Difference between revisions of "Diffuse scattering"

From GISAXS
Jump to: navigation, search
(Ornstein-Zernike model)
(Debye-Bueche random two-phase model)
Line 26: Line 26:
  
 
===Debye-Bueche random two-phase model===
 
===Debye-Bueche random two-phase model===
Yields correlation length
+
Yields correlation length (''a''):
 
::<math>
 
::<math>
 
\frac{\mathrm{d}S}{\mathrm{d}W} = \frac{A}{ (1 + (qa)^2 )^2 }
 
\frac{\mathrm{d}S}{\mathrm{d}W} = \frac{A}{ (1 + (qa)^2 )^2 }
Line 32: Line 32:
  
 
* References:
 
* References:
** Debye, Bueche [http://link.aip.org/link/doi/10.1063/1.1698419 Scattering by an Inhomogeneous Solid]
+
** P. Debye, A.M. Bueche "[http://scitation.aip.org/content/aip/journal/jap/20/6/10.1063/1.1698419 Scattering by an Inhomogeneous Solid]" ''J. Appl. Phys.'' 1949, 20, 518. [http://link.aip.org/link/doi/10.1063/1.1698419 doi: 10.1063/1.1698419]
** Debye, Anderson, Brumberger [http://link.aip.org/link/doi/10.1063/1.1722830 Scattering by an Inhomogeneous Solid. II. The Correlation Function and Its Application]
+
** P. Debye, H.R. Anderson, H. Brumberger "[ Scattering by an Inhomogeneous Solid. II. The Correlation Function and Its Application]" ''J. Appl. Phys.'' 1957, 28, 679. [http://link.aip.org/link/doi/10.1063/1.1722830 doi: 10.1063/1.1722830]
 
** [http://www.ncnr.nist.gov/resources/sansmodels/DAB.html NCNR page]
 
** [http://www.ncnr.nist.gov/resources/sansmodels/DAB.html NCNR page]
  

Revision as of 19:56, 3 June 2014

Diffuse scattering is the scattering that arises from any departure of the material structure from that of a perfectly regular lattice. One can think of it as the signal that arises from disordered structures, and it appears in experimental data as scattering spread over a wide q-range (diffuse). Diffuse scattering is generally difficult to quantify, because of the wide variety of effects that contribute to it.

Bragg diffraction occurs when scattering amplitudes add constructively. If there is a defect in a crystal lattice (e.g. atom missing or in a slightly 'wrong' position), then the amplitude of the Bragg peak decreases. This 'lost' scattering intensity is redistributed into diffuse scattering. The diffuse scattering thus arises from the local (short range) configuration of the material (not the long-range structural order).

In the limit of disorder, one entirely lacks a realspace lattice and thus scattering does not generate any Bragg peaks. However, a disordered structure will still give rise to diffuse scattering. The Fourier transform of a disordered structure will not give any well-defined peaks, but will give a distribution of scattering intensity over a wide range of q-values. Thus samples with an inherently disordered structure (polymer blends, randomly packed nanoparticles, etc.) will only generate diffuse scattering.

Causes

  • Thermal motion causes atoms to jitter about their ideal unit cell positions, which decorelates them. This suppresses the intensity of the Bragg peaks, especially the higher-order peaks (see Debye-Waller factor), and instead generates high-q diffuse scattering. (One can also think of this in terms of phonons: in ordered systems the diffuse scattering is probing phonon modes.)
  • Static disorder in crystals (vacancy defects, substitutional defects, stacking faults, etc.) similarly creates diffuse scattering.
  • Grain structure in otherwise ordered materials will also contribute. The grains themselves can count as 'scattering objects', but since their size is ill-defined, the grain boundaries give rise to diffuse scattering.
  • Nanoscale disorder gives rise to low-q diffuse scattering. For instance, a disordered polymer blend (or a bulk heterojunction) or a random packing of nanoparticles, will generate substantial low-q diffuse scattering.
  • Surface roughness in thin films measured by GISAXS gives rise to low-q diffuse scattering in GISAXS. Roughness will tend to broaden (and increase the intensity of) the specular rod, and will also generate intense low-q scattering.
  • Polymer chains in solution generate scattering without a well-defined size-scale. This is normally interpreted in terms of the form factor of the polymer chain. However one can also think of it as the polymer chains having disordered arrangements and thus giving rise to diffuse scattering. (C.f. definitional boundaries.)

Analysis

Diffuse scattering can be difficult to quantify, since so many different effects contribute to it. Nevertheless, if one has a good understanding of the expected kind of disorder, one can fit the diffuse scattering with a model.

Ornstein-Zernike model

Yields correlation length (ξ):

Debye-Bueche random two-phase model

Yields correlation length (a):

Guinier model

Yields average pore size.


See Also