What Have We Learned About Air Conditioning & The Coronavirus — Positive Energy

Proud signatories of the Carbon Leadership Forum’s MEP 2040 Challenge

Design Around People. A Good Building Follows.

Blog

Read more about a range of building science, engineering, and architecture topics on our company blog.

What Have We Learned About Air Conditioning & The Coronavirus

by Kristof Irwin and M. Walker

There’s an unprecedented unifying force in the world today and it’s the SARS-CoV-2 pandemic. If you’ve been paying attention to the news at all lately, you’ve likely been inundated with articles, news and facts (both real and alternative) about COVID. Young or old, rich or poor, we are all in this together. The virus has intersected with everyone’s daily life in myriad, unexpected ways and continues to do so. Never before in the history of our company have we heard from so many existing clients, potential new clients, and podcast listeners telling their pandemic stories, expressing concerns about their indoor air quality, and asking what they can do to create healthier home and office environments. 

In an effort to broadly provide resources to our clientele and audience, we’ve written articles on the topics of health precautions for construction job sites and designing for healthy environments while reducing pathogen spread. We’ve released podcast episodes on the impact of ventilation and filtration on virus transmission. But now it’s time to talk about a serious elephant in the room as it pertains to coronavirus spread - air conditioning. 

Those two words appear together so commonly that we scarcely think about them. Air is a relatively simple concept, but conditioning is worth exploring. Conditioning means to condition something for a purpose. You condition leather to make shoes, you condition limestone to make Portland cement. When it comes to air, you condition it for human occupancy. Conditioning is far involved more than simply cooling, it includes humidity control, ventilation, and filtration.

Setting The Stage - The Starbucks Case

Recently, we came across an eye opening case study in South Korea that took place at the beginning of August (note the link is to a Korean site, but Google Chrome’s automatic translation tool works quite well and is, as far as we can tell, accurate). The situation presented in the Starbucks case illustrates the interdependent relationship of air conditioning systems and protective masks with the spread and prevention of the SARS-CoV-2 virus. The basics of the situation are straight forward enough - 27 people in a 2 level Starbucks in Paju, South Korea tested positive for COVID-19 after 1 unknowingly infected patient visited the store and stayed for a couple of hours. However, each of the workers on shift during this time all tested negative. 

For context, South Korea is not under stay-at-home-orders, as much of the US and other countries are, so cafes, restaurants, and stores are all open, and people can eat or drink inside. Like much of the southern United States, South Korea is also steaming hot and humid in the summers, so air conditioners are always on during this time of the year. Also like the U.S., it is common in South Korea to see minimal levels of effective filtration and ventilation in public spaces (although the mileage varies greatly from place-to-place and building-to-building).  

Two things immediately stood out in the story that will lead us to our discussion on air conditioning:

  1. The employees, who tested negative, were wearing protective KF94* masks for the duration of the infected person’s stay at the coffee shop, while the infected visitors either did not wear them or removed them at some point. 

  2. There are notable differences in the space conditioning equipment between the 1st floor, where the employees were working, and the 2nd floor, where the majority of infections occurred.

*Quick side note to clarify a term: if you’re unfamiliar with KF94 masks, or Korea Filter masks, don’t worry. Essentially, they’re a Korean made version of their American counterpart, the N95 mask, with a few minor differences in performance and testing protocols. They look similar, and they filter a nearly identical percentage of particles—95% versus 94%. See the chart below from 3M for more specifics regarding the differences between these two types of mask or check out this link to learn more about other masks and their function. 

So based on the data we have at hand in the Starbucks case, backed by the growing body of evidence suggesting masks’ effectiveness at preventing transmission, we can reasonably infer that the masks were indeed effective in protecting the employees from infection. But what exactly happened with the rest of the store? How is it that 1 infected individual was able to transmit the virus to 27 other people in just a 2 hour period? Let’s take a look at some of the highlights from the Insight article (translated from Korean, of course) and use their reporting as a launching pad to look more critically at the science behind virus spread inside buildings:

A woman in her 30s stayed at the store on the second floor [of the starbucks] for about two hours…”

“On the second floor, where six ceiling air conditioners were distributed… infection was bound to spread quickly.

Surprisingly, the four employees who worked inside Starbucks were fine. [They] went to the second floor from time to time, but they wore KF94 masks throughout the working hours.

The story was eventually picked up by Bloomberg, who reported that the incident illustrates a lot about both the effectiveness of masks and the role of air conditioning in the spread of the disease.

The Starbucks case is one of “the most important opportunities to study risk factors among a more or less controlled cohort of people,” said Arnold Bosman, director at Transmissible BV, a Netherlands-based developer of training materials for outbreak control. “This Starbucks event will be a very valuable training exercise for future generations of epidemiologists.

