Nearfield calculations in nSpheres and Qrad

Nearfield calculations in nSpheres and Qrad

Matthew Crane (guest) 12 Feb 2019 05:21

Hi DDSCAT Community,

I have two questions:

1. I'm using the "nSphere" target generation to produce a target without overlapping spheres. I'd like to include nearfield calculations to examine if there's any nearfield interaction between the spheres. For a single sphere, such as the ellipsoid nearfield example, the nearfield calculation extends the computational volume beyond an individual sphere, as specified in the ddscat.par file. However, I'm confused how this works for nSpheres, where dipoles are only placed at sphere locations. Would a nearfield calculation with fractional extensions of 0.5 in all directions expand the computational volume around each sphere or would it somehow expand the computational volume around the entire TUC? Assuming its the former, if I wanted to include sphere-sphere nearfield interactions, should I specify the fractional extension to include nearest neighbor spheres? Or, am I misinterpreting these things?

2. I've been trying to extract the Qrad and Qtorque from calculations. When I include everything (i.e. "DOTORQUE"), the qtable still doesn't include those values. What setting should I double check to make ddscat write the Qrad and Qtorque?

Thank you for the help!
-Matt

Re: Nearfield calculations in nSpheres and Qrad

draine 12 Feb 2019 14:10

1. When DDSCAT is directed to do narfield calculations, the "computational box" is extended beyond the circumscribing rectangular volume (i.e., the minimum rectangular volume containing ALL of the dipoles in the target structure, whatever the "shape option"). The amount of extension is specified by the fractional extension in +x direction, -x direction, +y direction, -y direction, +z direction, and -z direction (in the "Target Frame"). The "fractions" in the fractional exension are, e.g., Delta L_x/L_x where L_x is the length in the x direction of the original circumscribing rectangular volume, and Delta L_x is the requested extension the +x direction, and similarly for the other five fractional extension parameters.
If you only want fields *between* spheres in the SPHERES_N option, then you do not need to extend the circumscribing volume, e.g., you can use option

2 = NRFLD (to calculate both E and B)

with

0.0 0.0 0.0 0.0 0.0 0.0

on the line in ddscat.par specifying the fractional extension of the calc. vol.

2. The Q_pr and Q_Gamma vectors are printed out in the files wxxxryyykzzz.sca

See the example SPHERES_N_DOTORQ, where the output file w000r000k000.sca
includes the lines

Qsca*g(1) Qsca*g(2) Qsca*g(3) iter mxiter Nsca
JO=1: 6.3826E-01 -5.1775E-03 1.1536E-02 3 300 769
JO=2: 6.5400E-01 6.4557E-03 -2.7960E-03 3 300 769
mean: 6.4613E-01 6.3907E-04 4.3700E-03
Qtrqinc(1) Qtrqinc(2) Qtrqinc(3) Qtrqsca(1) Qtrqsca(2) Qtrqsca(3) Qtrq(1) Qtrq(2) Qtrq(3)
JO=1: -1.168E-02 2.601E-02 1.089E-01 1.698E-02 -3.945E-02 -4.500E-02 5.304E-03 -1.344E-02 6.395E-02
JO=2: 1.600E-02 -8.209E-02 5.072E-02 1.906E-02 -2.139E-02 -3.471E-02 3.506E-02 -1.035E-01 1.601E-02
mean: 2.158E-03 -2.804E-02 7.983E-02 1.802E-02 -3.042E-02 -3.985E-02 2.018E-02 -5.846E-02 3.998E-02

The net radiative torque is given by the last three columns labeled Qtrq(1-3).
The ddscat.par file in this directory is set up to do the torque calculation.

Matthew Crane (guest) 13 Feb 2019 07:27

Profesor Draine,

I understand—thank you for the help. I should add thank you for creating and updating DDSCAT! It was a huge part of my PhD work and I've continue to use it in my postdoc.

Thanks,
Matt

Matthew Crane (guest) 03 Mar 2019 22:59

Thank you very much for the help.

I've got one last question. What frame does DDSCAT return results for the Q_pr and Q_gamma?

For some context, I'm studying the effects of Q_torque when rotating a cylinder relative to the direction of plane wave propagation, such that it sits in the x-y plane. To do this, I'm varying the target rotation by sweeping theta in the ddscat.par file from 0 to 180 degrees. I'm trying to figure out if a positive torque in z pushes the "top" of the cylinder toward positive x or negative x.

Thanks for the help!

draine 15 Mar 2019 15:19

The components of Q_pr and Q_Gamma are given in the "Lab Frame" (see section 16 of the UserGuide)

PASS ACCA/CIA/CMA EXAMS,Payment after PASS results WhatsApp [phone]

lordjerom7 11 Sep 2022 23:50

PASS ACCA/CIA/CMA EXAMS,Payment after PASS results
PASS CIPP,CIPM,CIPT EXAMS. PAY AFTER PASS RESULTS

PASS GRE/GMAT EXAMS AND PAY AFTER CONFIRMATION OF PASS RESULTS

BUY PMI (PMP / CAPM / ACP / PBA, RMP, PgMP)PAY AFTER RESULTS

PASS ISACA CERTIFICATES [CISA, CISM,CRISC] PAY AFTER RESULTS

GET EC-COUNCIL CERTIFICATIONS(CEH practical ,CCISO) PAY AFTER RESULTS

BUY ISACA CERTIFICATES [CISA, CISM,CRISC] PAY AFTER RESULTS WhatsApp [phone]