protein components in the artificial saliva, mucin (one of major components of saliva) was added

to the total protein concentration level in saliva. The mucin-containing saliva is the best

substitute of natural saliva in theological properties, and viscosity and elasticity of this medium

are responsible for the protective role of saliva against desiccation (Vissink et al. 1984).

Depending on the virus suspension in the nebulizer, the size of dry aerosols or droplet nuclei (dp)

can be calculated from the droplet diameter, dd, according to Eq. (4-2) (Hinds 1999c).

 dp = d (F,) 1 (4-2)

where Fv is the volume fraction of solid content in nebulizer suspension. The volume fraction of

solid content for MS2 suspension in DI water, 0.25% tryptone solution, and artificial saliva was

9.9x10-4, 3.5x10-3, and 2.1x10-2, respectively. After complete evaporation, the particle size of

MS2 aerosols generated from DI water, tryptone solution, and artificial saliva was 0.10 dd, 0.15

dd, and 0.28 dd, respectively.

 Seven particle sizes were selected including (1) 30 nm, which is close to the nominal

MS2 primary particle size, (2) 230 nm, which is the upper limit of particle size measured by the

SMPS when the sample flow of the electrostatic classifier is set at 1.5 Lpm, and (3) 60 nm, 90

nm, 120 nm, 150 nm, and 200 nm, which provides information of intermediate sizes.

PCR assay

 Before submission to PCR analysis, 4-mL samples were concentrated to 280 [L by

using an Amicon ultracentrifugal device (UFC 810096, Millipore, Bedford, MA, USA) followed

by RNA extraction with QIAamp Viral RNA mini kit (QIAGEN Inc., Valencia, CA, USA)

according to the manufacturer's instructions and stored at -80 OC. A previous study (O'Connell

et al. 2006) of real-time fluorogenic reverse transcriptase (RT-PCR) assays for detection of MS2

was followed for design of primer and probe sequences. The GenBank accession number was