Bioaerosol Release from Concentrated Microbial Suspensions in Bubbling Processes

Abstract

Bursting bubbles are one of the most common mechanisms in aerosols’ formation from natural and artificial waterbodies. The presence of microbial materials in the liquid could cause their aerosolization and generation of bioaerosols. The process depends on a number of parameters of the gas and liquid involved. This project investigated the influence of the air flow, bubble size, the temperature of the liquid and its surface tension on the efficiency of bioaerosol generation. It was found that the bioaerosol is more efficiently produced at higher air flow rates and smaller bubble size. The influence of the liquid temperature was also identified to be quite high, reaching an order of magnitude of the bioaerosol concentration over the temperature range from 4 °C to 38 °C. The addition of surfactants did suppress the foam formation, which was found to have a negative effect on the process; the rate of the bioaerosol generation increased with the increase in the antifoam concentration.

1. Introduction

Bioaerosols are suspended particles of biological origin, for example, viruses, bacteria, fungal spores, pollen and residue or products of organisms [1]. They could be released by natural or anthropogenic processes, which were comprehensively discussed in Mirskaya and Agranovski [2].

Bursting bubbles are one of the most common mechanisms in aerosols’ formation from natural and artificial waterbodies [3,4,5]. For instance, bursting bubbles from breaking ocean waves produce sea spray aerosols and condition the microorganism exchange between the ocean and the atmosphere [6,7,8]. Bubbles, formed by the air entrainment in liquid, absorb and carry microbial load from the liquid to the surface, where they burst, transferring their content to the atmosphere via hundreds of droplets [9]. A rupture of a bubble resting close to a gas–liquid interface can produce two types of droplets: (i) film drops and (ii) jet drops. Film drops are generally smaller than jet drops, and they are formed by fragmentation of the bubble film cup; meanwhile, jet drops can be up to 5–20% of the bubble diameter formed by a breakup of a central jet forming during the collapse of the bubble cavities [10,11].

The single bubble experiments conclude that the bubble size is strongly correlated with the number and types of aerosolized droplets [4,5,6]. Bubbles with diameters of less than 3–6 mm burst, generating jet droplets, while larger diameter bubbles burst with the fragmentation of their cap film, producing predominantly film droplets [6,9]. The number of film drops produced by a bursting bubble is proportional to the square of the bubble diameter [10].

However, fewer studies are focused on the cumulative effect of multiple bubble flow regimes, such as aeration, on bioaerosol production. For instance, Günther et al. [12] investigated a continuous flow of multiple bubbles with a mean diameter of 2.92 mm and showed an increase in the film drops’ formation compared to a single bubble of the same size and under the same parameters. The bubble–bubble interactions in the flow explained such results.

Apart from the bubble size, other parameters, such as surface tension, gas airflow, composition and concentration of the solution, relative humidity, travel distance to the surface and temperature affect the bioaerosol release [13,14]. Effects of atmospheric temperature on the bioaerosols’ properties have been actively investigated [15,16]. However, the effect of liquid phase temperature on the bioaerosol release is less represented in the literature. It was shown that the increased temperature of the solution prolongs the time the bubbles stay at the liquid–gas interface before bursting with a bigger droplets formation [9]. Sun et al. [17] observed the gas and liquid temperature impacts on the bubble formation and concluded that the gas phase temperature did not cause significant alterations in the bubble detachment compared to the liquid temperature.

Changes in the surface tension affect bubble formation and bursting mechanisms; order of magnitude variations in the efficiency of aerosol generation were observed with changes in the solution film pressure [11]. Foam formed as the result of bubbling tends to suppress droplets’ production [13]. Antifoam agents reduce surface tension and prevent foaming. The addition of surfactants significantly changes the bubble flow structure, slows the bubble rising motions and reduces the bubble–bubble interactions, especially for the bubbles with a diameter lesser than 1.5 mm [18]. The presence of surfactants in the solution stabilizes the bubble, which results in a thinner film and generates a higher number of smaller jet droplets [4,13].

The biological content of bursting bubbles can negatively affect human health by dispersing pathogenic microorganisms and contamination from polluted water bodies. Wastewater and sludge are well-known sources of clinical pathogens that may become airborne during wastewater treatment, causing severe health effects [19].

