DESIGN AND PERFORMANCE OF A MATRIX-ASSISTED LASER DESORPTION TIME-OF-FLIGHT MASS-SPECTROMETER UTILIZING A PULSED NITROGEN LASER

The design considerations and experimental performance of a linear time-of-flight mass spectrometer are reported for performing matrix-assisted laser desorption studies. A simple pulsed gas-discharge nitrogen laser (337.1 nm) is successfully used in contrast to the more widely used frequency-quadrupled (266 nm) or frequency-tripled (355 nm) Nd:YAG solid-state laser. Optical considerations in utilizing the pulsed nitrogen laser are discussed and a simple optical arrangement is described which allows for suitable imaging of the poor spatial beam profile of the pulsed nitrogen laser. Laser spot sizes of 150 x 450-mu-m are obtainable. As with the frequency-tripled Nd:YAG laser, sinapinic acid is found to be the most useful matrix for producing protonated molecular species from proteins. Appropriate laser power levels are determined, as are matrix/sample levels. Adequate response for most small to medium molecular weight proteins is obtained for less than 1 pmol of sample. A simple einsel lens incorporated into the ion source does not appear to provide any significant of the laser-desorbed ions; however, a constant d.c. voltage applied to beam steering plates enhances the ion signal significantly. Selective, pulsed deflection of the low-mass ions produced from the matrix is also utilized to prevent excessive saturation of the microchannel plate ion detector. High source potentials are found to provide improved resolution and sensitivity in comparison with lower source potentials combined with post-acceleration at the detector. Representative mass spectra of several proteins and peptides are presented. Increased formation of photoinduced adduct ions are observed in comparison with that reported for matrix-assisted laser desorption experiments utilizing a Nd:YAG laser and significant amounts of dimer and trimer ions are produced. The former may be due to non-optimized sample/matrix compositions as opposed to wavelength-induced effects, whereas the latter is only observed at higher protein concentrations. Significantly more peak broadening than would normally be expected is observed above 20 000 u. This may be due to the post-acceleration design of the microchannel plate detector used.