailed answer and the very useful links!
If ORBxxx and SPIxxx are in CCS, how to explain the fact that for, eg,
SPI005 in the first iteration
sqrt(0.46560**2 + 0.80642**2 + 0.53749**2) = 1.075
ie, exactly the projection on the M axis. I would not expect that if
0.46560, 0.80642 and 0.53749 were projections on the non-orthogonal
axes. That is for me the hardest thing to understand.
Best regards,
Kateryna
On 29/06/12 04:49, Gavin Abo wrote:
> 1) In which coordinate system are SPI005 and ORB005 given?
>
> In Appendix C (http://www.wien2k.at/reg_user/textbooks/) of "New notes
> about Hyperfinefield calculations (ps)", it mentions that the subroutine
> /couplx/ (of lapwdm) now calculates matrices of all components of spin
> and orbital momentum in the "crystal coordinate system
> (sx,sy,sz,lx,ly,lz)". Therefore, *I believe the x, y, and z values of
> SPIxxx and ORBxxx are also in the crystal coordinate system (CCS), while
> the M values ("PROJECTION ON M" values) are parallel to the
> magnetization. *
>
> If your good with reading fortan, you can look into the code. I don't
> full understand what is going on in the code, but I believe the
> "direction to M" (in your case: 1 1 -1) specified in case.inso is read
> in SRC_lapwdm/lapwdm.f. Then, the angles theta and phi between the
> "direction to M" and CCS are calculated in SRC_lapwdm/angle.f. Next, the
> x, y, and z values of SPIxxx and ORBxxx are calculated in the CCS. The
> x, y, and z values are written to case.outputdm(up/dn) and
> case.scfdm(up/dn), while a Cartesian to spherical equation [r =
> sin(theta)*(cos(phi)*x+sin(phi)y)+cos(theta)*z] is used to calculate the
> radius (M) using the x, y, and z, theta, and phi values before writing
> to the same output files as performed by SRC_lapwdm/output.f.
>
> 2) Why for the first iteration MMI005 is not even roughly equal to
> SPI005 + ORB005?
>
> SPIxxx is the spin moment calculated from selected electrons only
> (usually d or f).
>
> MMIxxx is the sum from all electrons (s, p, d and f states) inside the
> atomic sphere xxx.
>
> ORBxxx is the orbital magnetic moment.
>
> So*MMIxxx = SPIxxx + ORBxxx is not necessarily true.*
>
> See the reference links below for more information:
>
> http://zeus.theochem.tuwien.ac.at/pipermail/wien/2011-September/015296.html
> http://zeus.theochem.tuwien.ac.at/pipermail/wien/2008-April/010820.html
> http://zeus.theochem.tuwien.ac.at/pipermail/wien/2005-January/004399.html
>
> On 6/28/2012 9:18 AM, Kateryna Foyevtsova wrote:
>> Dear Wien2k developers,
>>
>> I use wien2k version 11.1 to run spin-polarized GGA+U calculations with
>> SO coupling for a molibdenum oxide.
>> The symmetry of the system is the following
>>
>> bleblebles-o calc. M|| 1.00 1.00 -1.00
>> P 15 2 P-
>> RELA
>> 13.669712 13.669712 13.669712 60.00 60.00 60.00
>>
>> As you see, I set magnetization axis to 1 1 -1, which should be in terms
>> of (non-orthogonal) lattice vectors.
>> With the help of xcrysden and case.outsymso, I can deduce that this
>> direction corresponds to the 0.577350, 0.816497, 0 direction in terms of
>> the cartesian global coordinate system.
>>
>> When I converge the electron density with (without using any previously
>> converged non-relativistic calculation)
>>
>> runsp_lapw -p -orb -so -dm
>>
>> I get the following data for the first and the last iteration on one of
>> the Mo atoms:
>>
>> 1. iteration:
>> :SPI005: SPIN MOMENT: 0.46560 0.80642 -0.53749 PROJECTION ON M
>> 1.07518
>> :ORB005: ORBITAL MOMENT: -0.08361 -0.01872 0.02851 PROJECTION ON M
>> -0.06454
>> :MMI005: MAGNETIC MOMENT IN SPHERE 5=1.86180
>>
>> last iteration (converged solution):
>> :SPI005: SPIN MOMENT: 0.61653 1.06239 -0.70860 PROJECTION ON M
>> 1.41804
>> :ORB005: ORBITAL MOMENT: -0.08361 -0.01872 0.02851 PROJECTION ON M
>> -0.06454
>> :MMI005: MAGNETIC MOMENT IN SPHERE 5=1.43149
>>
>> Now, I am struggling to understand two things:
>> 1) In which coordinate system are SPI005 and ORB005 given?