Industrial processes significantly contribute to the generation of bioaerosols [2]. Due to the high risks of pathogens’ transfer to the atmosphere, there are many studies focused on bioaerosol sampling and monitoring of the airborne particle concentration at contaminated sources [19,20]. Wastewater treatment plants are a significant anthropogenic source of bioaerosols. The highest bioaerosol concentrations at wastewater treatment plants relate to the treatment facilities with aeration and mechanical agitation of liquids [21]. The aeration system design can determine the rates of bioaerosol release. For example, an aeration system with horizontal rotors and surface turbines releases two orders of magnitude more bioaerosols than air diffuser aerators [22]. Aeration with horizontal rotors also generates a wider variety of pathogens [23]. An upgrade of old coarse aeration systems to fine bubble aeration resulted in a significantly reduced bioaerosol concentration [24]. Therefore, a better understanding of parameters influencing bioaerosol generation in different industrial processes can help create safer working environments by controlling the emission of bioaerosols.

This study investigated the bioaerosol release from a microbial suspension during continuous bubbling under controlled laboratory conditions. The experiments replicated industrial and natural processes of continuous multi-bubble flow. The effect of temperature of the liquid phase, bubble size, airflow and surface tension on the bioaerosol concentration has been investigated and discussed.

2. Materials and Methods

2.1. Microorganisms

Escherichia coli (E. coli), a gram-negative coliform bacteria, was selected as a model organism for this study based on its wide presence in the air of many occupational settings (i.e., hospitals, recycling points, wastewater treatment facilities) and its regular use in bioaerosol laboratory-based studies [25,26].

A fresh culture of E. coli (ATCC 11303) was grown by inoculating 200 mL of sterile nutrient broth (1.3 g/100 mL) (OXOID Ltd., Basingstoke, Hampshire, England) with a single bacterial colony and shaking at 150 rpm for 18 h at 37 °C. The culture was then centrifugated at 7000 rpm for 15 min (Centrifuge model 5810, Eppendorf, Hamburg, Germany), the supernatant was eliminated and the remaining cells were resuspended in 200 mL of sterile deionized water with following loosening of potential microbial clumps by treatment in the sonic bath for 10 min. Considering the fact that the concentration of aerosolized microorganisms in the proposed experiments strongly depends on the concentration of the microbial suspension, the following procedure was used to ensure uniformity of all experimental runs by maintaining the concentration of the bacterial suspension close (within 5% inter-batch discrepancy) to 10⁷ Colony Forming Units (CFUs) per millilitre. This concentration was selected because it is achievable in all microbial harvesting runs, ranging between 7.0 × 10⁷ and 1.0 × 10⁸ CFU/mL. On completion of the microbial harvesting procedure described above, 0.1 mL sample of the microbial suspension was acquired and 10-fold diluted with sterile water. Then, 100 µL aliquot of an appropriate dilution was spread on the surface of the Nutrient Broth (NB) plates. The remaining microbial suspension was placed in the fridge and kept overnight at 4 °C. In parallel, the plates were incubated at 37 °C overnight, and colonies were counted with a colony counter (Biolab, Clayton, VIC, Australia). The corresponding bacteria concentration was estimated, and the concentration was adjusted with sterile ionised water to make the volume of one litre with the bacterial concentration of 10⁷ CFU/mL, providing the integrity of all experimental runs. The suspension concentration was also checked on completion of experiments to ensure that it was stable during the daily experimental program. The results have never departed from the ones obtained at the beginning of the procedure by more than 5%, which could be considered satisfactory.

2.2. Experimental Setup

A schematic diagram of the experimental setup is shown in Figure 1. A heat resistant cylindrical glassware with the i.d. of 100 mm was used as a bottom part of the aerosol chamber. For observation purposes, the upper part of the chamber was made out of clear acrylic tightly slid into the glass cylinder to make the total height of 300 mm. Considering that the experimental program involved suspension heating, the bottom part was made out of heat resistant material.

The HEPA-filtered compressed air with the flow rate monitored by a mass flow meter (Model: #4043, TSI Inc., Shoreview, MN, USA) was discharged into the liquid 11 cm below the surface to provide the required regime of bubbling. Two different gas discharge devices were employed to investigate the influence of bubble size on bioaerosol generation. An orifice-type device with a diameter of 1 mm was used to produce larger bubbles, whilst a porous stone diffuser commonly used for fish tank aeration was used for producing smaller ones (pore size was measured to be 60 ± 18 µm). To enable a direct comparison of the results, six identical airflow rates (0.01, 0.25, 0.5, 0.75, 1 and 1.5 L/min) were used throughout the experiments, evaluating the relationship between the airflow rates and bacterial aerosolization capability for two selected bubble sizes.

