Miniaturizing Wet Scrubbers for Aerosolized Droplet Capture

Abstract

Aerosols dispersed and transmitted through the air (e.g., particulate matter pollution and bioaerosols) are ubiquitous and one of the leading causes of adverse health effects and disease transmission. A variety of sampling methods (e.g., filters, cyclones, and impactors) have been developed to assess personal exposures. However, a gap still remains in the accessibility and ease-of-use of these technologies for people without experience or training in collecting airborne samples. Additionally, wet scrubbers (large non-portable industrial systems) utilize liquid sprays to remove aerosols from the air; the goal is to "scrub" (i.e., clean) the exhaust of industrial smokestacks, not collect the aerosols for analysis. Inspired by wet scrubbers, we developed a device fundamentally different from existing portable air samplers by using aerosolized microdroplets to capture aerosols in personal spaces (e.g., homes, offices, and schools). Our aerosol-sampling device is the size of a small teapot, can be operated without specialized training, and features a winding flow path in a supersaturated relative humidity environment, enabling droplet growth. The integrated open mesofluidic channels shuttle coalesced droplets to a collection chamber for subsequent sample analysis. Here, we present the experimental demonstration of aerosol capture in water droplets. An iterative study optimized the non-linear flow manipulating baffles and enabled an 83% retention of the aerosolized microdroplets in the confined volume of our device. As a proof-of-concept for aerosol capture into a liquid medium, 0.5-3 μm model particles were used to evaluate aerosol capture efficiency. Finally, we demonstrate that the device can capture and keep a bioaerosol (bacteriophage MS2) viable for downstream analysis.

Figure 1.

Schematic representation of our battery-powered air-sampling device (small black box to the left of the computer) collecting aerosols in a home environment to provide information on the landscape of aerosols in the room. The concept is that our device can be placed in any environment (e.g., homes, schools, hospitals, playgrounds, farms) and then sent back to a lab for analysis. See Figure 2 for a photograph of the battery-operated, compact device.

Figure 2.

Portable droplet-based air-sampling device. (A) Photograph of device. (B) Schematic of capture region workflow. An ultrasonic atomizer generates 4 μm droplets (top), and horizontal (middle) and vertical (bottom) open mesofluidic channels collect coalesced droplets and aerosols. Aerosols enter through the fan, microdroplets intercept the aerosols, and six angled baffles guide the airflow to increase capture of aerosols. Front wall of the device has been removed for visualization of the device.

Figure 3.

Impinging jet, stagnation region, and horizontal mesofluidic channels. (Ai) Basic schematic representation of an impinging jet (left) and the stagnation region in the device (right). (Aii) Droplets accumulate at a stagnation region on the baffle and migrate down the edge of the baffle, where it travels into the horizontal and vertical channels. Images were taken with a device that did not have a back wall, and water was colored dark blue to enable better visualization of the droplet. Video of droplet formation at the stagnation region is available in Supplementary Video 1.

Figure 4.

Schematic (top) and photo (bottom) progression of filling the mesofluidic channel at the edge of a baffle. Video was taken using a clear device to enable better visualization of the channel and is available in Supplementary Video 2.

Figure 5.

Effect of baffle quantity and angle on airflow. Experimental photos showing droplets in device and mean airflow velocity maps of a 2-D cross section of the device with (A) 0, 2, 4, and 6 baffles (all at a 30° angle) and (B) 0°, 15°, and 30° baffle angles (Supplementary Video 3). Images were taken from different timepoints and should not be compared with one another. Colors represent velocity and indicate the fluid flow (airflow) modeled. White lines and arrows represent direction of flow. Devices used in photos do not have front or back walls to enable visualization of the droplets.

Figure 6.

Effects of baffle quantity and angle on microdroplet retention. Microdroplet retention trends were determined by weighing liquid retained in the device before and after running the ultrasonic atomizer (see Experimental section). (A) The six-baffle design has the highest microdroplet retention (81.6% ± 1.6%) and was used for further device iteration studies. (B) The 15° baffle angle design retained a higher percentage of microdroplets (83.5% ± 1.1%) than a 0° baffle angle (73.8% ± 3.6%). No significant difference (ns) was observed with an additional 15° baffle angle (30°; 81.6% ± 1.6%). Bar graph represents the mean ± SD of n=3 experimental tests. One-way ANOVA with post-hoc Tukey’s multiple comparison test; * p<0.05; ** p<0.005; *** p<0.001. Devices tested in (A) all had a 30° baffle angle.

Figure 7.

Capture efficiency of our portable air-sampling device (n=3 devices) for 0.5, 0.75, 1.0, 2.0, and 3.0 μm polystyrene particles. The highest aerosol capture efficiency observed was 17.5% with 0.5 μm particles and the lowest capture efficiency observed was 4.5% with 2.0 μm particles. Bar graphs represent mean ± SD of n=3 independent chamber runs for each device. Symbol shape indicates different device A, B, or C; symbol color indicates independent chamber runs.

Figure 8.

Bacteriophage MS2 captured in our air-sampling device remains viable. A plaque assay was performed using the liquid samples obtained from our device. A control liquid sample obtained from our device (no MS2 aerosolized in chamber) showed no plaques (left) when cultured with E. coli, demonstrating MS2 absence from the sample. In contrast, when MS2 was aerosolized in the chamber, plaques formed (dark grey circles; right) as a result of MS2 infecting E. coli, indicating MS2 presence in the sample. Representative images of n=2 independent chamber runs with one device.