Improvements in ion signal reproducibility obtained using a homogeneous laser beam for on-line laser desorption/ionization of single particles


  • Ryan J. Wenzel,

    1. Department of Chemistry, University of California at Riverside, Riverside, CA 92521, USA
    Search for more papers by this author
  • Kimberly A. Prather

    Corresponding author
    1. Department of Chemistry and Biochemistry, University of California, San Diego, La Jolla, CA 92093-0314, USA
    • Department of Chemistry and Biochemistry, University of California, San Diego, La Jolla, CA 92093-0314, USA.
    Search for more papers by this author


A major factor limiting on-line single particle mass spectrometry techniques from becoming more quantitative is the large shot-to-shot variability in ion intensities observed in the laser desorption/ionization (LDI) mass spectra.1,2 In previous work, lab-generated particles showed fluctuations of up to 152% in the absolute ion intensities in averaged spectra of 200–300 ‘identical’ particles.2 Most of these fluctuations were attributed to inhomogeneities in the laser beam profile, leading to significant differences in the power each particle encountered depending on the position in the LDI laser beam where it underwent analysis. The goal of the work presented herein is to determine whether a fiber optic actually reduces the observed variability in single particle LDI mass spectral data. Initial results are presented for individual single component organic particles composed of 2,4-dihydroxybenzoic acid (2,4-DHB) analyzed using a low-power flat-top laser beam profile created by sending an ultraviolet (266 nm) DI laser through a fiber optic. Relative standard deviations of the total ion intensities for peaks in individual spectra are reduced to 31%. Single particle spectra, compared with and without laser homogenization at the same nominal laser fluence, show a marked enhancement. Specifically, the ion signal patterns of the 2,4-DHB particle spectra obtained using a homogenous LDI beam look identical to one another (i.e. only one particle type was produced with a commonly used neural network grouping algorithm), whereas without beam homogenization 25 different particle types (based on ion intensity patterns) were obtained. Future publications will explore more particle types and matrices but the initial results described herein are quite encouraging. Copyright © 2004 John Wiley & Sons, Ltd.