The bioaerosol generated by bursting bubbles was monitored by a single-stage 400-holes BioStage Impactor (SKC Ltd., Eighty Four, PA, USA) operated by a BioLite+ vacuum pump (SKC Ltd., Eighty Four, PA, USA) at the air flow rate of 28.3 L/min. Before the experiment, an agar plate was placed inside the impactor, at the top of the aerosol chamber, as schematically shown in Figure 1. After sampling, the agar plates were labelled and incubated overnight at 37 °C. The number of culturable bacteria was assessed on completion of incubation using the colony counter; the results were statistically adjusted using positive-hole conversion tables [27], and the concentration of aerosolised culturable bacteria was calculated and presented in CFU/m³. Considering that the airflow rates used for the bubbling were much lower compared to the flow required for the impactor operation, 20 air make-up holes with 3 mm diameter were located peripherally, 30 mm above the liquid level. Such an arrangement enabled the supply of make-up air, ensuring smooth device operation at the required level of 28.3 L/min, and also played an important role in aerosol transportation to the impactor. The laboratory setup was placed in the Class II Biohazard cabinet (AES Environmental Pty LTD, Minto, NSW, Australia) to ensure no alien particles reaching the experimental flows. In addition, the cabinet was used to eliminate any bioaerosol breakthrough to the laboratory space.

The influence of the temperature of the microbial suspension on the bioaerosol generation rate was investigated at three distinctly different temperatures (4, 26 and 38 °C). Such selection was justified by the fact that the lowest one realistically represents a reasonable natural temperature of water reservoirs during winter time in northern countries. The highest is close to the maximum survival temperature of bacteria, and the middle one represents the temperature of reservoirs during summertime. In addition, all microbial suspensions existing in industrial settings would also be represented by the selected range of the temperatures. For this series of experiments, the suspension was removed from the fridge added to the aerosol chamber and immediately used for experiments undertaken at 4 °C. Then, the aerosol suspension was heated on a hot plate equipped with a stirring module (Model PC-420D, Labnet International Inc., Edison, NJ, USA) until the target temperature of 26 °C was reached. The experiments at this temperature were correspondingly undertaken. Finally, the procedure was repeated at the temperature of 38 °C. The suspension was continuously stirred at low revolutions to avoid local overheating and corresponding rapid microbial inactivation. The cumulative effect of the suspension temperature and the airflow rate was obtained.

An impact of the surface tension on the bioaerosol generation during bubbling was investigated by adding increasing doses of antifoaming agent Antifoam A concentrate (A6582 Sigma Chemical Company, St. Louis, MO, USA) to the bacterial suspension. In particular, 0.5 mL, 1.0 mL and 1.5 mL of the antifoaming agent were added to achieve the concentration of ~0.5 g/L, ~1.0 g/L and ~1.5 g/L (density of A6582 is 0.97 g/mL) of bacterial suspension to reduce the surface tension for consecutive experimental runs.

The environmental conditions were controlled and remained consistent during all experiments. The ambient temperature was 26 ± 0.5 °C, and the relative humidity was in the range of 50–55%. It is important to note that, considering the strong sensitivity of the impactor to the microbial concentration and the corresponding possibility of either under- or over-loading the agar plates, a number of preliminary experiments were conducted to determine the optimal sampling time to achieve a countable concentration of bacterial aerosols for all experiments described above.

3. Results and Discussion

3.1. Bubble Size and Airflow

The size and shape of bubbles for each bubbling regime were visually observed and assessed using a photography-based technique and image analysis. More than 20 high-speed photographs were taken for each set of experimental conditions, and some representative illustrations are shown in Figure 2. The results obtained are as follows: for the porous stone diffuser, the mean diameter was 1.9 ± 0.7 mm, and for the orifice, it was 8.7 ± 2.1 mm. Changing the air flow rate from 0.01 L/min to 1.5 L/min evidently resulted in different numbers of bubbles per unit volume; however, the mean bubble sizes remained relatively constant for both discharging devices and did not strongly depend on the air flow rate. This conclusion is confirmed by randomly selected pictures representing bubbling at different flow rates, shown in Figure 2. It should also be noticed that at the highest flow rate used in this study, bubbles tended to coalesce, making it harder to realistically assess their diameters. On this basis, no further increase in the flow rate beyond 1.5 L/min was used in this study.

