shadow3

Dear Mianzhen,

Mica cannot work with the preprocessor, even though is cubic, but it has more than two elements in the unit cell.

The only way, by now, is to use a macro that applies an external diffraction profile, that you should provide (experimental or calculated by xop, for instance).

The method, as you said, is limited to a single wavelength. Usually a small change of energy in the bandwidth accepted by the crystal, does not produce significant changes in the diffraction profile. So in good approximation, you can stay with a single diffraction profile ("monochromatic") but use a uniform spectral distribution (with a relatively small bandwidth).

I attach an example workspace that contains a macro that applies an external diffraction profile. I think you can use it as template for your case. I place this file in the "Example workspace" for further users.

Let me know if you encounter problems.

Best regards,

Manuel

On 10/26/2012 03:26 AM, Mianzhen Mo wrote:

Hi Manuel,

Sorry to bother you again.

I have a technical problem with running the Shadow program. I would like to simulate a X ray imaging system with Mica Muscovite crystal (002), which has a 2d spacing of 19.98 A. In the Bragg Preprocessor, I select the simple FCC as the structure type, however, the problem is the entry for "Symbol 1st element", what should I enter here? I can't enter single element since, as you must know, Muscovite is a compound with 5 kinds of atoms (the chemical formula is: K2O 3Al2O3 6SiO2 2H2O ).

One way I tried to get around this problem, but seemed to fail, is:
a) choose simple FCC
a) set the lattice constant as the right constant with Miller index of 004 for Mica (Here since Mica working on second diffraction order for 1.5 keV, the lattice constant is 5.0037 A)
b) set element as Si, (here I just picked up an element randomly)

But the problem with this trial is f0, f1&f2, these should be different from Si and Mica, so this should not represent correctly the behavior for Mica.

The other way I am thinking is whether I can generate the reflectivity curve using xop instead of using Bragg Preprocessor with shadow? but this method is limited to a single wavelength, in my simulation, i will set a broadband x ray source.

Anyway, I don't think I can get around this problem without your help, Could you please give me some hints and suggestions?

Thanks,
Mianzhen

On Thu, Oct 4, 2012 at 1:00 AM, Manuel Sanchez del Rio <srio@esrf.fr> wrote:

Hi Mianzhen,

For hexagonal crystals, the four index notation (h k i l) verifies h+k+i=0, so i=-h-k is redundant. Thus, for XOP forget the third index.

By the way, we are currently updated the XOP bent crystal codes including the crystal anisotropy, and full cylindrical/spherical shape, in principle for cubic crystals. We discovered some small problems in the current codes for multilamellar-Laue, and in the treatment of the Poisson ratio. If you use Bragg geometry I guess the dependency on anisotropy is small, so you are "safe". We will have the new code in about two months.

I enclose one of my papers on Quartz reflectivity, just in you may be interested....

Best regards,

Manuel

On 10/04/2012 07:17 AM, Mianzhen Mo wrote:

Dear Manuel,

My name is Mianzhen Mo, and I am a PhD student in University of Alberta, Canada.

Right now I am using xop to simulate the reflectivity of a spherically-bent quartz crystal. However, I don't quite understand the Miller index settings in the input panel for bent crystal.
The quartz that I wanna simulate is with a hexagonal lattice system, the lattice plane is 21(-3)1, and the 2d spacing is 3.082 angstrom. Here comes my question, the miller indices in the xop input panel only have options for (h k l), which seems to me only for a normal cubic lattice system. Is there any way that I can type in a four-index lattice plane such as the one that I have (21(-3)1)? or Is there any way I can convert a four-index lattice plane to the acceptable three-index lattice plane as shown in the panel and still simulate the right plane with right lattice spacing that I want?

Looking forward for your help. Many thanks!!

Cheers,
--
Mianzhen Mo
Laser Plasma
Electrical and Computer Engineering
ECERF W3-030
University of Alberta
Edmonton, Alberta, Canada, T6G 2V4