An electron microprobe study of the influence of beam current density on the stability of detectable elements in mixed‐salts (isoatomic) microdroplets

SUMMARY This paper shows that a means of accurately measuring beam current during microprobe analysis of inorganic fluid microdroplets is essential, since certain elements were sublimated from such specimens under easily achieved beam current densities, i.e. S at 1·8 nA/μm 2 , K at 2·5 nA/μm 2 , Na at 3·5 nA/μm 2 , P at 5·3 nA/μm 2 . In comparison, Cl was volatilized even under the mildest conditions used (0·35 nA/μm 2 ), and Ca, Mg and Co were stable under the severest operating conditions (7·1 nA/μm 2 ). Elements were less stable in large (3 μm diameter) droplets than in small (1 μm) droplets under identical irradiation conditions. The onset of volatilization is a direct function of the current delivered per unit area and not of the total integrated dose. The addition of 50 g/l of urea to the mixed‐salts (isoatomic) solution, or (a) the mounting of the droplets so that the carbon‐celloidin support film was interposed between them and the electron source, and (b) top‐coating the droplets with carbon, did not, in general, raise the threshold of volatilization of a given element, but did effectively retard the rate of loss at current densities above the volatilization threshold. A literature survey confirmed that similar losses can occur from biological tissue specimens, albeit at higher beam current densities. Finally, the possibility that local specimen heating during electron/specimen interaction is a cause of element loss during microprobe analysis is discussed.