With EDX about 1-2 μm depth resolution is possible. It uses high energy x-ray photons therefore have a large interaction volume, on the order of µm in SEM and 100s of nm in TEM. XPS will give you more accurate composition of surface as compared to EDX.

X-Ray excape depth is strongly affected by accelerating voltage of (SEM?) microscope and specimen density. You can estimate the depth by this equation:

R = 0.064 (E_0^1.68 – E_C^1.68) / ρ, where R = spatial resolution in um, E_0 = accelerating voltage [keV], E_c = critical excitation energy [keV] and ρ = mean specimen density in g/cc

(or you can just google "monte carlo simulation" and it will give you better explanation) :D

Agreed with the above answers, It is dependent upon accelerating voltage and sample type (conductor/insulator). SEM - In a 5kV, e-beam can penetrate to a depth of 0.5 microns, in a 15kV e-beam can travel to a depth of 1microns. With increasing a voltage, e-beam can go to a depth of 2microns and up to that depth, X-ray could be generated and therefore composition can be examined in that range. Therefore, EDAX is not a surface science technique as XPS.

For 20 years we used 1-1.2  micron-thick layers, and , in our  EDAX analyses, we never got beyond that depth. 

The excitation depth/volume of the primary electron also depends on angle of incidence.  And with modern guns lower primary energies can be reached as stated above.  The fluorescence attenuation lengths depend on the energy of the X-Ray photon emitted and the web CXRO database is useful I find for this.  The books are relatively old but are good

Reed, S. J. B., 1993, Electron Microprobe Analysis (2nd Ed.), Cambridge University Press

and for surface analysis

Practical Surface Analysis, Auger and X-ray Photoelectron Spectroscopy

by D. Briggs (Editor), M. P. Seah (Editor)  Wiley

and more recently for SAM

Scanning Auger Electron Microscopy

Martin Prutton (Editor), Mohamed M. El Gomati (Editor) Wiley

along with

Introduction to Surface and Thin Film Processes, John A. Venables, Cambridge University Press (2000)

but essentially the the low energy photoelectron gives you surface information and the photon bulk (microns but limited by excitation depth of primary electron and its energy). 

First of all, EDAX is the acronym of a Company, correctly you should write EDX- Analysis. This is only the half of the story. The most important part is the excitation. We distinguish the excitation by particles and by photons. Particles means protons and/or electrons and the result is the emission of X-rays (photons). The methods are PIXE, SEM-EDX. The excitation by photons results in X.ray fluorescence. Methosd like XRF, TXRF, and Synchrotron Radiation based XRF, etc. All These methods are combined with the same detection system EDX.

The Information depth is strongly dependent on the excitation (particles, photons) and of the material in question. There is no thumb-rule, each case have to be inspect individually. Therefore I recommend E.P.Bertin, Priciple and practice of X-ray spectrometric Analysis, Plenum Press.

Regards Alex