Dear sampada, Use the origin software, this is the best tool for such type of function. You can perform very easily with out any time consuming. This is suitable not only for XRD, as well XPS, XRF. These function as base line correction, normalisation can be efficiently done in origin.

It will depend on your software. It is a trivial matter in most programs. But if you want to do it manually in a spreadsheet then you could try the approach:

1 find the Max Y value in each file

2 divide Max Y high by Max Y low = multiplier

3 multiply each value of Y for Max low series by multiplier.

4 chart

See the example spreadsheet. 2 files raw data and calculations, initial chart, and chart after normalisation.

Thank you for the answer and the spreadsheets.

dear Sampada 

i agree with Ian 

yes it is possible

Thank you everyone!

Dear sampada, Use the origin software, this is the best tool for such type of function. You can perform very easily with out any time consuming. This is suitable not only for XRD, as well XPS, XRF. These function as base line correction, normalisation can be efficiently done in origin.

I think, the correct way would be first to subtract the baseline. The baseline is important since it is not obviuosly the same for each sample, although you repeat the XRD on the same sample without modifiying the parameters (sample position, same voltage and current for X Ray, same detector parameters. etc.). Then find a reference diffraction peak which is common of all your diffractograms. After the reference peak is choosen, then you take its intensity as the maximum intensity and do all the steps as proposed by Ian. You can use Excel or Origin. But I think it is difficult to subtract the baseline without a specific XRD software, if you want to do it correctly. You can use Match software which is free of charge for a short period (1 month I think).

Good luck   

I agree with the answer given by Atilla. The analysis of XRD spectrum starts from the removal of background. It is a crucial when sample filled by the operator is not uniform within the sample holder.

The best free software for the XRD data are:  

http://www.crystalimpact.com/match/

http://www.xpowder.com/

Its not completely free but one can do analysis within the time period given by the software company. 

Atilla is rigth, but if you are not using a zerobackground sample holder you will make some mistakes.

I think it is better to keep the original data, even for comparisson purposes, because it'll give u the information of instrument, data acquisition, sample preparation and nature of ur sample.

yes it can be

Here's how I prefer to perform normalization for reflection mode XRD conventional line profiles:

This, in my opinion, will preserve "the information of instrument, data acquisition, sample preparation and nature of ur sample"! It should also eliminate any differences in "Absolute Intensity" due to "sampling volume" changes between various specimens or the power settings used.

What do you think?

https://www.flickr.com/photos/85210325@N04/8434970459/in/photostream

Interesting, Ravi - step 3 in Excel?

Sure Ian! Add all values in the column :-) 

Ravi, I can't get that to work. The baseline subtract is ok, but divide by integrated intensity for each point does not seem to normalise. 

Please have a look at the sheets - same data. Have I made an error here?

Looks great! What is the difference between File 1 and File 2? What behavior were you expecting after "normalization"?

The contribution from the principal phase constituents in File 1 seems significantly diminished in File 2.

There was a zero value in File 1 so BG was 0.

I'd like to see the profiles using your method as well to try and understand the difference. I've always been more comfortable using the INTEGRATED INTENSITY over absolute intensity in powder diffraction data.

I'm still learning :-)

BTW this is not the Ravi method. It is a method that predates both of us combined potentially. It is quite a well known principle of conserving the total integrated intensity from equivalent diffracting volumes. Intuitively, I'd think that the reliability of this method  will increase with a larger angular range for the powder profile :-)

In my opinion, this is a better method to compare constituent composition of several powder profiles.

Thanks for that, I don't think this is what the original post by Sampada intended.

The two files are just two datasets I chose to use to demonstrate a manual normalisation. They both contain the one component in common. The difference is seen between the charts called 'raw data charts' and 'normalised'. So the intended behaviour is seen in the 'normalised' chart.

The zero value is a consequence of your step 2.

My method is shown in my original post and the sheet and first two charts above.

I think we might be suffering from understanding 'normalisation' differently. I was aiming to emulate manually what happens in EVA (Bruker) when I normalise two scans.

Ok, I see.

No problem Ian!

My response was to "would like to normalize the intensity of different XRD curves, to be able to compare them", which I'd like to do correctly as well. Especially, with "XRD curves for a polymer composite" with potential amorphous halo and possible preferred orientation, this would become even more critical. Yes?

I see a myriad of similarities and differences between the two normalization techniques. I have no explanation. I'm suspicious of any method in XRD powder analysis that uses absolute intensity values to do anything, instinctively. I might need to be a bit more open-minded perhaps. Generally, using absolute intensity value leads to a lot of "hand waving" and pontification after the fact. :-)

Happy Holidays! Best wishes!

https://www.flickr.com/photos/85210325@N04/31437185510/in/dateposted/

Thank you for the discussion everyone. I just wanted to state here that I have polymer films and not powder, so in that case will the above process still work?

In that case the normalization technique becomes even more critical. I would try both if possible. I'd also rotate the sample about its surface normal and repeat the diffractograms to IDENTIFY any preferred orientation, which in my opinion will be present in this case.

• What is the Nano structural morphology of the film?
• What substrate have you used if any?
• What chemistry?
• Processing history? How was the polymer film produced?
• Film thickness?
• How is the sample mounted in the diffractometer?

The more information you provide will mean the better response you will have.

• Posting original XRD data will help.
• Posting relevant literature will help.

I also suggest a transmission Laue image (see example below for Kevlar,  Aluminum foil and RDX/Al composite) using a simple dental film, unless you have the good fortune of working with a 2D detector :-) This will give you a better idea than numerous time intensive conventional powder diffractograms.

Just remember you are using a powder diffraction method good for randomly oriented crystalline powders to evaluate the Nano structure of a composite sample that may contain any combination of the following constituents:

Imagine trying to deconvolute the complexity of these 2D diffractograms with a single equatorial scan diffractogram. Tall order!

https://www.flickr.com/photos/85210325@N04/7975835813/in/album-72157632728981912/

https://www.flickr.com/photos/85210325@N04/7944785794/in/album-72157632728981912/

https://www.flickr.com/photos/85210325@N04/7988731098/in/album-72157632728981912/