InterPore2025

Application of Packed Bed Reactors and Microfluidic Devices to Simulate Multiscale Bacterial Transport Through a Contaminated Aquifer

21 May 2025, 14:50

15m

Oral Presentation (MS05) Microbial Dynamics in Porous Media: Advances in Biofilms, Biogeochemistry, and Biotechnology MS05

Speaker

James Marquis (Montana State University)

Description

Subsurface microbial communities play critical roles in the attenuation of anthropogenic contaminants, as well as global biogeochemical cycling [1]. It has been established that physiological state (whether the organism is sediment-attached or planktonic) may drastically affect metabolic activity levels and rate of bio-degradation [2]. However, due to inherent sampling challenges, accurate prediction of distribution and partitioning of microbial communities within the subsurface remains largely unresolved. Therefore, to determine the empirical relationships between individual microbial cells, growth, attachment, and detachment in a contaminated aquifer requires a laboratory-based approach guided by field observations.
The overarching goal of this work is to disentangle these relationships in the context of the highly contaminated Bear Creek Aquifer, located in Oak Ridge Tennessee. To accomplish this, two reactor systems that mimic the hydrology at different scales were used to simulate subsurface transport conditions and study the transport of the Gram-negative field isolate Stenotrophomonas GW821-FHT01H02 (H02)- a ubiquitous bacterium that is observed in both groundwater and sediments. Packed Bed Reactors (PBRs) were used to simulate the mesoscale; a primary advantage of these systems is the ability to accurately recapitulate much of the spatial and structural heterogeneity of the field. PBRs were packed with sand representing the approximate particle size distribution (75 to 300 µm, x̄ = 150 µm) and porosity (ϕ = 0.42) of sediments from the field site. PBRs were inoculated with a pulse of H02 and operated at two flow rates representing the upper and lower bounds of observed seepage velocities in the field. Results suggested that the partitioning of H02 is highly sensitive to changes in flow. Under non-growth static conditions, 94 ± 4% of H02 cells, regardless of inoculation density, bound to sand particles in the tested size range. This is compared to 55 ± 3% under the low flow condition and 20 ± 8% under the high flow condition in the PBRs. Additionally, breakthrough curves of both H02 and inert tracer bromide (Br-) were compared at both flow rates. Under both conditions, there was no appreciable difference in time to peak between H02 and Br-. However, observational challenges associated with mesoscale reactors made it difficult to distinguish between the effects of microbial growth and attachment kinetics throughout these experiments.
To observe the attachment and detachment processes in real-time while minimizing the effects of microbial growth, we have developed complementary silica oxide microfluidic devices. These devices were intentionally designed to simulate the porosity, mean particle size, and surface properties of the PBRs. Similarly, we have successfully inserted a fluorescent transposon into the genome of H02 to enable improved visualization and quantification of cells over short-time frames without the use of fluorescent dyes. Initial results indicate that attachment rates of H02 are highly sensitive to changes in seepage velocity and indifferent to changes in bacterial cell concentration. Furthermore, information regarding attachment and detachment rates gleaned from microscale experiments may improve our ability to predict transport times at larger scales where it is challenging to distinguish between the effects of growth and attachment kinetics.