I'm not sure about Chlorine, specifically, but a wide XPS peak usually means that you have that element bonded to more then one kind of atom, which shifts the orbital's energy. For example, carbon bonded to carbon has a 1s peak energy of ~284.5 eV. Carbon bonded to oxygen (more electronegative, pulling the molecular orbital cloud "more to it's side", so to speak) has a 1s peak energy of ~286 eV. So a sample which contains carbon bonded both to carbon and to oxygen would show a broad carbon 1s peak, which could be separated by a process called deconvolution, showing two peaks inside the broad measured one. This is a relatively well known procedure, you can find more about it by searching for "XPS deconvolution".

So, like i said, i don't know much about the Chlorine 2p XPS shifts when bonded to this or that element, but from what you described, it seems that increasing the chlorine content changes the fraction of it that reacts and bonds to a certain kind of atom in your sample. By deconvoluting and quantifying the fraction of each bonding state (there can be more then 2, by the way), you can see how that change is correlated to the concentration. Seems like higher concentration increases a peak with slightly lower energy (you said angle, but I'm assuming you mean Binding energy in eV?), so you see the broadening of the peak due to the appearance of a second (or third...) bonding state, and a shift for lower values due to increased fraction of the lower energy state. 

I hope that helps!

Best regards,

Rafael

PS: On a usual XPS system, a peak with up to 1.5 eV FWHM is normal, so if what you are seeing is much bellow that, maybe it's nothing... 

PPS: I did a quick search and found this, since you are looking at the 2p peak, there is also the spin-orbit splitting effect, which makes the analysis a bit more complicated: http://xpssimplified.com/elements/chlorine.php

I'm not sure about Chlorine, specifically, but a wide XPS peak usually means that you have that element bonded to more then one kind of atom, which shifts the orbital's energy. For example, carbon bonded to carbon has a 1s peak energy of ~284.5 eV. Carbon bonded to oxygen (more electronegative, pulling the molecular orbital cloud "more to it's side", so to speak) has a 1s peak energy of ~286 eV. So a sample which contains carbon bonded both to carbon and to oxygen would show a broad carbon 1s peak, which could be separated by a process called deconvolution, showing two peaks inside the broad measured one. This is a relatively well known procedure, you can find more about it by searching for "XPS deconvolution".

So, like i said, i don't know much about the Chlorine 2p XPS shifts when bonded to this or that element, but from what you described, it seems that increasing the chlorine content changes the fraction of it that reacts and bonds to a certain kind of atom in your sample. By deconvoluting and quantifying the fraction of each bonding state (there can be more then 2, by the way), you can see how that change is correlated to the concentration. Seems like higher concentration increases a peak with slightly lower energy (you said angle, but I'm assuming you mean Binding energy in eV?), so you see the broadening of the peak due to the appearance of a second (or third...) bonding state, and a shift for lower values due to increased fraction of the lower energy state. 

I hope that helps!

Best regards,

Rafael

PS: On a usual XPS system, a peak with up to 1.5 eV FWHM is normal, so if what you are seeing is much bellow that, maybe it's nothing... 

PPS: I did a quick search and found this, since you are looking at the 2p peak, there is also the spin-orbit splitting effect, which makes the analysis a bit more complicated: http://xpssimplified.com/elements/chlorine.php

Thank you so much Rafael.

It was very good, enough elaborated explanation.

Thanks a lot.

You are welcome!

Rafael has actually covered most of the information. The area under the curve is used to calculate the amount of species present (this is always calculated as a ratio). So FWHM can not be looked at in isolation. Basically, Cl is very reactive, hence Cl maybe present in your sample as ions or ZnCl2 or as both. Then you will have 4 peaks. Cl+/-p  orbital ion and Cl+/-p orbital in bonding with Zn (if you are very unlucky a complex Zn-Cl-O). Hence, if all four are clubbed together, it would be hard to deconvolve.

I would suggest that you take ZnCl2, make a pellet and take XPS of that just to over-rule the possibility that I have stated above.

Thank you very much for the suggestions @Arun Sir.

In addition to the aforementioned, I would say that you should take into consideration that the Cl2p peak at around 200 eV should to be deconvoluted as doublet peak, both of them assigned to the same Cl-X compound.