偏最小二乘(Partial least square,PLS)聚类法是一种全新的气溶胶单粒子光谱数据处理方法,是利用具有"自组织机制"的PLS回归算法去完成数据的聚类。阐述了PLS聚类对模拟数据集的运用以展示这种方法的一般特征及有效性,然后应用到气溶胶激光飞行时间质谱数据以展示PLS聚类的正确性及成功运用,最后将PLS聚类应用到氯化钙、氯化镁、氯化钠及氯化钾四种气溶胶单粒子激光击穿光谱混合数据集,通过分析聚类获得的树形图和图中节点的统计特性,剖析了正确聚类及发生错误划分的原因,表明了PLS聚类方法在气溶胶单粒子谱分析方面的应用潜力。
An evaporation/condensation flow cell was developed and interfaced with the matrix-assisted laser desorption/ionization （MALDI） time-of-flight mass spectrometer for on-line bioaerosol detection and characterization, which allows matrix addition by condensation onto the laboratory-generated bioaerosol particles. The final coated particle exiting from the con- denser is then introduced into the aerodynamic particle sizer spectrometer or home-built aerosol laser time-of-flight mass spectrometer, and its aerodynamic size directly effects on the matrix-to-analyte molar ratio, which is very important for MALDI technique. In order to observe the protonated analyte molecular ion, and then determine the classification of bi- ological aerosols, the matrix-to-analyte molar ratio must be appropriate. Four experimental parameters, including the temperature of the heated reservoir, the initial particle size, its number concentration, and the matrix material, were tested experimentally to analyze their influences on the final particle size. This technique represents an on-line system of detection that has the potential to provide rapid and reliable identification of airborne biological aerosols.
