Dear Umesh Kumar Singh , for EDS analysis we always use the highest accelerating voltage that allows our SEM, 30kV, independently of the sample nature. The optimum choice of accelerating voltage is determined by the elements present in the specimen. The accelerating voltage (in kV) should be not less than twice the highest excitation energy Ec (in keV) of any element present, in order to obtain adequate intensity.
The interaction volume (of the beam with the sample) depends on the beam current, so strong currents means large count rates and therefore faster analysis and better statistics. However these large interaction volumes cause a lower spatial resolution, which wouldn´t be a problem for an uniform sample. New EDS detectors, with large sensors, can achieve good results with lower currents.
It shows 2-6% error in major concentration elements but up to 99.99% error in minor or trace elements. So what is the exact percentage of error in EDS?
Dear Umesh Kumar Singh , about the error in EDS, it depends on the element and its concentration and also on the EDS detector sensitivity. Light elements (atomic number Z low) are harder to detect by EDS and therefore more error could be associated to them, but probably the main source of error is the low concentration of elements, because the abundant ones induce a X-ray background singal that hides the low intensity signals of these low concentration elements.
Reducing the detection limit and the error require more counts, wich can be achieved by increasing the measurement time or /and the accelerating voltage. The idea here is increasing the counting of X-ray photons, so longer time let accumulate more counts, the same effect causes a higher current beam, and also you can increase this counting rate by means of more efficient detectors (such as Si or Si(Li) Drift Detectors (SDD) or larger detectors given that the X-rays emitted by the sample can be emitted randomly the larger the detector area the more X-ray photons it can collect.