All Answers (29)
 Why people keep talking about "DebyeScherrer equation"? The equation is called simply Scherrer equation. Debye has nothing to do with that.
My suggestion is always: if you want to drive a car on a public road, first learn to drive.
So please, before using ANY tool, first open a book or browse through the Internet to find out what is all about and to find out the relevant literature IN the field (too many people look for literature on Scherrer equation in papers that have nothing to do with Xray diffraction). You can find the derivation of Scherrer equation in any Xray (powder) diffraction book and if you look at that derivation, you find the answer to your question as well (any student know that angles, especially in trig formulas are in radians). Plus β and FWHM are two different things (the integral breadth versus the full width at half maximum) that may or may not be the same ofr a given peak.
This said, Scherrer equation give you a number related to the distribution of length of columns composing your specimen. It CAN be related, under quite strict hypotheses, to the mean domain size. In general, knowing the domain shape (that you can hardly guess from Scherrer formula), the value you obtain is a ratio between high order moments (4th over 3rd) of the size distribution. For some discussion:
https://www.researchgate.net/publication/200045699_Line_broadening_analysis_using_integral_breadth_methods_a_critical_review
(Scherrer formula is the size part of the WilliamsonHall method).
So forget about "particle size". And forget about being able to characterize anything above 100 nm using Scherrer equation (the equation does not include instrumental and strain broadening effects always present in a pattern). Moreover, talking about "crystallinity" is also not appropriate as you characterize the crystalline part of your material only.
If you are still convinced that Scherrer formula is of any quantitative use in 2013 (it is good for qualitative comparison), then:
* D and λ have the same unit of measurement (e.g. nm as per SI)
* FWHM is the full width at half maximum of the peak (not half of it) in rad. As you have it in deg from the machine, just multiply by pi/180
* theta is half of Bragg angle in rad
If you want to work quantitatively then first open a good book on diffraction and then have a look at more modern analysis techniques (e.g. Whole Powder Pattern Modelling, https://www.researchgate.net/publication/11527348_Whole_powder_pattern_modelling) that can give quantitative information compatible with the microstructure of the material.  Perfect response @ Matteo. Is very usual find papers where 'particle size calculated by Scherrer formulae' is written... Very nice answer
 best answer Matteo Leoni
 Thanks a lot Matteo i am really grateful for your contribution. As Pablo said, one usually finds papers where the Scherrer fomula is used. There was also a controversy as to whether one should halve the FWHM..thanks a lot.
 Also one need to take the machine broadening and stress and strain factors into account as well when using the Scherrer equation. It is also good to compare the calculated particle size with something like TEM and other techniques of determining the particle size. Further more, some techniques apply to the primary particle size and others to the secondary particle size.
 Ali, no problem. In any case I keep insisting that XRD does not measure a particle size, but a domain size. Particles can be single domains, so the two numbers can be equal. Scherrer equation does not include any (micro)strain effect. It might include the instrumental part, but not many people do that correction (if you use it for comparison, you don't care if your values are all underestimated).
I would like to see a paper where people want to halve the FWHM... that's scary  I did see some literature mentioning HWHM instead FWHM. I think HWHM stands for "Half width at half maximum". If that is the case Ali is referring to, I am really curious to know about HWHM. I personally never used it anywhere.
I did see so many people reporting crystallite size in nano for an ultrafine grained powders. This just happens by not subtracting the instrument broadening from FWHM in scherrer equation. People do not bother about the correction for instrument while using Scherrer equation just because the word nano sounds crazy.  Radians.
 I had a quick check on google and indeed there are people using the HWHM (= FWHM/2).in Scherrer formula. That's complete nonsense. It could be that those authors had a wrong translation for "Halbwertsbreite"...
Sitarama, I think most people do not care about the instrument correction simply because they found that someone else was using Scherrer formula that way and they decided to do the same. I cannot count the number of wrong citations to Scherrer (1918) and I am pretty sure most people out there never saw or read it (they just keep citing something they found in other publications). Believe or not, in that article there is no mention on how to derive the formula!  Thanks you very much for yours articles Prof. M. Leoni.
