Data processing in on-line laser mass spectrometry of inorganic, organic, or biological airborne particles

A general solution for data processing of large numbers of micrometer- or submicrometer-particle mass spectra in aerosol analysis is described. The method is based on immediate evaluation of bipolar laser desorption ionization mass spectra acquired in an on-line (impact-free) time-of-flight instrument. The goal of the procedure is a characterization of the particle population under investigation in terms of chemical composition of particle classes, particle distributions, size distributions, and time courses, rather than an investigation of each individual particle. After automatic peak analysis of each newly acquired bipolar mass spectrum, the mass spectral information is statistically evaluated by a fuzzy clustering algorithm, providing for an immediate attribution of the particle to predefined particle classes. The particle distributions over these classes can be monitored as a function of time and particle size range. Definition of the particle classes as used for on-line evaluation is performed in an earlier step, either by manual approach, or by selection from a particle class database, or, as in most cases, by fuzzy clustering of a set of particle mass spectra from the population (the aerosol) under investigation. Definition of the particle classes is depending only on the distinguishability of the spectra patterns of different particles. It is not necessary for the clustering approach to fully “understand” the mass spectra. The range of possible applications of the method is therefore very broad. Particles dominated by inorganic components, as typically observed in aerosol chemistry for example, can be investigated the same way as organic particles (e.g., from smoke or automobile exhaust) or even biological particles such as bacteria, yeast, or pollen. The data processing method has been successfully applied in several fields of stationary applications and will be employed in mobile instruments for large scale field studies in atmospheric chemistry, engine combustion research, and the characterization of house dust.

[1]  T. Peter.  Airborne Particle Analysis for Climate Studies , 1996, Science.

[2]  Allen,et al.  Direct observation of heterogeneous chemistry in the atmosphere , 1998, Science.

[3]  S. Downey,et al.  Single particle characterization by time-of-flight mass spectrometry , 1995 .

[4]  S. Friedlander,et al.  Smoke, dust, and haze , 2000 .

[5]  M. P. Windham.  Cluster validity for fuzzy clustering algorithms , 1981 .

[6]  R. Grieken,et al.  Elucidating the composition of atmospheric aerosols through the combined hierarchical, non-hierarchical and fuzzy clustering of large electron probe microanalysis data sets , 1995 .

[7]  J. Heyder,et al.  The growth of atmospheric particles in moist air , 1995 .

[8]  K. Prather,et al.  On-Line Characterization of Individual Particles from Automobile Emissions , 1997 .

[9]  K. Prather,et al.  Real-time monitoring of pyrotechnically derived aerosol particles in the troposphere , 1997 .

[10]  René Van Grieken,et al.  IDAS : a Windows based software package for cluster analysis , 1996 .

[11]  Peter J. Rousseeuw,et al.  Finding Groups in Data: An Introduction to Cluster Analysis , 1990 .

[12]  B. Spengler,et al.  Source identification of single particles by on-line laser mass spectrometry , 1998 .

[13]  K. Prather,et al.  Real-Time Measurement of Correlated Size and Composition Profiles of Individual Atmospheric Aerosol Particles , 1996 .

[14]  D. Murphy,et al.  Aerosol characteristics at Idaho Hill during the OH Photochemistry Experiment , 1997 .

[15]  Anthony S. Wexler,et al.  Sulfur speciation in individual aerosol particles , 1996 .

[16]  R. Grieken,et al.  Hierarchical cluster analysis with stopping rules built on Akaike's information criterion for aerosol particle classification based on electron probe X-ray microanalysis , 1994 .

[17]  B. Turpin,et al.  Elemental composition and morphology of individual particles separated by size and hygroscopicity with the TDMA , 1996 .

[18]  R. Grieken,et al.  Characterization of individual particles in the North Sea surface microlayer and underlying seawater: comparison with atmospheric particles , 1992 .

[19]  B. Spengler,et al.  On-line measurement and characterization of single particles using laser mass spectrometry and multivariate data analysis , 1996 .

[20]  Daniel M. Murphy,et al.  On-line single-particle analysis by laser desorption mass spectrometry , 1991 .

[21]  Erik Swietlicki,et al.  Phase partitioning of aerosol constituents in cloud based on single-particle and bulk analysis , 1997 .

[22]  R. Grieken,et al.  Chemical characterization and source apportionment of individual aerosol particles over the North Sea and the English Channel using multivariate techniques , 1991 .

[23]  J. Martins,et al.  Fine mode aerosol composition at three long-term atmospheric monitoring sites in the Amazon Basin , 1994 .

[24]  Bernhard Spengler,et al.  Simultaneous Detection of Positive and Negative Ions From Single Airborne Particles by Real-time Laser Mass Spectrometry , 1996 .

[25]  J. Bezdek,et al.  FCM: The fuzzy c-means clustering algorithm , 1984 .

[26]  M. Andreae.  Soot Carbon and Excess Fine Potassium: Long-Range Transport of Combustion-Derived Aerosols , 1983, Science.

[27]  Matrix-Assisted Laser Desorption/Ionization of Size- and Composition-Selected Aerosol Particles , 1996 .

[28]  P. Verheijen,et al.  ON THE PERFORMANCE OF AN ON-LINE TIME-OF-FLIGHT MASS SPECTROMETER FOR AEROSOLS , 1997 .

[29]  L. Reimer,et al.  Scanning Electron Microscopy , 1984 .