Indeed, this scenario is an important case study for researchers across the scientific community to examine how pollutants and pathogens can be spread in indoor environments. And as far as Positive Energy is interested in this unfortunate case study, we want to examine the action of the building systems and their contribution to poor health outcomes. Like doctors, professional engineers need to at minimum “do no harm,” although this minimum is not a sufficient standard of care given how easily it can bias expectations toward cost-only-optimized-solutions.  When we identify what doesn’t work, it informs and refines our understanding of design strategies to help keep our clients comfortable, safe and healthy indoors. 

What Does The Starbucks Case Teach Us?

The Starbucks case seems to affirm a growing body of scientific research on the effectiveness of masks at preventing transmission, but the scenario also begs our core question - how does air conditioning impact transmission? The answer is related to the reasons why masks are beneficial. Both are operating to either move or prevent the movement of air. In the case of air conditioning systems the air they move and mix is a potential vector for spreading SARS-CoV-2 around a building and dispersing it into the volume of indoor air. Masks prevent this potentially virus-mixed air from entering our lungs. Again, air is the common link. A solution of solid or liquid particles suspended in air is an air-solution, or aero-solution, now commonly referred to as an aerosol. The important aspect of particulate or liquid matter in an aerosol is that it is a solution, this means the solid or liquid does not readily fall out, it stays suspended in the air for a long time, hours to weeks. The fact that the virus can be carried via aerosolization shapes how we understand and deal with it. 

A recent NYT Opinion Column by Dr. Linsey C. Marr, an engineering professor at Virginia Tech, articulates this well: 

As we cough and sneeze, talk or just breathe, we naturally release droplets (small particles of fluid) and aerosols (smaller particles of fluid) into the air. In a peer-reviewed study published in Scientific Reports on Wednesday, researchers at the University of Nebraska Medical Center found that aerosols collected in the hospital rooms of Covid-19 patients contained the coronavirus. This confirms the results of a study from late May (not peer-reviewed) in which Covid-19 patients were found to release SARS-CoV-2 simply by exhaling — without coughing or even talking. The authors of that study said the finding implied that airborne transmission “plays a major role” in spreading the virus.

Given that the virus is airborne, it makes sense to employ our knowledge of the behavior and flow of air in indoor spaces (or better yet, use modeling tools to do so), but that is not as simple as it may seem. From a recent study on droplet behavior: 

The dynamics of virus transmission is not well understood, with one challenge being the complicated fluid and flow characteristics involved in the fate and transport of virus, including source dynamics (e.g., exhale velocity and temperature, droplet sizes, virus load, and droplet–virus correlations), ambient conditions (e.g., mean and turbulent flows, temperature, and humidity), and virus dynamics (e.g., virus viability and infectious rate) (e.g., Lindsley et al., 2015; Feng et al., 2020; Dbouk and Drikakis, 2020a; and Mittal et al., 2020). Understanding the fundamental fluid dynamics of expiratory virus-laden droplets is critical to the prediction of the transport and fate of droplets and associated potential threats of infectious disease transmission and will provide quantitative guidance for making a public health policy for disease mitigation, e.g., decisions on social distancing and face covering in various indoor and outdoor environments (Dbouk and Drikakis, 2020b;  Verma et al., 2020).

So, given these facts, just how dangerous can air conditioners really be? As you might expect, it highly depends on how well designed and installed those systems are. Air conditioners are not themselves inherently problematic, but left to the devices of traditional industry practices, they can be disastrous for human health.

We can safely assume that many buildings are not employing robust filtration or ventilation strategies, which are both known to be effective in mitigating airborne particulates on which the SARS-CoV-2 virus is carried. If you have not yet listened to our recent podcast episode on this topic with Dr. Ty Newell, PhD, P.E., it is a true education on the matter. Conditioned spaces create unique hygrothermal conditions and the behavior of pollutants in a given space is largely determined by its conditioning strategies and how well they were implemented. This is important to note primarily because most conditioned spaces have systems that are insufficient to protect human health and do no harm. 

In fact, the first COVID-19 patient in Wuhan spread it to others via an air conditioning unit even though they were more than 6 feet away. In a published study of the patient one scenario, a swab sample from the air conditioning system near the patient tested negative, indicating that the virus droplets indeed were not filtered and likely circulating around the restaurant via the air conditioner’s blower. We’re inferring here that the COVID laden particles were being circulated by, not through, the air conditioning system. 

