I am looking at determining the strain in Bi4Ti3O12 thin films on Si/Ti/Pt substrates using novel doping. I deposited the thin film using a d RF magnetron sputtering, and producing a non-doped target for a reference thin film would take to long.
What is the best way to quantify strain from the peak shift for my thin film? would it be acceptable to use compare it to the power diffraction pattern as calculated using the Mercury software. Or should I compare it to a thin film using platinumised substrates reported elsewhere?
Best Wishes
David
Dear David,
You can try williamson hall analysis for strain using XRD data.
All the best
Rehan
Dear David,
you can calculate strain from the peak broadening (not from peak shift!). As Rehan mentioned, Williamson Hall plots are a convenient way to do so.
You firstly need to determine the instrumental peak broadening from your diffractometer. If it is difficult for you to prepare a strain free sample of similar composition, you can use another material. For example Si or LaB6 is often used as reference. The only thing which is important, is that your reference sample is free of microstrain and is not nano-crystalline. The broadening you measure with your reference sample then will only be due to instrumental effects.
This broadening depends of the diffraction angel so you have to interpolate the values of instrumental broadening at the angles of interest
After correcting the total peak broadening from instrumental effects, you can use the Williamson Hall (WH) method to separate peak broadening due to strain from peak broadening due to size effects.
You can find plenty of information about the WH method in the internet, for example see attached pdf.
Best, Clemens
http://cbc-wb01x.chemistry.ohio-state.edu/~woodward/size_str.pdf
Dear David
Strain in thin films is inevitable and occurs mainly because of differential thermal coefficient and lattice mismatch between the film and substrate, especially in epitaxial thin films. The effect of strain (compressive as well as tensile) lead to change in the lattice parameters of unit cell which becomes evident from the shift in the position of Bragg's peak in XRD. The shift in the position will give you the change in the value of d (inter planner spacing), which in turn is related to the lattice parameters of the unit cell. Hope that may have answered your Q.
Regards
suhail
You''re getting some answers that say stain will cause peak shift and some that say no, it will not shift the peak, it will broaden it.
(Clemens Wall's answer is correct if microstrain only is considered)
The key to reconciling the two sets of answers is distinguishing between microstrain and macrostrain, that is the length scale over which the differences from equilibrium stress-free uniformly pure material lattice parameters occur. Stresses and strains that occur over long distances (many crystallites) within a material are referred to as macro-stresses. Stresses or strains that exist only locally (either between grains or inside a grain) are called micro-stresses or microstrains. Temperature change in a multiphase mixture where the phases are mixed will result in microstrain.
Microstrains cause peak broadening. Microstrains are caused by nonuniformities withing the crystallite, such as nonuniform composition (on the size scale of the crystallites), dislocations and other crystal imperfections such as grain boundaries.
Macrostrains cause peak shifts due to a possibly anisotropic change in lattice parameter. Macrostrains are caused by things like: in a single phase material, a change in temperature from reference temperature or change in composition from reference composition will result in uniform hydrostatic strain. In a multiple phase sample where the phases are at different locations (like a film on a substrate), a change in temperature with a coefficient of thermal expansion mismatch between the phases will result in biaxial strain. Physical deformation will cause a direction-dependent macrostrain (in addition to microstrain caused by dislocations created by the deformation, if it extends beyond elastic to plastic deformation). If the thermal history includes melting, such as welding, there are likely to be very large macrostrains.
Hopefully this helps to reconcile the disparate answers you've received, so far. Feel free to ask further questions.
Dear David,
I agree with Dr Rehan You can use williamson hall analysis for strain using XRD data. Also, you can cheek
https://www.researchgate.net/publication/288073811_Structural_linear_and_nonlinear_optical_properties_of_co-doped_ZnO_thin_films
https://www.researchgate.net/publication/282155487_Influence_of_annealing_temperature_on_structural_and_optical_properties_of_nanocrystalline_Platinum_octaethylporphyrin_PtOEP_thin_films
Article Structural, linear and nonlinear optical properties of co-do...
Article Influence of annealing temperature on structural and optical...
I totally agree with dear Boyer, but I want to mention here, if we look at the question of dear David, it is all about the thin films on different substrates and in thin films strain is visible from the shift in diffraction peaks corresponding to different reflections.
Dear All,
thank you very much for you input, I will be sure to give these a read and to try and understand the williamson hall method.
best Wishes
Jamie
- Badges
- Science method
- Similar topics
- Methodology
- Crystallography
- X-ray Diffraction