Ionization of vapor molecules by an electrospray cloud

Abstract

We first summarize the early work by Fenn and colleagues on vapor ionization by an electrospray cloud (subsequently dubbed secondary electrospray ionization, or SESI), followed by analysis via an atmospheric pressure ionization mass spectrometer (API-MS). It was in part reported in Ph.D. theses and presented to ASMS conferences, but remains largely unpublished. After spending 20 years in limbo, various aspects of their method have begun to be used, leading recently to outstanding limits of detection of ambient volatiles (parts per quadrillion; ppq). There is still much room for improvement of the method, as the ionization probability (defined as the concentration ratio n s / n v between ionized vapor and neutral vapor) is p ∼ 10 −3–10 −4. This result follows from recent approximate measurements, as well as from a newly derived expression for the equilibrium value p e under space charge dominated conditions typical of an ES cloud (probably also of a corona discharge): p e = kɛ o /( Z s q). This simple expression is derived from a balance between space charge dilution ( dn s / dt = − Z s n s n i q/ ɛ o ) and the rate of ionization of neutral vapor ( dn s / dt = kn v n i ). It is independent of the concentration n i of the charging drops (or ions), but depends on the electrical mobility Z s of the ionized vapor and the net charge q on the charging species (ions or drops). ɛ o is the electrical permittivity of vacuum. Still unresolved is the important mechanistic issue of whether the charge-exchange rate coefficient k corresponds to vapor collisions with ES drops ( k d ), or rather with individual ions formed after complete drop evaporation ( k p , based on the ion-induced-dipole interaction model). Coincidentally, an upper limit obtained for k d is comparable to k p . The ionization efficiency of SESI is compared to that of radioactive and corona sources. Appendices include information on the various rate coefficients fixing p e . They extend the ion-induced-dipole interaction model to account for the relatively large size of most vapor molecules of interest. Size effects on k p are found to be modest, in contrast with the strong size dependence of the mobility of large ions.