 I think the formula makes sense to use in a few special cases, with due caution. For instance the 'particles" should be small, so that this dominates the peak broadening compared to strain. It should be realized that it is not immediately possible to distinguish crystal size and correlation length. For instance the broad peaks of a fluid could be translated as " very small crystals" which obviously makes no sense. The XRD basically tells you about lack of correlation, either because you have small crystals, or you have substantial uncertainty in the regularity of the packing, be it stress induced or via multiple conformations (in molecular systems), or due to quenching  take your pick . Also, worth stressing that particle size is not necessarily the same as crystal size. Having said the Scherrer equation is quite useful for comparative purposes (comparing like with like).
In some cases you may know apriori what sort of defects are possible & then the interpretation becomes considerably easier.  In addition to the considerations already described, I would like to speak up for the oftenignored importance of K the Scherrer constant. In most papers published that use the Scherrer formula, the authors assume K=1. This has led to a number of publications that do not include K in the Scherrer formula, which has led to more and more people ignoring the importance of this constant.
For an excellent paper describing the Scherrer formula and the Scherrer constant, I recommed J.I. Langford and A.J.C. Wilson, “Scherrer after Sixty Years: A Survey and Some New Results in the Determination of Crystallite Size,” J. Appl. Cryst. 11 (1978) pp 102113.  Can someone explain what causes " XRD peak splitting"
 Scott, you are right, very few people consider the K (actually there are also various definitions and values according to the way Scherrer formula is derived).
The meaning of the result does not change if you use 1 or 0.89. The result you obtain is in both cases not what you think it is! Moreover most people do not even consider the fact that that peak is often made up of two wavelength components (separated, thus giving broadening), that have an intrinsic Lorentzian breadth. You should also consider those before dealing with K....
Ali, what do you mean? usually peaks split when you have a reduction in symmetry of the lattice (e.g. the 001 010 and 100 are superimposed i.e. make one peak in cubic and can be 2 or 3 if you have tetragonal or orthorhombic distortion)  @ Matteo Leoni, Dear sir can you please explain that  in some cases the whole pattern is shifted a bit from its JCPDS data towards higher values of 2theta.
 Anil, Is the shift uniform or does it have any polynomial dependencr across the 2theta range. That can be due to zero error and/or sample displacement error. If the shift exists even afte correcting for both, your sample might have a slightly different lattice parameters, for example bcz of strain due to different processing conditions. A full pattern fitting will help you to refine ur lattice. Also cross check the standard silicon shifts with your diffraction pattern. This will help calibrate your sample pattern.
 Dear S. KadaThe shift is uniform throughout the range.
 If the shift is uniform over the entire range of 2theta, that would suggest a 2theta zero error. But f you only see this "in some cases" then that is very strange, unless your "entire range" is too limited  how high do you go in 2theta? A sample displacement error (very common) leads to a shift with 2theta dependence  more significant at low angles than at high angles.
As Matteo and other have noted, the description "JCPDS" (Joint Committee on Powder Diffraction Standards of the ASTM) has long been superseded by "ICDD" (the International Center for Diffraction Data).  Anil! I've used internal standards to eliminate variations due to both "zero error and/or sample displacement error" as Sitarama has pointed out. Here is an example of how we used the invariant diffraction pattern from the sample holder to calibrate multiple conventional (with 0D detector) diffractograms:
http://www.flickr.com/photos/85210325@N04/8434970459/in/photostream
Aluminum foil and scotch tape are reasonably good calibration standards:
1. Kevlar+Al foil: http://www.flickr.com/photos/85210325@N04/7975835813/in/set72157632728981912
2. Energetic composite + Al Foil:
http://www.flickr.com/photos/85210325@N04/7988760159/in/set72157632728981912/
Thanks Scott Speakman! I use your PP presentation generously:
a. Basics of XRD: http://web.pdx.edu/~pmoeck/phy381/Topic5aXRD.pdf
b. Often used slide: http://www.flickr.com/photos/85210325@N04/8592218974/in/photostream
Matteo Leoni! Thanks for your excellent comments. I've learnt a lot from this discussion. Thanks to all contributors!