Dr. Marr again:

Consider the case of a restaurant in Guangzhou, southern China, at the beginning of the year, in which one diner infected with SARS-CoV-2 at one table spread the virus to a total of nine people seated at their table and two other tables.

Yuguo Li, a professor of engineering at the University of Hong Kong, and colleagues analyzed video footage from the restaurant and in a preprint (not peer reviewed) published in April found no evidence of close contact between the diners.”

“Droplets can’t account for transmission in this case, at least not among the people at the tables other than the infected person’s: The droplets would have fallen to the floor before reaching those tables.”

But the three tables were in a poorly ventilated section of the restaurant, and an air conditioning unit pushed air across them. Notably, too, no staff member and none of the other diners in the restaurant — including at two tables just beyond the air conditioner’s airstream — became infected.

All evidence considered, the Starbucks case in South Korea is strikingly similar to the case of patient one in Wuhan. Air conditioned spaces with insufficient strategies employed for human health can and do cause serious health issues. 

What Could Have Prevented These Infections?

To state the obvious, staying home or utilizing a curbside pickup system would have certainly prevented this particular infection cluster, but since many people are opting to continue some degree of public life as it was before the pandemic, let’s look at the other strategies available in hopes that more buildings can “bake in” protective measures without relying on occupant behavior.

Profs. Linsey Marr (Virginia Tech), Shelly Miller (CU Boulder), Kimberly Prather (UC-San Diego), Charles Haas (Drexel University), William Bahnfleth (Penn State), Richard Corsi (Portland State), and Jose-Luis Jimenez (CU Boulder) have written a fantastic and exhaustive FAQ document with lots of really great information. We’ve simplified a few salient points for those who aren’t able to dive in to that depth just yet.

Protective Masks

Wearing protective masks is a demonstrably effective strategy as evidenced by the Starbucks employees who did not become infected. Researchers have, for quite some time, known that masks can prevent people from spreading airway germs to others. These findings have driven much of the conversation around masks during the coronavirus pandemic and have been the catalyst for further research. As cases have continued to rise across the world (and especially here in the US), experts are pointing to a growing body of evidence suggesting that masks also protect the people wearing them, lessening the severity of symptoms, or in some instances, staving off infection entirely. This a growing body of research spans disciplines of virology, epidemiology, and ecology and the results so far suggest that universal masking not only protects others from a potentially infected individual, but also protects the mask wearer. The mechanism of protection is the reduction of the “inoculum” or dose of the virus for the mask wearer, leading to more mild and asymptomatic infection manifestations. Ideas about the importance of viral dose in the development of various diseases have been studied since the 1930s and what we have learned has contributed to the development of strategies to protect us against other airborne pollutants as well. 

With regard to the SARS-CoV-2 virus, there is a notable new paper out on the effectiveness of mask wearing. Dr. Monica Gandhi, an infectious disease physician at the University of California, San Francisco wrote in a recent article:

As governments and workplaces began to recommend or mandate mask-wearing, my colleagues and I noticed an interesting trend. In places where most people wore masks, those who did get infected seemed dramatically less likely to get severely ill compared to places with less mask-wearing.

It seems people get less sick if they wear a mask.

When you wear a mask – even a cloth mask – you typically are exposed to a lower dose of the coronavirus than if you didn’t. Both recent experiments in animal models using coronavirus and nearly a hundred years of viral research show that lower viral doses usually means less severe disease.

No mask is perfect, and wearing one might not prevent you from getting infected. But it might be the difference between a case of Covid-19 that sends you to the hospital and a case so mild you don’t even realize you’re infected.

There you have it. Protective masks are a simple, relatively straightforward and inexpensive strategy to protect yourself and others from viral transmission. 

Humidity Control

The impact of humidity on human comfort and health is important to understand and important to include in mechanical system designs. Humans and viruses prefer different indoor temperatures and humidities to thrive. Keeping indoor spaces in the Goldilocks zone of 40-60% relative humidity  is an effective way to mitigate the spread of viruses like COVID. Our bodies natural defenses, our cilia and mucous tissues air impaired when the air gets too dry. Too wet, and the resultant microecology of damp buildings creates an ecosystem for a host of microbes, including fungi, bacteria and viruses impact the indoor microbiome in ways that negatively impact our health. 

There are new approaches to modeling airborne droplet behaviors that illustrate the expelled droplets that carry the SARS-CoV-2 virus are sensitive to environmental conditions, including temperature, humidity, and ambient flows. Since these droplets play a key role in viral and other pollutant spread, we should have a keen sensitivity to controlling humidity in indoor environments. Further convincing evidence suggests this modeling strategy’s accuracy as noted in another study: 

At a higher humidity, the droplets grow faster, fall to the ground earlier and can be inhaled less by healthy people. A humidity level of at least 40 percent in public buildings and local transport would therefore not only reduce the effects of COVID-19, but also of other viral diseases such as seasonal flu.

