Verifying the viable particle counts of biofluorescent particle counters by using inkjet aerosol generators

Abstract

A method was developed to verify the viable counts of BioTrak^TM (TSI Model 9510-BD) without using test bacteria. In this method, an inkjet aerosol generator (IAG) was used to deliver a known number of test aerosol particles to BioTrak in the given sampling time. An IAG generates two types of monodisperse test particles, inert particles and fluorescent particles. Inert particles consist of a chemical substance that does not emit fluorescence when particles are irradiated by the laser beam of BioTrak. Inert particles were used to test whether a BioTrak generates false positive in viable counts or not. Depending on the version of the discrimination algorithm, BioTrak generated false positive in viable counts over 2 − 6-µm particle-diameter range. Fluorescent particles are monodisperse particles with a trace mass of fluorophores, which simulate the autofluorescence emission of surrogate bacteria. In this study, two types of fluorophore pairs, β-NADH and riboflavin sodium phosphate (riboflavin in short) and Coumarin 30 and riboflavin, were used. The counting efficiency of viable particle (CE_viable) were evaluated by generating fluorescent particles at a 3.0–8.3-µm particle diameter. The average of CE_viable, and its uncertainty among three BioTrak used in this study, were 0.701 ± 0.090 (k = 2), which is close to the physical sampling efficiency of BioTrak reported by the manufacturer. The relative standard uncertainty of CE_viable due to day-to-day variation was 0.8% and 3.5%, when the first and second fluorophore pairs were used, respectively. These results proved that the proposed method can be regularly used to verify the viable counts of BioTrak.

Graphical Abstract

EDITOR:

• Hans Moosmüller

Acknowledgement

This work was partly performed under the “Budget for Promotion of Strategic International Standardization” commissioned by the Ministry of Economy, Trade and Industry (METI), Japan. The authors would like to express their gratitude to Dr. Shinichiro Fujii at the Biomedical Standard Research Group at the National Metrology Institute of Japan for letting us use their fluorescence spectrometer during this project.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 CE_viable were not evaluated using inert particles as test particles during BT-1 availability. However, the trend could be the same as BT-3 because the version of the BT-1 algorithm was the same as BT-3.

2 These values were riboflavin at 37 mg kg–¹ and β-NADH at 1.8 g kg–¹ for BT-1, 43–55 mg kg–¹ and β -NADH at 2.2–6.9 g kg–¹ for BT-2, 55 mg kg–¹ and Coumarin-30 at 1.8–2.2 mg kg–¹ for BT-2, 55 mg kg–¹ and β -NADH at 6.9 g kg–¹ for BT-3, and 55 mg kg–¹ and Coumarin-30 at 1.8–2.2 mg kg–¹ for BT-3. These values are also given in the caption of Figure 9.