The concentration of airborne bacteria at different air flow rates is shown in Figure 3. The bioaerosol concentration was increased with the increase in the flow rate, reaching the maximum value for the highest flow used in this study (1.5 L/min). Similar trends were reported in the literature [14] and are especially crucial for processes where bubbling is used for laboratory-based bioaerosol generation for research needs. It is also important to highlight that the amount of bioaerosol generated by bursting of smaller bubbles was significantly higher as compared to the bioaerosol production by the larger ones at the same corresponding flow rates. Such a situation was observed regardless of the fact discussed above, that bursting of the larger bubble produces more droplets, as their amount is proportional to the square of the bubble diameter [10]. However, for the same air flow rate, the number of smaller bubbles and their total surface area was significantly higher (~2 orders of magnitude) compared to the orifice generated larger ones, which offsets the outcomes reported by Wu [10]. Statistical tests (single factor ANOVA tests with a confidence interval of 95%) demonstrated that the difference between the results obtained for two bubble sizes was significant along the entire range of the air flow rates. Interestingly, as is shown in Figure 3, both graphs are steeper at the flow rates of up to 1 L/min and then become flatter for larger flows. This trend could be explained by much more intensive and violent bubbling at higher flow rates, where intensive interactions between bubbles could minimize the influence of the flow rate increase.

3.2. Effect of the Temperature of Microbial Suspension

The results of the bioaerosol release during bubbling through the microbial suspension at different liquid temperatures are presented in Figure 4. Both bubble sizes, small (Figure 4A) and large (Figure 4B), demonstrated increased bioaerosol release with the temperature rise. The difference was relatively small at the lowest flow rate of 0.01 L/min; however, it was increased with the flow rate increase, reaching the maximum values at the highest flow used in this study. The difference exceeded an order of magnitude for the temperature ranging between 4 °C and 38 °C. It must be noticed that, unlike the air flow rate, which could potentially increase beyond the maximum value used in this project, the microbial suspension temperature range is the maximum possible in the area of the bioaerosol studies; the freezing point restricts it at the bottom, and a dramatic microbial inactivation would occur if it increased beyond 38 °C.

This effect can be explained by significantly more efficient heat and mass transfer in heated liquids with higher vapour pressure and, correspondingly, elevated fluxes from hot liquid to the clean gas carrier. The results are in good agreement with reports recently published in the literature. In particular, Sun et al. [17] reported that the temperature increase in the liquid phase from 25 °C to 65 °C led to a 30% increase in bubble detachment volume attributed to the evaporation. Poulain et al. [9] showed that increased temperature of the solution led to more persistent bubbles that burst with the formation of bigger droplets.

3.3. Effect of the Surface Tension on the Bioaerosol Generation Rates

Figure 5 presents the changes in bioaerosol release from the bacterial suspension after the addition of surfactants. The observed results showed that the concentration of the airborne bacteria was generally higher for the increased amount of the antifoaming agent. A similar effect was observed for both smaller (Figure 5A) and larger bubbles (Figure 5B). The highest concentration of antifoam used in this study (1.5 mg/L) was associated with the largest increase in the bioaerosol generation. In contrast, the lowest antifoam concentration of 0.5 mg/L did not cause any changes above the reference level taken at zero antifoam concentration. Such an outcome is quite interesting, showing that excessive foam formation observed for the cases of low/zero antifoam presence is beneficial, causing some suppression of the bioaerosol generation. These results are in good agreement with the results obtained by Ke et al. [4], who reported that the mass of emitted particles per bubble is increased with the decreased surface tension.

