Maryam, your problem lies in insufficient data. The counts are low, so you have two options: 1 open up the slit size to allow a greater X-ray flux, this will increase the counts and the data quality, but at the expense of some peak resolution. 2 increase the counting time in order to get greater signal/noise, the cost here is time. If you want good data, then spend a long time gathering it. Or, a combination of 1 and 2 - experiment! What are your current data gathering conditions: step size, time per step, detector type, slit sizes, generator settings and tube power rating (I have seen some people under powering the tube and getting poor results).

To your question, grinding will depend on the type of material, most samples will benefit from being 1-10microns, but over grinding can induce amorphicity and reduce your crystallinity.

'Noise' in XRD pattern is relative entity. It appear large or small in comparison to peak data of sample. Ideally, one should not use 'programs' to 'reduce noise'. Noise will be 'constant background' for given set of parameters or settings in X-Ray diffraction experiment. What researcher can do is try to set parameters or settings so as to maximize 'peak data' collection. This can be done by various strategies and it depends on type of material, quantity and nature of its properties such as particle size, composition etc. In your case, you are using 10 gram ceramic compound. Even if it is in the form of either powder or pellet this quantity is more than enough for XRD study. (I don't know what do you mean by 'second treatment'). If it is in powder form and if its particle size range in nanometer then usually 'peak heights' will be small and peaks will be broader than those observed in XRD scan of micron sized powder. You can get better quality data by doing longer scan by reduced step size and more 'time per step'. Take necessary precautions while preparation of sample and better use back-loading technique and 'spinner' set-up. Longer is scan time more is the quality of data collection. Make sure you are taking wider scan range and not missing out on important peaks at lower angles such as between 10 degree to 20 degree. In case of nanopowder special care must be taken to avoid moisture and make sure your powder is dry.

In case you are using sintered or heat-treated pellet of ceramic compound, peak heights will depend on grain-size. Similar to nanopowder if grain size is in sub-micron range one has to use longer scan time.

More specific advice can be given if composition and properties of ceramic compound is known. Hope this helps.

Peak height of XRD pattern of any sample depends on: a) Sample constituent (Higher the atomic no. of the elements present in the sample, higher will be the intensity), b) Unit cell parameters of the sample (For cubic one gets high intense peaks because of multiplicity whereas lowers symmetric structure has lower multiplicity and hence low intense peaks) c) Crystallinity of the sample (same sample treated at two temperatures- generally higher temperature treated samples give good data), d) Scan duration (longer the duration, higher will be the signal/noise ratio), e) Nature of the sample (nano ceramics give broader peaks). So please try to play with either the heat treatment or longer duration scan and you will be able to get better diffraction data. Hope that helps.

Chandrashekhar Sadashiv Kumbhar's answer contains some useful information, follow that and you should be able to produce better results. It might help if you show us your data then it may be possible to diagnose a particular problem.

Mohsin Jafar - some of your points are not correct. Your point a) is incorrect, there is no direct relationship between high atomic number of the elements present and high peak intensity (where did you get this?). Secondly your point b) is incorrect, the unit cell parameters only affect the peak positions, not the peak intensity. However, your other points are well made.

I have got some general observations that:

Gd2Ti2O7 and Gd2Zr2O7 both crystallizes as same pyrochlore structure. Taking same amount of sample I get more intense peaks of the latter.

It can be explained like that: for two similar structures, structure factor equation has to be same. Structure factor equation takes care of the summation of all the from factors and their respective positions.But form factor is related to atomic no. So, because of high Z, form factor will be greater which in turn will lead to a higher value of the structure factor thereby leading to Higher intensity. I correlated it like this. If this logic has any wrong explanation, kindly correct me.

As regards point b. it is my mistake. It should be "crystal structure type". In the parentheses I explained with cubic and monoclinic systems in terms of their multiplicities.

I apologize to all the readers for this unwanted error.

It is necessary that you do a second annealing (calcination) at a temperature higher than that of the first time. The phase did not formed

 Thank you so much for your respons ! it was a misstake in the parameter that my professor notice. Hopefully it will work, it gave better peaks for other samples. Again thank you ! 

In my experience, the main reason for the strong background, along with poor crystallinity of the sample, is the wrong choice of X-ray tube anode. If you have elements in the sample to one or two sites lighter than the material of the anode tube, they give a strong fluorescent background. For example, it is undesirable to use a tube with a cobalt anode for investigating samples containing iron. If anode is copper, the large background obtained in the presence of nickel in samples.

Alexander what I remember its copper that we use in the lab, and I don't have nickel in my samples but its definitively good advice to learn.

 XRD diffraction did not show that good results and good peaks! So Im wondering how can I improve the crystallinity in my powder ? can I save the results and get better XRD patterns if I for example grind it more ? or increase temperatur or time ? I played with parameter and it worked for my other sample but not for the scandium one. 

Thank you.! 

