Re: Auger electron spectroscopy (AES) spectrum question

[EMAIL PROTECTED] (Marko) wrote in message news:<[EMAIL PROTECTED]>...
> Hello!
> 
> In Auger electron spectroscopy (AES) which in used in analyzing
> surfaces of conducting and semiconductions solids, Auger electrons are
> emitted and their energy is measured, since every different chemical
> element (not hydrogen or helium) gives different energies for emitted
> electron, we are given a method of determining the chemical
> composition and the concentration of chemical elements at surface of a
> solid sample.
> 
> The result is presented often as a graph, with ordinate N(E), giving
> intensity (counts per second) and abscissa E, gives the energy in eV.
> 
> The graph shows how the intensity varies with energy, this is called
> Auger spectrum in AES.
> Where the intensity is highest there is highest concentration of that
> specific chemical element which emitted Auger electron energy is equal
> to the value of energy, E giving maximum value of intensity, N(E).
> 
> To better see were the maximum variation are, the graph given instead
> show differentiated Auger spectrum, that is the ordinate is N'(E) and
> absicssa is E.
> What is interesting here is where the peaks are lowest, this according
> to Auger Handbook gives what chemical element are at surface.
> 
> My question is why are the lowest peaks, minimum values of
> N'(E)interesting?
> 
> Since, according to differential calculus, this gives the minima of
> N(E), at points left of lowest point of N'(E), N'(E) is decreasing,
> and at right of this point N'(E) is increasing, this gives that this
> peak gives minumun of N(E), so why are minimum peak interesting in
> Auger electron spectroscopy, since lowest peaks in differentiated
> Auger spectrum doesn't show where the intensity is maximum instead it
> shows where the intensity is minimum.
> 
> Thanks for your help!
> Marko

Coming from the XPS-corner, where the Auger Peaks are a by-product,
i'll try an educated guess.
The lowest absolute values in the differentiated graph are of interest
since they give the maxima as well. You can differentiate between
minima and maxima easily. the minima will be very wide (diferentiated
"noise" base line) while the maxima will be more discrete.
Since Auger peaks are wide signals, i guess it makes sense to work
with the diff-method.
I am afraid that you somehow mixed up something. If N'(E) is
decreasing left of the "lowest point" it means that the increase is
getting less, the closer you get to the lowest point. The decrease of
the increase means that the original is still going up -> towards a
maximum.
I am really sorry for that last sentence ;-)