Interestingly, the difference in the rate of the bioaerosol production from the suspension with the highest surfactant concentration compared to the one with the lowest amount of the antifoam was more substantial for the regime involving smaller bubbles (Figure 5). To explain such an outcome, as discussed above, small bubbles’ bursts cause the generation of the jet droplets, while larger diameter bubbles predominantly produce film droplets. The presence of surfactants lowers the surface tension, causing the rupture of liquid film, decreasing the efficiency of the process. Such a conclusion is in good agreement with the results reported by Néel and colleagues [13], who observed increased numbers of smaller size jet droplet production for bubbles with radii of ~1.5 mm with the addition of surfactants. However, it must also be noted that, regardless of the fact that the graphs depart from each other in Figure 5, the statistically significant difference was only observed for the highest flow rates for small bubbles, whilst the results were not statistically different for all flow rates involving the larger bubbles.

3.4. Contribution of Different Mechanisms to the Bioaerosol Release during Bubbling

Considering mass transfer processes during the bubbling of the gas through microbial suspension, one could conclude that two processes contribute to the bioaerosol release: bioaerosol generated by the bursting bubble/jet and bioaerosol transferred from the liquid to the gas during bubble rise through the suspension and released after the bubble burst. A series of experiments were undertaken to evaluate the contributions of both mechanisms. It is widely reported in the literature that the droplets produced during bubble burst are quite large and could reach a few dozen micrometres in diameter [28]. Obviously, it is much larger than the size of the E. coli particle (~1 µm) transferred to the gas phase during the bubble rise, and they could be simply and sharply separated from each other. To ensure sufficient separation, porous media with the fibre size of 20 µm, 2% packing density and of 2 mm thick were secured right above the air make-up hole’s level, providing an efficiency of around 3% for a 1 µm particle and >95% for particles larger than 10 µm. Obviously, the results obtained for the procedure when the filter was not installed represent the contribution of both mechanisms, whilst the scenario with the filter only illustrates bioaerosols released by the gas phase. Two air flows (0.25 L/min and 1.0 L/min) were used to evaluate the bioaerosol release with and without the porous medial installed. The results of the experiments are presented in

Table 1. Contribution of the bioaerosol release by the gas carrier and bursting film.

	Air Flow 0.25 L/min		Air Flow 1.0 L/min
	Small Bubble	Large Bubble	Small Bubble	Large Bubble
Without filter, CFU/m3	21,302 ± 3546	4322 ± 356	181,551 ± 21,453	117,111 ± 12,756
With filter, CFU/m3,	1632 ± 194	270 ± 33	7208 ± 678	3918 ± 322
% in the gas phase	7.66	6.25	3.97	3.35

. As is seen, the contribution of the “gas phase” bioaerosols is quite low, varying between 3.35 and 7.66% for various scenarios. In addition, the bubble size is not very crucial; the outcomes are much more strongly influenced by the air flow rate. Obviously, the estimate is somewhat indicative, as the separation procedure is not perfect, and some large particles could potentially penetrate the media. In contrast, some small particles could be accidentally removed. However, a simple estimate could confirm that even in the case of some large particle penetration, the outcomes would not be strongly affected. It also should be noticed that the bubble size is not very important, and the results are much more strongly influenced by the air flow rate. In particular, the concentration of the suspension selected for all experiments, as discussed above, was ~10⁷ CFU/mL. Then, the average droplet carrying 1 CFU would have a diameter of ~55 µm, so even in the case of penetration of such a large droplet, it would only contribute a single CFU. The other important issue is related to the fact that the amount of the bioaerosol transfer to the gas phase is related to the bubble travel time or, respectively, to the height of the liquid column. The corresponding study was beyond the scope of the research and might be undertaken in the future.

4. Conclusions

This study investigated the impact of several parameters on the bioaerosol release from a bacterial suspension of E. coli during bubbling regimes under controlled laboratory conditions. The influence of different process parameters, including the air flow rate, the temperature of the microbial suspension, and presence of surfactants, has been evaluated, and the results were discussed. It has been established that the effect of the air flow on the release of bioaerosol is significant and could reach an order of magnitude in terms of the concentration of suspended particles within the flow range used in the work. A rather significant difference in the bioaerosol release is also observed with an increase in the temperature of the microbial suspension, while a larger number of microbes is released from hotter liquids. In contrast, the use of surfactants has not been found to be beneficial in minimizing microbial release; the addition of the antifoaming agent caused an increased release of bacterial particles. These outcomes are important as they could be directly used in industrial settings to minimize the microbial release to the surroundings reducing the risk of transferring harmful pathogens and substances from the contaminated liquids to the air.