"Plus β and FWHM are two different things (the integral breadth versus the full width at half maximum) that may or may not be the same ofr a given peak." Helpful link:
http://pd.chem.ucl.ac.uk/pdnn/peaks/gauss.htm
Ali Wako! "Peak Splitting" in monocrystals occur due to subgrain structure:
http://www.flickr.com/photos/85210325@N04/8294404556/sizes/h/in/photostream/
Bragg XRD Microscopy  YouTube play list for many materials with various nanostructural morphologies:
http://www.youtube.com/watch?v=ykSjd2ZZn64&list=PL7032E2DAF1F3941F
Please feel free to join us and contribute with your knowledge & expertise. You are all preapproved and welcome to join "Xray Diffraction Imaging for Materials Microstructural QC" group 
http://www.linkedin.com/groupAnswers?viewQuestionAndAnswers=&discussionID=218925637&gid=2683600&commentID=128396954&trk=view_disc&ut=3Xe63G3z1I5RI1
Presently we are discussing: " XRD Methods in "Materials Analyses": How many of you have used or are currently using and/or would like to use in the future? XRD methods are generally portable, real time, noncontact, NDE & in situ."
Ravi
BraggXRDMicroscopy@GMail.com
P.S: I'm still reviewing some of all your comments/links/references and continuing to learn.  Volker Klemm! Please forward PDF files for the references you've provided, when convenient.Thanks!
Ravi
BraggXRDMicroscopy@GMail.com
http://www.linkedin.com/profile/view?id=63121884&trk=hb_tab_pro_top  I agree with Edward Andrew Payzant!
I've been still living in the past and using JCPDS. I shall henceforth switch to ICDD.  Ali Wako! More on Bragg Peak Splitting.
Quartz Bragg XRD Micrograph: http://www.flickr.com/photos/85210325@N04/8449083122/in/set72157632724063041/
In this case the incident beam contained both K Alpha 1 & 2. Only a Ni foil K Beta filter was used. Each spatial pixel yields a RCP (Rocking Curve Profile) with the well separated K Alpha 1 & 2 peaks. We used this "INVARIANT" separation to recalibrate each RCP along the Omega scale. Very precise!  Mateo! "Moreover most people do not even consider the fact that that peak is often made up of two wavelength components (separated, thus giving broadening), that have an intrinsic Lorentzian breadth."
In most Xray rocking curve analyses the incident beam is absent the K Alpha 2 component and a very low inherent FWHM (instrumental). However, in the case of Quartz above, we actually turned "lemon to lemonade" by using the separation between the two topographs (K Alpha 1 & 2) to recalibrate the Omega scale.
Obviously, the presence of two superimposed topographs may appear to be confusing but when viewed in the Omega space the individual topographs are easily deconvoluted. The inherent FWHM is at least an order of magnitude higher in this case compared with the monochromated K Alpha 1 only case. But since the Quartz was of such high quality the peaks were amply well separated for deconvolution. Here the RCP profile shape was more akin to Pearson VII m=5 (just tested it): http://pd.chem.ucl.ac.uk/pdnn/peaks/pvii.htm
Here are some examples of other RCP shapes that we encountered:
http://www.flickr.com/photos/85210325@N04/8561945111/sizes/k/in/photostream/
Ravi
BraggXRDMicroscopy@GMail.com  Thanks Ravi Ananth and Edward Payzant
 DebyeScherrer is an optical configuration and not an equation to calculate particle size! I totally agree with M Leoni, you really have to read before doing anything. You can obtain accurate measurements of crystallite size (not particle size! A particle sizer can give you particle size. XRD may be able to give you crystallite size. They are not the same thing.) but be very careful. You have to use a standard to obtain broadening due to the instrument. The equation is:
Bo = Bi+Bst+Bs
where Bo = total peak broadeing (=observed peak broadening), Bi=broadening due to instrument, Bst=broadening due to lattice strain, Bs=broadening due to crystallite size. So many people just use the FWHM directly and plugged the value into the Scherrer equation! This gives inaccurate results which undestimate the actual value.  That is correct N. Kamarulzaman! The "particle size effect" and the "strain effect" must be deconvoluted besides the instrumental factors for correct results. It is tough to do this with the DebyeScherrer technique using film. Multiple (hkl)'s need to be analyzed to potentially separate the effects of the two factors on the FWHM. Lots of computation! The particle size effect should be independent of the (hkl) chosen. The strain effect would certainly be (hkl) dependent.
