Total Concentrations of Virus and Bacteria in Indoor and Outdoor Air

Appropriate song to play while reading this post: Mat Kearney – Breathe In, Breathe Out

This paper came out last month, and I thought it would be nice to briefly mention it here, even though many other papers have looked at the concentrations of airborne bacteria and viruses as well.

In this study, done by Aaron Prussin et al. from the Department of Civil and Environmental Engineering at Virginia Tech, air samples were collected in triplicate at 9 locations in Blacksburg, VA. The sample sites were: a classroom, a daycare center, a dining facility, a health center, three single-family houses, an office, and an outdoors at an unspecified university campus.

For each sample, about 100 liter of air was pumped through a 0.2 μm pore size filter, and nucleic acids on the filters were stained with a fluorescent dye and counted. Pinprick sized particles (between 0.02 and 0.50 μm) were counted as virus-like particles (VLPs), while the larger signals (between 0.50 and 5.00 μm) were counted as bacteria-like particles (BLPs).

Both BLP as well as VLP concentrations were found to be between 105 and 106 per cubic meter, and not significantly different between the indoors and outdoors locations, although there was a trend for higher concentrations outdoors. Notably, the lowest concentration of both particle types was in the health center, and the highest concentration indoors was found in the classroom and homes. Overall, the differences between the locations were very small.

The authors state that not many other studies have looked at virus concentrations in air. Unfortunately, filtering through a 0.2 μm filter is not the best choice to catch viruses (Update: see below). Many viruses will be able to pass that filter, and recent work from the Banfield lab has shown that even some very small bacteria will pass through these filters (Luef et al., Diverse uncultivated ultra-small bacterial cells in groundwater, Nature Communications 6: 6372, 2015). So it is likely that the actual virus concentration both indoors and outdoors is much higher.

Total Concentrations of Virus and Bacteria in Indoor and Outdoor Air – Aaron J. Prussin, II, Ellen B. Garcia, and Linsey C. Marr – Environmental Science & Technology Letters 2015, 2, 84−88. Work founded by the Alfred P. Sloan Foundation, the National Science Foundation, and the National Institutes of Health.

Update: I stand corrected. Filtering air and water are two very different processes, and in the conditions used in the paper, filtering air over a 0.2 μm filter will actually retain most of the viruses. For a detailed explanation, see the comments section below.


Virus and Bacteria in Indoor and Outdoor Air, DOI: 10.1021/acs.estlett.5b00050, or 7-legged spiders invading a cute house?
Virus and Bacteria in Indoor and Outdoor Air, DOI: 10.1021/acs.estlett.5b00050

8 thoughts on “Total Concentrations of Virus and Bacteria in Indoor and Outdoor Air

  1. I haven’t read the study yet… but I have to admit I am very puzzled by the pore size of the filters as you report them. I worked extensively with viruses during my PhD we certainly would expect many of them to come through a .2um pore.

    1. Hi – I wanted to pass along some information I once received from William Lindsley at the National Institute of Occupational Safety and Health (NIOSH). His explanation of “The meaning of filter pore size in filter specifications” is so clear, I’ll just quote him here.
      “The term “filter pore size” can be very confusing, because it doesn’t actually refer to the size of the pores in most filters, nor does it indicate the size of the particles that the filter will retain. A much better term would be something like “equivalent pore size”, because the pore size really refers to an equivalent ideal sheet with uniform pores.
      Aerosol filters are able to collect much smaller particles than would be suggested by the size of the passageways through them. This is because small particles are not collected primarily by sieving (i.e., a particle is too big to get through the openings, like a sieve used for sand). Rather, they collect on the filter surfaces by impaction, sedimentation and diffusion. A 3 µm PTFE filter, for example, has been shown to retain >96% of aerosolized MS2 virions, which are <80 nm in diameter (NC Burton et al. 2007, The Annals of Occupational Hygiene 51(2): 143-51). An extensive discussion of aerosol filter theory can be found in Aerosol Technology: Properties, Behavior and Measurement of Airborne Particles by William C. Hinds, while collection efficiency data for many filter types are given in Aerosol Measurement: Principles, Techniques and Applications by Paul A. Baron and Klaus Willeke.
      Perhaps some engineers can jump in here, but from what I have learned, the main mechanism that these vacuum filters work is through impaction, which mean that even particles smaller than 0.2um would get caught, at some collection efficiency.

      1. Thanks, Rachel, for this helpful explanation! It is good to learn that the pore size is not exact, and that the filter does not really work like a sieve. That helps to explain why smaller particles such as viruses can be retained on the filter. Having said that, there is also evidence that the filters do make a difference, and that they at least enrich for bigger particles, while letting smaller particles through. The Leuf paper I mentioned above (from the Banfield lab) shows a significant difference between the microbial communities retained on the 0.2 and 0.1 μm pore size filters, resp. The bacterial cells that went through the 0.2 μm filters were indeed found very small. Their average size, as measured by electron microscopy was estimated to be around 200 x 250 x 300 nm. So if these tiny bacteria can make it through 200 nm filters, it would be plausible that a significant portion of viruses, which are even smaller (size range of most viruses is 20-300 nm, according to Wikipedia), can pass. It would be hard to predict the percentage of VLPs that can pass, but I would not be surprised it if was the majority.

