Viruses have the potential to be transmitted or to become transmissible through air among animals or between animals and people, or they have potential to develop transmissibility, posing real or potential risks to swine, poultry, and veterinary workers. Animals in agricultural facilities generate virus-containing particles from their respiratory tracts or from their fecal matter. Many of these particles are small enough to be transported substantial distances. Few measurements have been made of the airborne concentrations, sizes, and infectivity of these virus-containing particles. Particle size is especially important because it helps to determine how far virus-containing particles can travel through air, where virus-containing particles deposit in the human respiratory tract, and technologies that can remove the particles from air. The objective of this research is to identify or develop a high-volume, field-portable, size-differentiating viral aerosol sampler and use it to measure worker exposures to live airborne influenza viruses in animal agriculture facilities.

The first step in this research was to assemble a wide range of existing samplers that collect airborne particles by a variety of principles, and to test the samplers side-by-side in an isolation room using artificially generated influenza virus aerosols. The aim for these tests was to determine the types of samplers that collect the greatest quantity and measure the highest airborne concentrations of viral RNA and live virus. Samples were analyzed to determine quantities of live virus, using isolation techniques, and total virus, using RT-PCR. Findings indicated that higher virus titers and more RNA copies are recovered from high flow rate samplers than from low flow rate samplers, likely because high flow samplers consolidate their samples more than low flow samplers. On the other hand, the highest airborne virus concentrations were measured by lower flow rate samplers, suggesting that sample consolidation may cause greater inactivation of virus and damage to viral RNA. Additionally, results suggest that impingers may keep viruses live more effectively than other types of samplers. These results suggest to us that a two-sampler strategy may have benefits during outbreak investigations. One sampler might be a high flow, non-sizing impingement sampler for detection of virus; the second sampler might be a lower flow, size-separating impingement sampler for concentration measurements.

The second step of the research, which has been the focus of the past year’s activities, has been to design a novel size-separating impingement sampler. Our approach is to develop a multi-stage virtual impactor system that concentrates particles in different size intervals and collects them in impingement samplers. Inertia is the most practical way to separate particles for analysis by size. A virtual impactor consists of an acceleration inlet nozzle and two outlets: a collection probe which draws away a minor portion of the incoming flow – usually about 10% — and a bypass outlet which draws away the remaining major portion of the incoming flow, which turns 90° from its original direction. Airborne particles are accelerated through the inlet nozzle with the incoming air and directed towards the collection probe. Larger particles with enough inertia are separated in the probe, concentrated, and carried away with the minor flow while smaller particles follow the turning air and are carried away with the major flow.
The sampler that we are designing will contain a series of stages with progressively smaller nozzles placed in series to process the same aerosol flow, with the particles separated by size into several samples that can be collected and analyzed individually. We plan to separate particles into at least four aerodynamic size intervals (>10 µm, 3-10 µm, 1-3 µm, and <1 µm) using a series of virtual impactors. We are still determining if we will include a virtual impaction stage that separates particles 0.3-1 µm, in which case the last interval would be <0.3 µm. Our plan is to collect particles using aerosol impingers such as an AGI-30 or SKC Biosampler. The final stage will be collected using a gelatin filter. The sampler is currently planned to have an incoming flow of 150 L/min.

The multi-stage virtual impaction sampler is being designed using Ansys computational fluid dynamics (CFD) modeling software. The steps needed to design the size-separating impactor sections are to draw the geometry of each section in three dimensions, lay out a three-dimensional mesh within the geometries, model the airflow throughout the mesh, and superimpose particle motion into the airflow. Design parameters will be adjusted to achieve the desired size separation with aerodynamic diameters of 10 μm, 3 μm, 1 μm, and, if we decide to use five stages, 0.3 μm. Initial values, taken from the scientific literature, indicate that 10 μm particle separation will be achieved with 2 nozzles approximately 12 mm in diameter, 3 μm separation with 12 nozzles of 2.9 mm in diameter, 1 μm separation with 36 nozzles of 0.97 mm in diameter, and, if we decide to use five stages, 0.3 μm particle separation with 36 nozzles that are 0.46 mm in diameter. Subsequently, a sampler will be fabricated according to the design developed with CFD modeling by the University of Minnesota College of Science and Engineering Machine Shop. The performance of this novel sampler will be verified in laboratory tests and compared to existing samplers in laboratory and field tests.