Maryam, your problem lies in insufficient data. The counts are low, so you have two options: 1 open up the slit size to allow a greater X-ray flux, this will increase the counts and the data quality, but at the expense of some peak resolution. 2 increase the counting time in order to get greater signal/noise, the cost here is time. If you want good data, then spend a long time gathering it. Or, a combination of 1 and 2 - experiment! What are your current data gathering conditions: step size, time per step, detector type, slit sizes, generator settings and tube power rating (I have seen some people under powering the tube and getting poor results).

To your question, grinding will depend on the type of material, most samples will benefit from being 1-10microns, but over grinding can induce amorphicity and reduce your crystallinity.

Looking at XRD scan name, I can guess that you have already heat treated the ceramic powder at 1000 degree C for 12 hours and 1200 degree C for 24 hours(This you have to confirm with me).

If that is correct, then following advice will help you: 

If somebody heat treat ceramic powder (like oxides) for that long period and at those temperatures, there is no apparent reason not to have crystallinity. In fact your scan suggest there is very little amorphous content. As far as size of crystallites is concern, that too seems to be in range of microns. (That is a fact in most cases, as I assume temperature and soaking time is pretty high).

This kind of sample should easily produce counts in order of thousands. Still you are getting maximum peak height of only 500 counts; means it seems your 'scan duration' runs  very short time like few minutes if not in seconds.

First make sure you are using sufficient amount of powder in sample preparation.

Then, if you are using point detector, replace it with line detector. (Or sometimes called as Fast detector)

Keep 'scan duration' parameter for sufficiently longer time.

You will find, as you increase scan time, there is proportional increase in peak heights. Soon you will realize what scan duration is required to get peak heights in 5000 to 10000 counts. This much 'maximum peak height' in you scan means you have good quality data for all basic purposes of XRD analysis. Phase analysis of this scan then can give you information about number of phases and their compositions. Take help of your XRD technicians and your pears/supervisors to better set suitable combination of 'step size', 'time per step' and 'scan duration' while doing XRD scan.

I don't think you need to grind the sample or make more heat treatment cycles to get good quality XRD data.

However, if you want to do those heating cycles because you have realized from XRD scan that, you have not obtained desired phase composition in previous heating cycles, then that is a separate issue.

As far as getting good quality data in XRD is concerned, it seems you need to make changes in your XRD scan parameters and settings only.

Three main reasons:

1) Not well-crystallized sample

2) Not adjusting Z properly

3) Not adjusting apertures and mask or discriminator of detector well (especially for magnetic materials) 

Hi if you need to improve the pattern without changing the characteristics of the sample, it is very simple:

1. There is the sample holder called zero background which helps to minimize the amorphous sample.

2. Increase the account numbers during the measurement, decreasing the background in the pattern. This is possible increasing time by step of the pattern.

3. Grinding the sample further to ensure a finer grain.

I hope I've helped

Background in the XRD pattern is coming from the sample only and I think it can not be suppressed by the zero background sample holder (e.g. zero diffraction plate made of Silicon cut at special orientation, etc.).

Zero diffraction holder is preferable when the sample quantity is less and sample holder cannot be filled completely.  Metallic sample holder gives its own background and relative sharp peaks.

In your case, you have sufficient amount of sample quantity.

The Background is always present in the XRD but its amount varies. In highly crystalline samples it can not be seen clearly due to the presence of highly intense Bragg peaks.

When materials structure moves from crystalline to amorphous, sharp peaks goes down and the background looks easier.

There are more than one parameters which affect the XRD background (base line). For example, how you feel the sample in the holder (Packing Density)  

This is my view.

Thanks

Ian J Slipper, Majid Abrishami and Heiddy Quiroz highlighted some important precautions which XRD operator normally take but which you could have missed out. I agree with them and these mistakes are frankly not expected and did not came to mind earlier.

1) Generator settings and tube power settings (under powering the tube leading to not enough X-Ray source photons)

2) Not adjusting 'Z' axis setting in XRD leading to errors in peak position and also number of counts.

3) precautions for magnetic materials.

4) Using metal holder for powder should be avoided if amount of powder is minuscule. One of the suggestions also mention use of 'zero background holder' for this case.

However above mistakes are not expected from trained XRD operator and most researchers are aware of these. Please double-check whether inadvertently you have done some of these mistakes.

I agree with Ian J Slipper. Diffraction pattern looks like insufficient intensity set at each point. This is indicated not the value of the background, but his strong fluctuations.

I suggest a more fundamental step, index the pattern.

Have you used Reitveld analysis yet? 

If your samples are reasonably consistent then you may compare profiles to get a relative idea. Use an internal/external standard if possible to estimate better. See powder XRD example below with "no slits" (source & receiving)

https://www.researchgate.net/post/How_do_I_calibrate_the_diffractometer_2Theta_angles_with_this_powder_diffractogram_of_corundum_standard_from_a_Bruker_D8_XRD_instrument

Increase the  time of measurement to get better pattern

https://www.linkedin.com/pulse/incident-beam-characterization-sample-alignment-ravi-ananth?trk=mp-reader-card

https://www.researchgate.net/post/Is_this_a_reasonable_way_to_eliminate_effects_of_sample_surface_misalignment_with_the_diffractometer_axis_Theta-2Theta_Diffractograms

Ravi, yes I agree - careless words. For 'noise' read 'background' and 'flux' read 'intensity'.