>> If they were given in the global cartesian coordinate system, they would
>> be parallel to 0.577350, 0.816497, 0, but they are not.
>>
>> 2) Why for the first iteration MMI005 is not even roughly equal to
>> SPI005 + ORB005?
>>
>> Thank you very much!
>> Kateryna Foyevtsova
>>
>> P.S. When I perform relativistic calculations starting with a
>> preconverged electron density of the non-relativistic solution I get the
>> same final result.
>> ___
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>
>
>
>
> ___
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
[Wien] spin and orbital moments
Do you have a case.outputdm, case.outputdmup, or case.outputdmdn file?
Can you see if the THETA and PHI is different from that in case.outsymso?
How to explain the 1st iteration ORB005, since sqrt((-0.08361)**2 +
(-0.01872)**2 + (0.02851)**2) = +0.0903 != -0.06454
For those angles, I also get 0.927 for SPI005 and -0.06356 for ORB005.
If THETA and PHI in case.outputdm are slightly different, then both
calculations could work out.
Kind Regards
On 6/29/2012 7:36 AM, Kateryna Foyevtsova wrote:
> Dear Gavin,
>
> that's the point: sqrt(x**2 + y**2 + z**2) works! I indeed get 1.075
> when I insert my x, y and z into this equation!
>
> >From case.outsymso:
>
> THETA, PHI 1.57079632679490 0.955316618124509
>
> and using your formula I get 0.927.
>
> Bests
>
> On 29/06/12 14:49, Gavin Abo wrote:
>> That should be because the equation is not sqrt(x**2 + y**2 + z**2).
>>
>> The equation that it seems to use is
>> sin(theta)*(cos(phi)*x+sin(phi)*y)+cos(theta)*z for both ORBxxx and SPIxxx.
>>
>> So, sin(theta)*(cos(phi)*0.46560+sin(phi)*0.80642)+cos(theta)*0.53749 =
>> 1.075 (projection on the M axis).
>>
>> What are the values of phi and theta? I believe they are given in
>> case.outputdm(up/dn). Hopefully the values satisfy the equation, else I
>> must have overlooked something.
>>
>> On 6/29/2012 1:54 AM, Kateryna Foyevtsova wrote:
>>> Dear Gavin,
>>>
>>> thanks a lot for your detailed answer and the very useful links!
>>>
>>> If ORBxxx and SPIxxx are in CCS, how to explain the fact that for, eg,
>>> SPI005 in the first iteration
>>>
>>> sqrt(0.46560**2 + 0.80642**2 + 0.53749**2) = 1.075
>>>
>>> ie, exactly the projection on the M axis. I would not expect that if
>>> 0.46560, 0.80642 and 0.53749 were projections on the non-orthogonal
>>> axes. That is for me the hardest thing to understand.
>>>
>>> Best regards,
>>> Kateryna
>>>
>>>
>>> On 29/06/12 04:49, Gavin Abo wrote:
1) In which coordinate system are SPI005 and ORB005 given?
In Appendix C (http://www.wien2k.at/reg_user/textbooks/) of "New notes
about Hyperfinefield calculations (ps)", it mentions that the subroutine
/couplx/ (of lapwdm) now calculates matrices of all components of spin
and orbital momentum in the "crystal coordinate system
(sx,sy,sz,lx,ly,lz)". Therefore, *I believe the x, y, and z values of
SPIxxx and ORBxxx are also in the crystal coordinate system (CCS), while
the M values ("PROJECTION ON M" values) are parallel to the
magnetization. *
If your good with reading fortan, you can look into the code. I don't
full understand what is going on in the code, but I believe the
"direction to M" (in your case: 1 1 -1) specified in case.inso is read
in SRC_lapwdm/lapwdm.f. Then, the angles theta and phi between the
"direction to M" and CCS are calculated in SRC_lapwdm/angle.f. Next, the
x, y, and z values of SPIxxx and ORBxxx are calculated in the CCS. The
x, y, and z values are written to case.outputdm(up/dn) and
case.scfdm(up/dn), while a Cartesian to spherical equation [r =
sin(theta)*(cos(phi)*x+sin(phi)y)+cos(theta)*z] is used to calculate the
radius (M) using the x, y, and z, theta, and phi values before writing
to the same output files as performed by SRC_lapwdm/output.f.
