Biosurfactants Change the Thinning of Contaminated Bubbles at Bacteria-Laden Water Interfaces

Bubbles reside at the water surface before bursting, emitting droplets that can contain chemicals and pathogens linked to disease and contamination. We discover that bacterial secretions enhance the lifetime of bubbles. We also reveal and elucidate two distinct regimes of thinning for such contaminated bubbles. Initially, marginal regeneration governs their thinning rate, similarly to clean water bubbles. However, due to their enhanced lifetime, it is eventually evaporation that governs their thinning, thus also dramatically decreasing their thickness at burst. We derive and experimentally validate the expression for the critical timescale at which the transition between the two regimes occurs. The shift in thinning law makes the droplets produced by contaminated bubbles smaller, faster, and more numerous than those produced by clean bubbles. Our findings suggest that microorganisms can manipulate the aging physics of surface bubbles to enhance their own water-to-air dispersal.

Figure 1

Bubbles bursting after (a) 4 s and (b) 55 s at the surface of water with 0.61 and 0.21 ms between frames, respectively. The burst of the older bubble produces many more and smaller droplets. Scale bars are 1 mm.

Figure 2

Fifty second-old bubble observed using monochromatic light (sodium lamp, 589 nm) at the surface of water contaminated with E. coli: the bright spots are only observed when bacteria are present and reveal their location and persistence on the cap. Scale bar is 1 mm.

Figure 3

(a) Thickness evolution of bubbles with $R=5.5\pm 0.2\mathrm{mm}$ at ambient humidity in fresh deionized water, M9 medium [36, 37] diluted 20 times, and stagnant water. The diluted M9 culture medium does not significantly affect the lifetime nor thinning evolution of bubbles. (b) Thickness evolution of bubbles with $R=5.4\pm 0.2\mathrm{mm}$ derived from solutions of E. coli at $c_1=3\times {10}^7$ and $c_2=6\times {10}^6$ cells ${\mathrm{mL}}^{-1}$, with or without filtration of the bacteria. Solutions of filtered bacteria are obtained using a $0.22\mu \mathrm{m}$ pore-size filter. The ambient humidity is $H=22\pm 5\%$, and $H>95\%$ for saturated air conditions. All experiments in this study are conducted at ambient temperature $T=24\pm 1°\mathrm{C}$. In this figure and the next, filled and open symbols show natural and manual bursts, respectively: due to their more deterministic lifetimes, bubbles from concentrated solutions of surfactants or bacterial secretions must be pierced manually with a needle to obtain their thickness at an early time. (c) Probability distribution function of bubble lifetimes $t_b$ in ambient air, with deionized water. Bubbles in pure water scarcely live longer than 10 s, while bubbles in filtered or unfiltered solutions of bacteria can live up to 90 s. (d) Similar results are obtained with tap water, for which bubbles have the same thickness evolution as with deionized water [30]. Each curve in (c) and (d) represents one dataset consisting of an average of 1600 bubbles per curve.

Figure 4

(a) and (b) Thickness evolution of bubbles in solutions of SDS with $R=4.8\pm 0.1\mathrm{mm}$ in (a) ambient air ($H=47,32,40,53\pm 2\%$, 32, 40, $53\pm 2\%$ for increasing SDS concentration) and (b) in saturated air ($H>95\%$). Solid and dotted lines are solutions of Eq. (2) for (a) $H=50\pm 30\%$ and (b) $H=98\pm 1\%$. The dashed lines corresponds to fully saturated air, $H=100\%$, when the thickness evolution is solely governed by drainage ($h\sim t^{-2/3}$). (c) Thickness evolution of bubbles with $R=5.2\pm 0.4\mathrm{mm}$ in solutions of ${\mathrm{C}}_{14}\mathrm{TAB}$ at 5%, 50%, and 10 cmc in ambient air ($H=26\pm 2\%$) and at 50% of the cmc in saturated air. The plain and dotted line are the prediction for $H=26$ and 96% from Eq. (2).