On the mechanisms by which the charged droplets produced by electrospray lead to gas phase ions

A brief description of the process leading to the formation of the very sma ll droplets which ultimately produce gas phase ions is followed by a discus sion of the ion evaporation model (IEM) and the charge residue model (CRM). The IEM developed for small ions by Iribarne and Thomson is a well develop ed model which provides quantitative predictions for the rates of evaporati on of ions. The CRM attributed to Dole and extended by Rollgen does not pro vide a detailed consideration as to how the 'final' droplets containing onl y one ion are formed. Several experimental tests of IEM and CRM are discussed: The rates of forma tion of gas phase alkali ions M+ (Li+, Na+, K+, Cs+) determined from mass s pectra, are compared with predicted rates by the Iribarne and Thomson equat ion. Unfortunately, the available thermochemical data required for the eval uation of the theoretical rates are of insufficient accuracy. Therefore, th e theoretically predicted rates exhibit a large scatter. However, also, the experimentally determined relative rates on the basis of mass spectra are not in agreement. Relative rates for the alkali ions from this laboratory p redict nearly equal rates for the alkali ions while data by Leize et al, pr edict an increase from Li+ to Cs+. Both sets of data are compatible with th e theoretical predictions of the IEM equation. The CRM, when extended to in clude the effect of surface activity of ions, is compatible with the result s of nearly equal rates, but not compatible with the increasing rates of Le ize et al. It is very desirable to establish in future work as to which set of experimental data is correct. If the increasing rates of Leize et al. a re found to be true, a strong argument in favor of the IEM will be provided . Mass spectrometric observations of the intensities of the ions Na+ and Na(N aCl)(n)(+) obtained from aqueous solutions of NaCl at different concentrati ons are compared with ion distributions expected on the basis of IEM and CR M. in general, larger intensities of Na+ relative to Na(NaCl)(n)(+) would b e expected on the basis of IEM. The experimental data are found to correspo nd much more closely to the predictions of IEM. However, various assumption s had to be made in order to be able to make the predictions. Most importan t of these is the history of the droplets as they undergo Rayleigh fission, and in particular, the size, charge and number of offspring droplets. Sinc e accurate values for these quantities are not available, the conclusions i n favor of IEM are not definitive. The final stages of CRM where droplets have radii of a few nanometers and s everal charges and solute molecules cannot be treated with the Rayleigh equ ation. Loss of single charges (ions) is possible in this stage. if this is the case, the last stage of CRM will be IEM-like. Such a stage will lead to a blurring of the distinction between IEM and CRM, for small ions. Fernandez de la Mora has provided very strong evidence that globular, not d enatured, proteins are produced by CRM. More open, multiply-charged macroio ns could be produced by either IEM or CRM. The ions produced in the gas phase are not necessarily those present in the solution. Stable ions like the alkali ions are transferred without change to the gas phase. However, protonated bases may undergo changes as a conseq uence of the different basicity orders in solution and in the gas phase and other processes. (C)2000 Elsevier Science B.V. All rights reserved.