        In addition, and even more importantly, people who want to look at the viral population in an sample, will often filter their sample over an 0.2 μm filter to get rid of (most) bacteria, and to enrich for viruses. See e.g. this paper from a former co-worker: David T. Pride et al, Evidence of a robust resident bacteriophage population revealed through analysis of the human salivary virome. So I agree that some viruses will be retained on the filter, but there must still be many that can pass.

        But maybe there is a difference in filtering watery samples vs. air samples. I can imagine that the flow strength could make a difference as well.

        1. Elisabeth and David (and others who may be wondering),

          Don’t confuse filtering particulate matter (biological or otherwise) in water and in air. The mechanisms are fundamentally different in ways that matter. Don’t assume that experiments done in water are at all predictive of what happens in air.

          When using fibrous or porous filters to capture airborne particles, the efficiency as a function of particle size tends to be “U-shaped.” The minimum efficiency is typically in the size range 0.1-0.5 µm diameter. Smaller particles diffuse (undergo Brownian motion) across flow streamlines and when they hit a solid component of the filter they stick. The smaller the particle, the higher the efficiency when diffusion dominates. Larger particles drift across fluid streamlines by impaction, or else they may be intercepted by fibers/solid filter materials because of their finite size. For these larger particles, the bigger they are the more efficiently they are collected.

          Membrane filters with a 0.2 µm pore size can be expected to capture essentially all particles when used for air sampling. At the minimum efficiency particle size, in the 0.1-0.5 µm diameter range, the expected capture efficiency is > 99%. The same idea would apply for “high efficiency particle arrestance” (HEPA) filters for air cleaning. The most penetrating particle size for these filters is typically at a diameter of ~ 0.3 µm.

          Similar U-shaped behavior is exhibited for other processes that are important in indoor (and outdoor) bioaerosol dynamics. Deposition in the respiratory tract with particles are inhaled is more efficient for the very small and for larger particles than it is for particles in the 0.1-0.5 µm size range. Particles in the 0.1-0.5 µm diameter range deposit more slowly onto room surfaces than either smaller or larger particles.

          When dealing with small particles in air (say, smaller than about 10 µm in diameter), it is generally correct (although not strictly always) to assume that if the particle strikes a surface, it will adhere. The critical determinant for all manner of removal processes is the transport of particles from bulk air flow to a point where contact is made with some collecting surface. The same physics controls the outcome whether we are considering airborne fate, filtration efficiency (for sampling or for control purposes), or deposition in the respiratory tract on inhalation.

          – Bill

    2. The question about filter collection efficiency came up in the reviews of our manuscript, and I think it arises because filtration in water is very different from filtration in air. In water, filtration works by straining, but in air, the dominant collection mechanisms are impaction, interception, and diffusion, and the latter mechanism actually favors collection of the very smallest particles. There have been multiple studies showing that membrane filters of similar or even larger pore size than we used (>= 0.2 micrometers) have >99% collection efficiency in air for virus-sized particles (Burton et al., 2007, Annals of Occupational Hygiene; Liu and Lee, 1976, Environmental Science and Technology; John and Reischl, 1967, Atmospheric Environment).

      1. Thanks, Linsey and Bill, both for your helpful answers. I learned something new, and I stand corrected. I had never realized that filtering water vs air is so different, and had always assumed that it was a more or less sieving-type of selection for both sample types. Thanks for stepping in and explaining!

      2. To echo the comment by Elisabeth… I also had no idea how different the mechanics of filtration were between air and water. This was an extremely educational discussion! Thanks Bill, Linsey, and Rachel!

  2. I was quit happy to read this article published by Aaron J. Prussin et al.,.He has tried to collect the particle (Viruses and bacteria) many not suitable for complete study on viral particle. But the authors done a work to prove that virus also one of the major bio aerosol in indoor and outdoor. I will appreciate this work and the author could add more detailed work o the same objective.

Leave a Reply

Elisabeth Bik

After receiving my PhD at Utrecht University in The Netherlands, I worked at the Dutch National Institute for Health and the St. Antonius Hospital in Nieuwegein. In 2001, I joined the Department of Microbiology and Immunology at Stanford, where I have worked on the characterization of the human microbiome in thousands of oral, gastric, and intestinal samples. I currently study the microbiome of marine mammals. When I am not in the lab, I can be found working on my blog Microbiome Digest , an almost daily compilation of scientific papers in the rapidly growing microbiome field, or on Twitter at @MicrobiomDigest.