There is, of course, nuance here (this is a tricky set of topics). Take into account Stephanie H. Taylor MD M Architecture, CIC and her work in creating sufficient levels of humidity to support healthy immune function. Generally speaking, viruses thrive in dry conditions because they aerosolize and thus stay in the air longer. It’s also such that when your mucus tissues dry out, the cilia (which protect against viruses and other pollutants) don’t work like they should; the microbiome on the surfaces of your muco-cilia system don’t produce the right recipe to fight viruses. 

From Dr. Taylor’s IAQ Radio interview: 

When our mucous lining becomes thick effective,particle capture is reduced. Particle capture becomes ineffective at as little as a 6% increase in mucous viscosity. Cystic fibrosis patients experience more infections because infectious particles settle and macrophages and dendritic cells don’t secrete needed proteins.

“Low ambient humidity impairs barrier function and innate resistance against influenza infection.” Akiko Iwasaki study found that the mammalian immune system is impaired at 10%-20% RH

The comfort zone is 40%-60%. Staying within the comfort zone is the goal. Staying within the comfort zone reduces infectivity.

But in the case of SARS-CoV-2, and to add even more complexity to the topic of humidity and viral spread, its’ worth noting that the virus in question seems to behave a bit differently than its counterparts. This was recently described by Lew Harriman on an episode of IAQ Radio, in which he discussed the new ASHRAE document “Damp Buildings, Human Health and HVAC Design”. Harriman reminded listeners that, while Dr. Taylor’s findings are true, the SARS-CoV-2 virus is actually able to remain in the air for hours at a time at 50%RH. He also noted that the level of humidity control really depends on the building typology; grocery stores have different usage patterns than a home, for example. 

So while studies of other viron can and do provide meaningful insights to reduce  transmission in general terms, it is important to understand the specifics of the viral behavior in question before recommending or adopting a strategy. And, as with all science, the research body grows and our understanding will change. Remember that nothing is final, but in the applied science profession, we do our best to recommend solutions that help people based on the latest peer-reviewed research. 

Ventilation

We’ve mentioned this strategy previously in this article and other articles we’ve written and podcast episodes we’ve recorded and we cannot overstate the importance of sufficient ventilation. Researchers, such as Jeffrey Siegel, are taking this message mainstream. The NPR segment, Marketplace, recently aired an interview with Siegel about the state of ventilation in buildings and how it’s negatively impacting virus transmission indoors. 

From that interview: 

“Molly Wood: It is my understanding that a lot of existing [heating, ventilation and air conditioning] systems, particularly in commercial buildings, do recirculate a lot of air in order to keep either cooled or heated air in the system. That was for efficiency purposes?

Siegel: That’s absolutely correct.

Wood: So in hindsight, was that a terrible mistake?

Siegel: No, absolutely not. We have a climate crisis. Energy use associated with buildings is a very big part of our energy footprint. Conditioning that air is one of the big users of energy within a building. So it’s important that we do it in an energy-efficient manner. I think that the bigger problem is that we have to be much more cognizant of how we’re managing ventilation. I think COVID-19 adds some variables to how we might manage ventilation. But in general, I think that we have the tools to do it. It might take some investment and so on, [but] we just have to be a little bit more proactive and engaged in how we manage the ventilation in our systems.”

This begs a common question we get from practitioners across the AEC industry - if we’re going through all the effort to design and build energy efficient buildings, but we’re also being told to ventilate, how do we reconcile those outcomes? And the truth is that it takes some careful consideration and calculation, but that is exactly the role of a good mechanical designer who has a sympathetic understanding of enclosures, energy performance, and human health. With the right framework, communication flows, and process, multiple simultaneous positive outcomes (energy efficiency, healthy air, budget sensible approach) are achievable. 

If you were wondering about the ventilation of the building in the Starbucks case, Starbucks was specifically asked about their ventilation, and noted that "... the windows were opened for more than 10 minutes twice a day to ventilate," but that most of the windows were fixed glass and the only operable windows opened a narrow width of 30cm. There was no functional or known mechanical ventilation strategy. 

When we spoke with Ty Newell recently about the role of ventilation in virus prevention, he walked us through a graph (see below) that was presented in a webinar he’d given (based on his research) in early August that may surprise you. Providing sufficient fresh air in indoor spaces is a clearly effective strategy in preventing virus transmission. 