2) Why for the first iteration MMI005 is not even roughly equal to
SPI005 + ORB005?
SPIxxx is the spin moment calculated from selected electrons only
(usually d or f).
MMIxxx is the sum from all electrons (s, p, d and f states) inside the
atomic sphere xxx.
ORBxxx is the orbital magnetic moment.
So*MMIxxx = SPIxxx + ORBxxx is not necessarily true.*
See the reference links below for more information:
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2011-September/015296.html
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2008-April/010820.html
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2005-January/004399.html
On 6/28/2012 9:18 AM, Kateryna Foyevtsova wrote:
> Dear Wien2k developers,
>
> I use wien2k version 11.1 to run spin-polarized GGA+U calculations with
> SO coupling for a molibdenum oxide.
> The symmetry of the system is the following
>
> bleblebles-o calc. M|| 1.00 1.00
> -1.00
> P 15 2 P-
>RELA
>13.669712 13.669712 13.669712 60.00 60.00 60.00
>
> As you see, I set magnetization axis to 1 1 -1, which should be in
> terms
> of (non-orthogonal) lattice vectors.
> With the help of xcrysden and case.outsymso, I can deduce that this
> direction corresponds to the 0.577350, 0.816497, 0 direction in
> terms of
> the cartesian global coordinate system.
>
> When I converge the electron density with (without using any previously
> converged non-relativistic ca
[Wien] spin and orbital moments
That should be because the equation is not sqrt(x**2 + y**2 + z**2).
The equation that it seems to use is
sin(theta)*(cos(phi)*x+sin(phi)*y)+cos(theta)*z for both ORBxxx and SPIxxx.
So, sin(theta)*(cos(phi)*0.46560+sin(phi)*0.80642)+cos(theta)*0.53749 =
1.075 (projection on the M axis).
What are the values of phi and theta? I believe they are given in
case.outputdm(up/dn). Hopefully the values satisfy the equation, else I
must have overlooked something.
On 6/29/2012 1:54 AM, Kateryna Foyevtsova wrote:
> Dear Gavin,
>
> thanks a lot for your detailed answer and the very useful links!
>
> If ORBxxx and SPIxxx are in CCS, how to explain the fact that for, eg,
> SPI005 in the first iteration
>
> sqrt(0.46560**2 + 0.80642**2 + 0.53749**2) = 1.075
>
> ie, exactly the projection on the M axis. I would not expect that if
> 0.46560, 0.80642 and 0.53749 were projections on the non-orthogonal
> axes. That is for me the hardest thing to understand.
>
> Best regards,
> Kateryna
>
>
> On 29/06/12 04:49, Gavin Abo wrote:
>> 1) In which coordinate system are SPI005 and ORB005 given?
>>
>> In Appendix C (http://www.wien2k.at/reg_user/textbooks/) of "New notes
>> about Hyperfinefield calculations (ps)", it mentions that the subroutine
>> /couplx/ (of lapwdm) now calculates matrices of all components of spin
>> and orbital momentum in the "crystal coordinate system
>> (sx,sy,sz,lx,ly,lz)". Therefore, *I believe the x, y, and z values of
>> SPIxxx and ORBxxx are also in the crystal coordinate system (CCS), while
>> the M values ("PROJECTION ON M" values) are parallel to the
>> magnetization. *
>>
>> If your good with reading fortan, you can look into the code. I don't
>> full understand what is going on in the code, but I believe the
>> "direction to M" (in your case: 1 1 -1) specified in case.inso is read
>> in SRC_lapwdm/lapwdm.f. Then, the angles theta and phi between the
>> "direction to M" and CCS are calculated in SRC_lapwdm/angle.f. Next, the
>> x, y, and z values of SPIxxx and ORBxxx are calculated in the CCS. The
>> x, y, and z values are written to case.outputdm(up/dn) and
>> case.scfdm(up/dn), while a Cartesian to spherical equation [r =
>> sin(theta)*(cos(phi)*x+sin(phi)y)+cos(theta)*z] is used to calculate the
>> radius (M) using the x, y, and z, theta, and phi values before writing
>> to the same output files as performed by SRC_lapwdm/output.f.
>>
>> 2) Why for the first iteration MMI005 is not even roughly equal to
>> SPI005 + ORB005?
>>
>> SPIxxx is the spin moment calculated from selected electrons only
>> (usually d or f).
>>
>> MMIxxx is the sum from all electrons (s, p, d and f states) inside the
>> atomic sphere xxx.
>>
>> ORBxxx is the orbital magnetic moment.
