search

SasView / sasmodels

Package for calculation of small angle scattering models using OpenCL.
BSD 3-Clause "New" or "Revised" License
15 stars 27 forks source link

Orientational Distribution for Magnetic Models (Trac #713) #113

Open sasview-bot opened 5 years ago

sasview-bot commented 5 years ago

Magnetic models need the angular orientational distribution of magnetic moments around the magnetic field axis.

Migrated from http://trac.sasview.org/ticket/713

{
    "status": "new",
    "changetime": "2019-03-02T01:04:39",
    "_ts": "2019-03-02 01:04:39.736112+00:00",
    "description": "Magnetic models need the angular orientational distribution of magnetic moments around the magnetic field axis.",
    "reporter": "dirk",
    "cc": "",
    "resolution": "",
    "workpackage": "SasModels Infrastructure",
    "time": "2016-10-06T14:00:53",
    "component": "sasmodels",
    "summary": "Orientational Distribution for Magnetic Models",
    "priority": "major",
    "keywords": "",
    "milestone": "sasmodels Next Release +1",
    "owner": "dirk",
    "type": "enhancement"
}
pkienzle commented 5 years ago

Trac update at 2016/10/06 17:39:04: pkienzle commented:

Describe in more detail.

Do you want all orientations within pd degrees of the magentic orientation? Do you have a hard axis and soft axis in a direction independent with more pd in the soft direction than the hard? Do you want this independently for each sld component? Presumably not for the applied field, unless you have problems with field uniformity? Or is it something more complicated?

In any case, this will require some effort since some of the magnetic properties are precomputed outside of the kernel in sasmodels.details.convert_magnetism. This code could be moved into the kernels (increasing code size and runtime cost), or one could move the magnetic orientation polydispersity calculation outside of the kernel, and adjust sasmodels.details, sasmodels.sasview_model.SasviewModel.calculate_Iq and sasmodels.direct_model.call_kernel to suit.

pkienzle commented 5 years ago

Trac update at 2018/04/06 19:25:07: pkienzle commented:

Tying magnetism to particle orientation as suggested in SasView/sasview#993 may be sufficient. It has the advantage of fewer parameters and all magnetic orientations are correlated. It has the disadvantage that all magnetic orientations are correlated

butlerpd commented 5 years ago

Trac update at 2019/03/02 01:04:39: butler changed workpackage from "SasModels Redesign" to "SasModels Infrastructure"

dehoni commented 5 years ago

The commit has a documentation defining magnetisation vector and polarisation in line with particle orientation notation (code has to be changed). Now all what is needed is to figure out how to implement an isotropic ensemble of particle magnetisation ( similar for particle orientations). This is needed to generate two populations with spin pointing in opposite directions in order to describe field-dependence correctly, i.e. the different variation of mean magnetisation vs square mean quantities. With proper generalised orientation distribution, you cover all "normal" use cases.