Comparison of methods for generating bioaerosols

Bioaerosols include viable bacteria, viruses, dead bacterial cells, pollen, fungi and cell fragments, as well as numerous organic compounds derived from biomolecules as, for example, sugars, amino acids and methyl-derivatives. It has been shown that airborne bacteria may be viable also in the harsh conditions at high altitudes in the atmosphere and transmit diseases as well as act as cloud condensation nuclei (CCN) and ice nuclei (IN). Furthermore, they are able to influence cloud chemistry (Santl Temkiv, 2012). In order to study the properties of bioaerosols in a laboratory setting it is necessary to use a generation methodology that preserves bacterial viability and minimize coating with redundant solvent impurities as this will increase particle size and distort chemical analysis. Few studies describe generation of bioaerosols, especially with respect to aerosol coating, background concentration and viability. The objective of this work is to compare various aerosol generators for bacteria, spores and vesicles. Seven aerosol generators are investigated: atomizer (TSI Inc., Model 3075), Collison nebulizer (BGI Inc., 3-jet), vibrating orifice aerosol generator (VOAG, TSI model 3450), electrospray (TSI Inc., Model model 3480) and a bubble flask, a medical nebulizer (Dolema, Aeroneb ProX) and a custom-built liquid bubbler. At this point, exclusively Rhodoccocus sp. (a common soil bacterium) is studied. Particle number size distributions were measured with a scanning mobility particle sizer (SMPS, design: Lund University) and an aerodynamic particle sizer (APS, TSI Inc., Model 3321) in a 1 m stainless steel chamber (“Bio-cube”) with controlled ventilation, temperature and relative humidity. Bacterial viability is analysed by counting colony forming units (CFU) on gelatin membrane filters and – in future studies – possibly also by fluorescence. Figure 1 shows the particle number size distribution of the particles produces with the TSI atomizer. There is a peak of bacteria at 0.75 μm. The number concentration of bacteria were around 4 cm, while the background salt particles in the range 10 to 200 nm (GMD 40 nm) had a concentration around 3000 cm. The bubble flask generated lower bacterial concentrations, but also lower background. The VOAG produced uniform, but comparably large droplets (typically 20-40 μm). When dried, the background particles therefore had a size around 500 nm, while the bacteria had a peak at about 1.7 μm which indicates a significant coating. The electrospray could not be used for bacteria, since they are too large, but may be appropriate for bacterial outer membrane vesicles and spores. The full matrix of generation methodologies is not yet available, but the preliminary results clearly show that many of the methods commonly used for generation of bioaerosols are unsuitable for examination of key properties of the particles such as CCN or IN ability. A surface coating of bioaerosols may significantly alter these abilities especially when soluble substances are involved. Any background in the measurements may also make a systematic evaluation difficult.