 Thank you all your suggestions are valuable.

I think, everybody doing very well in xrd tunnel
 Views 18982
 Followers 68
 Answers 29
Popular Answers
My suggestion is always: if you want to drive a car on a public road, first learn to drive.
So please, before using ANY tool, first open a book or browse through the Internet to find out what is all about and to find out the relevant literature IN the field (too many people look for literature on Scherrer equation in papers that have nothing to do with Xray diffraction). You can find the derivation of Scherrer equation in any Xray (powder) diffraction book and if you look at that derivation, you find the answer to your question as well (any student know that angles, especially in trig formulas are in radians). Plus β and FWHM are two different things (the integral breadth versus the full width at half maximum) that may or may not be the same ofr a given peak.
This said, Scherrer equation give you a number related to the distribution of length of columns composing your specimen. It CAN be related, under quite strict hypotheses, to the mean domain size. In general, knowing the domain shape (that you can hardly guess from Scherrer formula), the value you obtain is a ratio between high order moments (4th over 3rd) of the size distribution. For some discussion:
https://www.researchgate.net/publication/200045699_Line_broadening_analysis_using_integral_breadth_methods_a_critical_review
(Scherrer formula is the size part of the WilliamsonHall method).
So forget about "particle size". And forget about being able to characterize anything above 100 nm using Scherrer equation (the equation does not include instrumental and strain broadening effects always present in a pattern). Moreover, talking about "crystallinity" is also not appropriate as you characterize the crystalline part of your material only.
If you are still convinced that Scherrer formula is of any quantitative use in 2013 (it is good for qualitative comparison), then:
* D and λ have the same unit of measurement (e.g. nm as per SI)
* FWHM is the full width at half maximum of the peak (not half of it) in rad. As you have it in deg from the machine, just multiply by pi/180
* theta is half of Bragg angle in rad
If you want to work quantitatively then first open a good book on diffraction and then have a look at more modern analysis techniques (e.g. Whole Powder Pattern Modelling, https://www.researchgate.net/publication/11527348_Whole_powder_pattern_modelling) that can give quantitative information compatible with the microstructure of the material.
Article: Line broadening analysis using integral breadth methods: a critical review
ABSTRACT: Integral breadth methods for line profile analysis are reviewed, including modifications of the WilliamsonHall method recently proposed for the specific case of dislocation strain broadening. Two cases of study, supported by the results of a TEM investigation, are considered in detail: nanocrystalline ceria crystallized from amorphous precursors and highly deformed nickel powder produced by extensive ball milling. A further application concerns a series of FeMo powder specimens that were ball milled for increasing time. Traditional and modified WilliamsonHall methods confirm their merits for a rapid overview of the line broadening effects and possible understanding of the main causes. However, quantitative results are generally not reliable. Limits in the applicability of integral breadth methods and reliability of the results are discussed in detail.
Article: Whole powder pattern modelling
ABSTRACT: A new approach for the modelling of diffraction patterns without using analytical profile functions is described and tested on ball milled f.c.c. Ni powder samples. The proposed whole powder pattern modelling (WPPM) procedure allows a onestep refinement of microstructure parameters by a direct modelling of the experimental pattern. Lattice parameter and defect content, expressed as dislocation density, outer cutoff radius, contrast factor, twin and deformation fault probabilities), can be refined together with the parameters (mean and variance) of a grainsize distribution. Different models for lattice distortions and domain size and shape can be tested to simulate or model diffraction data for systems as different as plastically deformed metals or finely dispersed crystalline powders. TEM pictures support the conclusions obtained by WPPM and confirm the validity of the proposed procedure.
Sitarama, I think most people do not care about the instrument correction simply because they found that someone else was using Scherrer formula that way and they decided to do the same. I cannot count the number of wrong citations to Scherrer (1918) and I am pretty sure most people out there never saw or read it (they just keep citing something they found in other publications). Believe or not, in that article there is no mention on how to derive the formula!