Long-term experience with rapid air filtration (6 to 15 air changes per hour) in a K-5 elementary school using HEPA and Do-It-Yourself (DIY) air purifiers during the COVID-19 pandemic

Introduction

In 2021, the Marche region of Italy invested in ventilation systems in schools in advance of the Delta and Omicron waves during which transmission within classrooms was tracked with testing, and analysis of classroom infection rates revealed they were over 80% lower in classrooms with 6 air changes per hour compared to schools without ventilation upgrades [16] [17] [18]. Frequent air cleaning may not stop spread of SARS-Cov-2 at near-field (e.g. talking close by), but can potentially emulate far-field equivalent of N95 with 95% reduction of particles (20x) in a well-mixed room. Using 0.6 ACH as a baseline for unventilated rooms at least 12 ACH is required for far-field protection equivalent to N95 (95%), and 11-12 ACH was measured aboard three passenger airplanes in flight [19].

Portable air cleaners and central HVAC systems are two valid ways to achieve the ACH targets necessary for infection control for Covid or many other harmful aerosols. Portable air cleaners that “recycle” the air in small to medium sized rooms may in many instances be cheaper (more cost-effective) than HVAC which consume high levels of energy and investment to transport, heat, and cool the “fresh” air from outside through existing ductwork that in turn may need to be upgraded to meet higher ACH targets demanded for infection control. While owners of some buildings can perhaps afford “fresh” air using the central HVAC to meet these higher ACH targets, “recycling” the air in each room with portable air cleaners, like plugging in lamps or fans in each room, can be more economical and scalable in terms of up front investment, energy consumption, and operating costs. A minimum 6 and preferably 12 air changes per hour is recommended by California Department of Public Health (CDPH) [5] to reduce and prevent airborne transmission of SARS-CoV-2 virus. Based on CDC recommendations relating square footage to clean air delivery rate of HEPA purifiers to prevent COVID-19 transmission, one interpretation of CDC guidance is equivalent to 5 ACH for use of HEPA purifiers in buildings [13] and 7.5 ACH in homes [14]. In general, 6 ACH requires 100 cfm per 1000 cubic feet of space, and 12 ACH is double that.

Many HEPA air cleaners (purifiers) sold in the United States are tested for clean air delivery rate (CADR) by the Association of Home Appliance Manufacturers (AHAM). Multiple research groups have similarly evaluated the effectiveness of Do-It-Yourself (DIY) air cleaners in the lab confirming their effectiveness to remove aerosolized particulate matter if used correctly [1] [2] including with filters capable of high filtration efficiency at the most penetrating particle size of 0.3 μm [3]. HEPA purifiers that deliver useful CADR on their highest fan speeds may be rendered much less effective for COVID-19 prevention when running on their lower fan speeds or auto speed, and end-users may forget to turn up their fan speeds to maximum. For COVID-19 prevention, one of the advantages of the DIY purifiers based on the box fans and well-chosen filters in [3] is they provide useful levels of CADR in a classroom without excessive noise generation even when they are used on the lowest fan speed -- eliminating uncertainties of speed settings commonly seen with HEPA purifiers. However, EPA recommends DIY air cleaners as a temporary alternative to commercial air cleaners [15], and a recent EPA study presented at the annual American Association of Aerosol Research (AAAR) in 2022 reported inconsistent usage and dissatisfaction of DIY air cleaners in homes during wildfire season in California. The most common complaint being excessive noise generated [4], likely because end-users were advised to set them at the highest speed and reduce speed if necessary. Furthermore if DIY air purifiers were to be used daily questions still remain about how long the filters will last or how often they need to be replaced.

To mitigate airborne spread of SARS-CoV-2 virus during the COVID-19 pandemic, both HEPA and DIY air purifiers were installed and used daily by teachers and staff at an elementary school in California during the academic school year from spring 2021 through fall of 2022 (ongoing) in 16 classrooms (30’x30’x30’ or 9000 cubic feet each), a library, an auditorium, a lunchroom, and in a hallway as shown in Figure 1. These included four models of HEPA purifiers (A, B, C, D) and DIY air cleaners (with Lennox and Nordic Pure filters). To address the questions about usage and filter longevity of both HEPA and DIY air purifiers, herein we report the real-world experience operating them at the school and assess the “wear and tear” of daily usage on the filters through measurements of filtration efficiency using the techniques described in [3].

• Download figure
• Open in new tab

Figure 1: Map of school and placement of HEPA and DIY air purifiers

Installation of HEPA A was conducted in Spring of 2021, HEPA B in Fall of 2021, and HEPA C and D in February, 2022. The DIY purifiers with two brands of filters operated at the low speed as illustrated in Figure 3. These DIY air cleaners were installed in or around February of 2022, with additional DIY purifiers (5” Lennox MERV 16) installed in or around October of 2022.

In Figure 2, the ACH from running air purifiers in each classroom is estimated using the CADR for each purifier from Table 1 and the volume of the classrooms (9000 cubic feet) ranging from 5.5 to 15.1 ACH, averaging 10.2 ACH across the 16 classrooms.

• Download figure
• Open in new tab

Figure 2: Estimated air changes per hour (ACH) in the 16 classrooms based on Table 1

• Download figure
• Open in new tab

Figure 3: DIY air purifiers made from Lasko fans and 5” Lennox MERV 16 (left) and 2” Nordic Pure MERV 13 (right)

Based on CADR testing results of DIY air purifiers at low speed [3] and to avoid the dissatisfaction of running them at high speed reported in [4], the teachers and staff were advised to use the DIY purifiers on the lowest speed only, and to run HEPA A on its maximum speed. The vast majority of teachers reported following those recommendations daily because the noise generated by those purifiers was tolerable in the classroom at these speeds. Whereas the vast majority of teachers also find HEPA B to be too noisy on its maximum speed and turned it down to the medium or low speed. The HEPA B are designated to be kept on 24×7, and all other purifiers are turned on/off by teachers and staff at the start and end of the school day respectively.