>>
>> So*MMIxxx = SPIxxx + ORBxxx is not necessarily true.*
>>
>> See the reference links below for more information:
>>
>> http://zeus.theochem.tuwien.ac.at/pipermail/wien/2011-September/015296.html
>> http://zeus.theochem.tuwien.ac.at/pipermail/wien/2008-April/010820.html
>> http://zeus.theochem.tuwien.ac.at/pipermail/wien/2005-January/004399.html
>>
>> On 6/28/2012 9:18 AM, Kateryna Foyevtsova wrote:
>>> Dear Wien2k developers,
>>>
>>> I use wien2k version 11.1 to run spin-polarized GGA+U calculations with
>>> SO coupling for a molibdenum oxide.
>>> The symmetry of the system is the following
>>>
>>> bleblebles-o calc. M|| 1.00 1.00 -1.00
>>> P 15 2 P-
>>> RELA
>>> 13.669712 13.669712 13.669712 60.00 60.00 60.00
>>>
>>> As you see, I set magnetization axis to 1 1 -1, which should be in terms
>>> of (non-orthogonal) lattice vectors.
>>> With the help of xcrysden and case.outsymso, I can deduce that this
>>> direction corresponds to the 0.577350, 0.816497, 0 direction in terms of
>>> the cartesian global coordinate system.
>>>
>>> When I converge the electron density with (without using any previously
>>> converged non-relativistic calculation)
>>>
>>> runsp_lapw -p -orb -so -dm
>>>
>>> I get the following data for the first and the last iteration on one of
>>> the Mo atoms:
>>>
>>> 1. iteration:
>>> :SPI005: SPIN MOMENT: 0.46560 0.80642 -0.53749 PROJECTION ON M
>>> 1.07518
>>> :ORB005: ORBITAL MOMENT: -0.08361 -0.01872 0.02851 PROJECTION ON M
>>> -0.06454
>>> :MMI005: MAGNETIC MOMENT IN SPHERE 5=1.86180
>>>
>>> last iteration (converged solution):
>>> :SPI005: SPIN MOMENT: 0.61653 1.06239 -0.70860 PROJECTION ON M
>>> 1.41804
>>> :ORB005: ORBITAL MOMENT: -0.08361 -0.01872 0.02851 PROJECTION ON M
>>> -0.06454
>>> :MMI005: MAGNETIC MOMENT IN SPHERE 5=1.43149
>>>
>>> Now, I am struggling to understand two things:
>>> 1) In which coordinate system are SPI005 and ORB005 given?
>>> If they were given in the global cartesian coordinate system, they would
>>> be parallel to 0.577350, 0.816497, 0, but they are not.
>>>
>>> 2) Why for the first iteration MMI005 is not even roughly equal to
>>> SPI005 + ORB005?
>>>
>>> Thank
[Wien] spin and orbital moments
1) In which coordinate system are SPI005 and ORB005 given?
In Appendix C (http://www.wien2k.at/reg_user/textbooks/) of "New notes
about Hyperfinefield calculations (ps)", it mentions that the subroutine
/couplx/ (of lapwdm) now calculates matrices of all components of spin
and orbital momentum in the "crystal coordinate system
(sx,sy,sz,lx,ly,lz)". Therefore, *I believe the x, y, and z values of
SPIxxx and ORBxxx are also in the crystal coordinate system (CCS), while
the M values ("PROJECTION ON M" values) are parallel to the
magnetization. *
If your good with reading fortan, you can look into the code. I don't
full understand what is going on in the code, but I believe the
"direction to M" (in your case: 1 1 -1) specified in case.inso is read
in SRC_lapwdm/lapwdm.f. Then, the angles theta and phi between the
"direction to M" and CCS are calculated in SRC_lapwdm/angle.f. Next, the
x, y, and z values of SPIxxx and ORBxxx are calculated in the CCS. The
x, y, and z values are written to case.outputdm(up/dn) and
case.scfdm(up/dn), while a Cartesian to spherical equation [r =
sin(theta)*(cos(phi)*x+sin(phi)y)+cos(theta)*z] is used to calculate the
radius (M) using the x, y, and z, theta, and phi values before writing
to the same output files as performed by SRC_lapwdm/output.f.
2) Why for the first iteration MMI005 is not even roughly equal to
SPI005 + ORB005?
SPIxxx is the spin moment calculated from selected electrons only
(usually d or f).
MMIxxx is the sum from all electrons (s, p, d and f states) inside the
atomic sphere xxx.
ORBxxx is the orbital magnetic moment.