Ian! No problem. I was just clarifying for my own gratification.

I always appreciate you sharing, without reservation, your incredible knowledge and experience in XRD for the benefit of the entire XRD community on RG.

Warm regards!

Thank you Ravi for your kind words. After 28 years in the business I am still learning, there is so much more to know in the field of XRD. Who knows, one day I might get around to a 2-D detector ;-) But as you may have seen if you read my posts following your link, my principal field is in micropalaeontology, a long way from X-rays. I have a problem in that my time is now limited - I have terminal cancer and may not be around for very much longer.  I am very busy finishing a manuscript for what is effectively my scientific life's work. I doubt very many will read it but I would like to see it published.  I will keep going as long as I can. There is a lot of accumulated knowledge in the participants on these pages and I have benefitted greatly from their input. But I will keep posting here, trying to spread what limited knowledge I have around to those that can benefit.

"I try living life as if each day was my last day", Anonymous Maximus :-) Share more of your love randomly! My mom was victorious in her struggle with cologne cancer but is now challenged with Parkinson. Do look into the use of Turmeric (with little bit of black pepper) and Ginger powders for ameliorating side-effects of the medication used for your ailment as well as general health. I've been using 3-5 grams of dry powder per day for the past 3 years with incredibly awesome "side-effects". My challenges are elevated A1C1, cholesterol, hypertension, rheumatoid arthritis. I managed somehow without the prescribed medications for  these conditions. Read about it and try. The concoction I use includes 2-3 tsp turmeric, 1-2 tsp ginger, a sprinkle of black pepper, 1/4 lime/lemon, and honey or sugar for taste (which I do not use) plus boiling water. Just drinking a beer mug full of it 6:30AM after tea when cold enough to drink :-)

We all appreciate your spirit. Let us know when this publication is ready. I'd love to take advantage!

BTW if you are interested in playing with our 2D devise, we could arrange it. Let us discuss via email. I may be able to organize a cost-free evaluation period for educational institutions quite rapidly. As long as you do not live in a country where US sanctions may be applicable. We would need some paper-work to initiate this and confirm the return after the evaluation period. We may need some sort of NDA (non-disclosure except for research data generated) to protect both our interests.

I think it would be feasible to determine the conditions when rocks formed by using real time 2D Bragg XRD Microscopy. A bold statement, yes? There is a whole lot that I know NOT than I do. That is exciting even for a sexagenarian!

Ravi

732.429.3538

https://www.linkedin.com/pulse/bragg-xrd-microscopy-quantifying-deviations-from-ideal-ravi-ananth?trk=mp-reader-card

https://www.linkedin.com/pulse/scroll-through-3d-reciprocal-space-using-pixcel-area-detector-ananth?trk=mp-reader-card

http://www.webmd.com/vitamins-supplements/ingredientmono-662-turmeric.aspx?activeingredientid=662

http://www.webmd.com/vitamins-supplements/ingredientmono-961-ginger.aspx?activeingredientid=961

Back to topic - The Roast in the interest of learning!

"Background in the XRD pattern is coming from the sample only", that is yet to be established. The use of "sample holder called zero background" will lead us closer to your assertion.

"There is the sample holder called zero background which helps to minimize the amorphous sample". Not so! If there is amorphous constituent in the "sample", its XRD component signal will remain and need not be eliminated.

Hi Maryam,

I'be been experienced a lot of solutions for more than 18 years, and the best I found is simply to change the holder.

By choosing a zero bakcground holder, you remove noise of the background. More information can be read in one of my publications (2007 Influence of Structural Properties of Pristine Carbon Blacks). The zero background holder method ( ZBH)  combines the advantages of the external standard method with the accuracy of the internal-standard method. A zero background holder is a single crystal

which has been cut along a nondiffracting crystallographic direction and then polished to optical flatness. Any X-rays falling on this crystal will be completely extinguished by Bragg extinction. The use of such a crystal as a sample support has become a popular way to improve the signal to background in a powder pattern.

Stephane, if you examine Maryam's diffractogram, the background only contained 30-50 counts, this is already very low. I don't think that simply using a low background holder in that instance is going to significantly improve the result. I regularly use these holders when I have very limited quantity of material or when I am examining organic samples - to minimise the sample transparency effect. But Maryam's post suggests that 10g of material are available. It should not be necessary to use a low background holder for such a quantity. There is enough to get good particle statistics. One of her posts indicated that there was a mistake in the original diffractometer parameters, and this has now been corrected for other samples.

The problem is, or I see it as a problem, that I've used the zero background sample holder.Ive worked with similar material and managed to get good results but now when my project is increased in mass to 10 gram it wasn't easy. 

10 gram! Synthesized? How? Processing history?

What do you expect the average "grain"/"particle"/"crystallite" size?

Got any micrographs of the powder?