From Dr. Newell’s recent paper, Killing Ourselves With Comfort

Reduction of disease transmission within buildings and homes requires increased fresh air flow rates (at least doubling to 40cfm per person) and improved air filtration (to at least MERV11 filtration). Carbon dioxide concentration monitoring of every indoor building space is the key to reducing indoor virus transmission rates. Carbon dioxide concentration is a direct measure of human respiration rates, and therefore, virus concentration in the indoor environment. Maintaining indoor carbon dioxide concentrations below 800ppm, equivalent to doubling today’s inadequate, odor-based ventilation rates, will reduce disease transmission rates below the limit required for decay of Covid-19 transmission.

But of course, like all subjects, there is complexity to consider in some situations. The unhealthy air from raging fires in California can actually make people more susceptible to COVID-19 as their lungs and immune systems can become overtaxed with the presence of toxic particulates via smoke inhalation. So robust filtration comes into focus as a crucial strategy for good indoor air quality. 

Filtration

As we have pointed out in previous articles, ASHRAE suggests using filters with a minimum MERV-13 rating. Condensing significantly, MERV ratings are based on a filter’s performance/ability to filter out particles between 0.3 and 10 microns. SARS-CoV-2 can be found in respiratory droplets or attached to other pollutants in this size range; the higher the MERV number, the higher the probability that the filter will remove these droplets. However, the solution to all problems is not to install a higher MERV rated filter to a building’s central air conditioning system, filters are part of a system and as such the parameters of the rest of the system can aid or impair a filter’s ability to capture pollutants (which are substances that are harmful to human health). The filtration system can’t leak air or let air bypass the filter and find another path to the conditioned space. Filtration efficacy is also dependent on the type of filtration media, it’s electrostatic properties and the velocity (speed and direction) of the particles as the approach the filter. 

An interesting quirk of the physics of filtration is the very smallest particles are actually easier to filter out than the 0.3 micron ones. The smallest particles get pushed toward filter fibers because of their collisions with gas molecules in the air. We recorded a fascinating, but relatively slow podcast episode on the wild world of filtration some years ago that is worth your consideration. Even Vox is getting in on the air quality conversation, with a recent article about the effectiveness of air filtration and virus transmission prevention. These ideas are not only on the radar of scientists anymore, but major media outlets. 

Remember that simply replacing a non-HEPA with a HEPA filter in existing equipment may worsen the problem. Make sure your system can accommodate the air flow needs of a HEPA filter. If your system can’t, you can explore a more decentralized approach through portable room air cleaners instead. Take a look at The Wirecutter’s recent review of portable room air cleaners for a pretty comprehensive list of consumer grade pieces of equipment you can buy online today.

In summary, we were already in the midst of a revolution of understanding in the field of IAQ when the SARS-CoV-2 virus abruptly entered our lives and brought the field more sharply into focus. As is evidenced by the Starbucks case, studying the impact of HVAC systems on human health, especially during a pandemic, is crucial to protect us against future outbreaks. With the data gathered from the diligent research currently taking place, we will continue to understand a more complete picture of how we can use indoor air quality as a public health tool that’s “baked in” to our society’s buildings. We have a lot of work to do, a lot to learn and understand, but we have the tools, the data and the motivation like never before. Pandemics don’t just work themselves out - they end when smart people take good science, communicate it effectively to the public, and we work together to take care of one another.

On The Horizon - Emergent Knowledge

The following are some examples of topics in the emerging research field of indoor air quality. 

  1. Better sensors and analytic tools - NGS next generation sequencing equipment.

  2. New data streams (IAQ data) - PTR-ToF-MS (proton transfer reaction, time of flight, mass spectrometers).

  3. Rapid IT development - we can handle big data sets and find the needles in the haystacks.

  4. Goal to personalize Healthcare - human genome unlocks microbial genomes as well.

    1. Metagenomics is the study of genetic material recovered directly from environmental samples. The broad field may also be referred to as environmental genomics, ecogenomics or community genomics.

    2. Epigenetics focuses on processes that regulate how and when certain genes are turned on and turned off, while epigenomics pertains to analysis of epigenetic changes across many genes in a cell or entire organism. ... The epigenome can mark DNA in two ways, both of which play a role in turning genes off or on.

    3. Metabolomics is the large-scale study of small molecules , commonly known as metabolites, within cells, biofluids, tissues or organisms. Collectively, these small molecules and their interactions within a biological system are known as the metabolome.

    4. Proteomics is the large-scale study of proteomes. A proteome is a set of proteins produced in an organism, system, or biological context.

    5. Glycomics is the comprehensive study of glycomes (the entire complement of sugars, whether free or present in more complex molecules of an organism), including genetic, physiologic, pathologic, and other aspects.

Positive Energy5 Comments