So*MMIxxx = SPIxxx + ORBxxx is not necessarily true.*
See the reference links below for more information:
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2011-September/015296.html
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2008-April/010820.html
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2005-January/004399.html
On 6/28/2012 9:18 AM, Kateryna Foyevtsova wrote:
> Dear Wien2k developers,
>
> I use wien2k version 11.1 to run spin-polarized GGA+U calculations with
> SO coupling for a molibdenum oxide.
> The symmetry of the system is the following
>
> bleblebles-o calc. M|| 1.00 1.00 -1.00
> P 15 2 P-
> RELA
> 13.669712 13.669712 13.669712 60.00 60.00 60.00
>
> As you see, I set magnetization axis to 1 1 -1, which should be in terms
> of (non-orthogonal) lattice vectors.
> With the help of xcrysden and case.outsymso, I can deduce that this
> direction corresponds to the 0.577350, 0.816497, 0 direction in terms of
> the cartesian global coordinate system.
>
> When I converge the electron density with (without using any previously
> converged non-relativistic calculation)
>
> runsp_lapw -p -orb -so -dm
>
> I get the following data for the first and the last iteration on one of
> the Mo atoms:
>
> 1. iteration:
> :SPI005: SPIN MOMENT: 0.46560 0.80642 -0.53749 PROJECTION ON M
> 1.07518
> :ORB005: ORBITAL MOMENT: -0.08361 -0.01872 0.02851 PROJECTION ON M
> -0.06454
> :MMI005: MAGNETIC MOMENT IN SPHERE 5=1.86180
>
> last iteration (converged solution):
> :SPI005: SPIN MOMENT: 0.61653 1.06239 -0.70860 PROJECTION ON M
> 1.41804
> :ORB005: ORBITAL MOMENT: -0.08361 -0.01872 0.02851 PROJECTION ON M
> -0.06454
> :MMI005: MAGNETIC MOMENT IN SPHERE 5=1.43149
>
> Now, I am struggling to understand two things:
> 1) In which coordinate system are SPI005 and ORB005 given?
> If they were given in the global cartesian coordinate system, they would
> be parallel to 0.577350, 0.816497, 0, but they are not.
>
> 2) Why for the first iteration MMI005 is not even roughly equal to
> SPI005 + ORB005?
>
> Thank you very much!
> Kateryna Foyevtsova
>
> P.S. When I perform relativistic calculations starting with a
> preconverged electron density of the non-relativistic solution I get the
> same final result.
> ___
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
-- next part --
An HTML attachment was scrubbed...
URL:
[Wien] spin and orbital moments
Dear Wien2k developers, I use wien2k version 11.1 to run spin-polarized GGA+U calculations with SO coupling for a molibdenum oxide. The symmetry of the system is the following bleblebles-o calc. M|| 1.00 1.00 -1.00 P 15 2 P- RELA 13.669712 13.669712 13.669712 60.00 60.00 60.00 As you see, I set magnetization axis to 1 1 -1, which should be in terms of (non-orthogonal) lattice vectors. With the help of xcrysden and case.outsymso, I can deduce that this direction corresponds to the 0.577350, 0.816497, 0 direction in terms of the cartesian global coordinate system. When I converge the electron density with (without using any previously converged non-relativistic calculation) runsp_lapw -p -orb -so -dm I get the following data for the first and the last iteration on one of the Mo atoms: 1. iteration: :SPI005: SPIN MOMENT: 0.46560 0.80642 -0.53749 PROJECTION ON M 1.07518 :ORB005: ORBITAL MOMENT: -0.08361 -0.01872 0.02851 PROJECTION ON M -0.06454 :MMI005: MAGNETIC MOMENT IN SPHERE 5=1.86180 last iteration (converged solution): :SPI005: SPIN MOMENT: 0.61653 1.06239 -0.70860 PROJECTION ON M 1.41804 :ORB005: ORBITAL MOMENT: -0.08361 -0.01872 0.02851 PROJECTION ON M -0.06454 :MMI005: MAGNETIC MOMENT IN SPHERE 5=1.43149 Now, I am struggling to understand two things: 1) In which coordinate system are SPI005 and ORB005 given? If they were given in the global cartesian coordinate system, they would be parallel to 0.577350, 0.816497, 0, but they are not. 2) Why for the first iteration MMI005 is not even roughly equal to SPI005 + ORB005? Thank you very much! Kateryna Foyevtsova P.S. When I perform relativistic calculations starting with a preconverged electron density of the non-relativistic solution I get the same final result.
