I need to analyze tens of XPS spectra from the papers. Moreover, I do not have numerical data of the spectra. For now I can see two options of how to do this. I could use Origin Pro to digitize the data of each spectrum and then past the digitized data to a free software like XPSPEAK41 and after fitting each peak, I could finally compare them. Another option is comparing the peaks' areas in Photoshop(of course after considering the background by a linear approximation).
I have tried both options and the error was around 10% of the values of at. % given in the papers. Nevertheless, some papers do not provide the values and I do not want to waste time on something that just does not work.
What do you think about analyzing XPS spectra using Photoshop?
I old times people just printed the spectra in identical scales, then they cut out the paper and weight it on a sensitive balance. And they came up with excellent compositions.
I think that your procedure using photoshop is more or less a digital version of the "old-style" way! I think it may work properly, regards, Dirk
I think it is reasonable to use the both ways you described, although the accuracy strongly depends on the quality of published plots.
I use Paint.NET for graphical integration of MS plots. It is a free software:
http://www.getpaint.net/
Here are a few other digitizing software apps ...
http://digitizer.sourceforge.net
http://www.silkscientific.com/graph-digitizer.htm
http://www.uniphiz.com/findgraph.htm
http://www.igorexchange.com/node/2014
I think that the error of 10% is very realistic. You will have always this order of magnitude analyzing the data from the small graphs published in the papers.
A very robust software for XPS analysis is AAnalyzer. You can try the trial version for free during one month: http://rdataa.com/aanalyzer/aanaHome.htm
Mihail, maybe it is too late but I know that sometime people show the XPS spectra for presentation preposes. This means that the spectra were scaled already somehow (!!!???). For calculations (at %) the original data were used (so everything is fair). To used "scaled" data (typical you see in arb. units) could be problematic for post-analysis. Otherwise, as stated above Phototoshop should to the job.
Good luck
Webplotdigitizer is the best and free. It works online as a well as offline
https://youtu.be/PW1nYO8ayMc
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Cell Biology
- Culture Cells
- Cell Line