Dear Manila, the main reasons are a good energy resolution, and in the constant analyser energy mode, the resolution is constant over the entire energy range measured. Second, and even more important is the higher transmission, i.e. larger possible count rates enabled by the hemispherical analyser, since it is focussing in two directions, while a cylindical mirror analyser of a segment of a cylinder only focusses in one direction. Thus the output signal is easier to measure, and the signal to noise ratio is better! Check the literature for more (I recomment any textbook on electron spectroscopies like Ertl-Küpper or Briggs (Practical surface analysis). Hope this helps, Dirk
But would like to add one more reason - the so-called acceptance angle (i.e. the angle under which the majority of emitted photoelectrons will be collected by analyzer and recognized by detector). HSA usually has much wider acceptance angle than CMA due to construction pecularities. In brief, the larger acceptance angle, the stronger XPS signal.
Also with CMA is not possible to make Angle Resolved XPS measurements.
Technickal reason: the HSA has much wider parameters to tune - energy window (pass energy), retarding potential setup, .... .
All these allow to get the state-of-the-art XPS signal for precised XPS analysis (for instance, even in the case of element traces).
This is one more important reason for preffering the HSA for an XPS.
A couple of decades ago, CMA were used for Auger spectroscopy.
But don't worry, as far as I know, the modern XPS systems are not using any more CMA for XPS - only HSA. Three typical placements of HSA are now used: vertical (Thermofisher), under angle (ULVAC-PHI) and horisontal.
Most accepted the first and second approaches (they are decribed in my XPS lectures presented here. Please don't think that it is an advertisement).
The best HSA is believed to be the VG Scienta because of the nearly ideal transmitting fuction and the most possible wide acceptance angle.
Dear Manila, to add to Dirk’s answer: The focal point of this analyzer (at the entrance to the analyzer) is also ideally to incorporate an input lens system for imaging and retardation. The lens system also increases the specimen-analyzer working distance and thus the flexibility of the instrument. Another important property, unlike the CMA, is that the analyzer does not suffer from the specimen positional dependency of energy and resolution. Regards Wiets
One thing I also forgot: There are further options to easily adapt lateral resolution to XPS: Either by a (mechanical) variation of the entrance aperture of the analyzer (as it was the case already in the old XPS systems from the late 1980s e.g. in the VG ESCALAB V) or, in a state of the art way, by using a microchannel-plate or different type of 2D detector and simple correlate the location of the pixel on the detector to a location of the sample surface without moving sample, slits, or detectors. Thanks to the properties of the HSA, such a correlation can easily be done, and thus, a lateral resolution in XPS down to the micrometer scale is feasible! Hope this is of value as well, Dirk