The DIY purifiers were made from Lasko box fans using a shroud and included an aluminum safety screen (pizza screen) in front of the fan since it is a K-5 elementary school with young children [6]. Each DIY purifier can be assembled in 8 minutes [7]. The progressive installation of air purifiers over time at the school reflects a desire to meet the minimum 6 and preferably 12 air changes per hour recommended by California Department of Public Health (CDPH) [5]. Additional DIY air purifiers are currently being installed in an effort to meet or exceed the preferred 12 ACH recommended by CDPH. Each HEPA or DIY purifier in use at the school is estimated to deliver anywhere between 100 cfm to 400 cfm depending on the speed on which it is operated at. For a 9000 cubic foot classroom in the school that is estimated to be over 6 ACH in most classrooms. At the lowest speed, the power consumption of Lasko fans was reported to be approximately 50 watts or 8 kilowatt-hours (kWh) per month if operating 8 hours per day in a school. At $0.10 to $0.25 per kWh operating the DIY purifier costs $1-$2 per month.

Methods

As described in [3], for each purifier measurement of filtration efficiency used an 7-channel optical particle counter (Temptop PMD 331) to count the ambient aerosol at input and output of the purifiers and measurement of airspeed at output using used an anemometer (Btmeter BT-100). In results below, all filtration efficiency measurements were based on recording 30 seconds of particle counts at input and output of the portable air cleaner or vent. Further research is needed, but published evidence suggests there is risk that significant quantities of SARS-CoV-2 virus as well as viable virus may be present in submicron aerosolized particles exhaled from people [9] [10] [11] [12]. In this study we compare the filtration efficiency of each filter at the 0.3 μm channel because it is known to be the most penetrating particle size [8], and in addition the 1 μm and 5 μm channels.

The method used to measure air changes per hour (ACH) in a classroom with a 7-channel optical particle counter (Temptop PMD 331) is described in [19]. The technique tracks the decay rate of ambient aerosols once the air purifiers are turned to arrive at an estimate of ACH and the “leak” of ambient aerosol particles into the classroom coming from the outside air ventilation system (doors and windows closed). The leak is in turn useful to estimate the rate of airflow from the ventilation system itself, separate from (in addition to) the air filtration provided by the purifiers within the room.

Results

Table 2 below shows the average and standard deviation of the filtration efficiencies for each air purifier at 0.3. Figures 4, 5, and 6 show the filtration efficiency for each of the purifiers in the school at 0.3 μm, 1 μm, and 5 μm respectively with x-axis representing each purifier ordered by the y-axis (efficiency). The measured efficiency for the Lennox filters is shown separately for those in use since February, 2022 (measured in or around October, 2022) and for new filters measured just after they were installed in October, 2022. Figure 7 shows the airspeed measured by an anemometer at the output of the air cleaner. The airspeed for each model of HEPA or configuration of DIY will vary based on its design or filter used, and airspeed is generally not directly comparable between different models or configurations.

• Download figure
• Open in new tab

Figure 4: Filtration efficiency at 0.3 μm of HEPA and DIY air purifiers at the school

• Download figure
• Open in new tab

Figure 5: Filtration efficiency at 1.0 μm of HEPA and DIY air purifiers at the school

• Download figure
• Open in new tab

Figure 6: Filtration efficiency at 5.0 μm of HEPA and DIY air purifiers at the school

• Download figure
• Open in new tab

Figure 7: Airspeed of HEPA and DIY air purifiers at the school

In one of the classrooms, the combined clean air delivery rate (CADR) of seven purifiers operating is estimated to be 2260 cfm from Table 1 (one each of HEPA A, B, C and four DIY purifiers with 5” Lennox MERV 16 filters). Based on this estimate, the ACH contribution from the seven purifiers would be 15 ACH (= 2260 cfm × 60 mins / hr / 9000 cubic feet).

Figure 8 shows the decay in particle counts measured at 0.3 μm in this classroom of size 9000 cubic feet (30’x30’x10’). Even though doors and windows were closed during the test, there were also four ceiling vents from the HVAC system that were supplying “fresh” air continuously into the room at an unknown airflow rate. In spite of “fresh” air from outside through the HVAC, the vents can be modeled as a “leak” of ambient aerosols into the room. Using the model in [19] based on particle decay rate, the ACH of the seven purifiers was estimated to be 18.3 ACH with a leak of 193,200 particles per liter per hour. The estimate of 15 ACH based on Table 1 is within 20% of the measured ACH of 18.3.

• Download figure
• Open in new tab

Figure 8: Filtration efficiency at 5.0 μm of HEPA and DIY air purifiers at the school

Just before the test, the 0.3 μm particle counts at the output of the HVAC vents were measured at 26,214 +/- 1,133 per liter (3 measurements), significantly lower than outside particle count measured at 59,466 per liter indicating the HVAC is likely filtering the outside “fresh” air. If the leak is mostly due from HVAC (not from closed doors or windows), the airflow rate in cubic feet per minute can be estimated by dividing the aggregate particle leak rate into the classroom (per 9000 cubic feet) by the concentration of particles per cubic foot measured at the output at HVAC vents (using appropriate conversions between liters to cubic feet and minutes to hours). This works out to be 1,100 cfm = (193,200 particles per liter per hr) X (28 liters per cubic ft) X (9000 cubic ft / 60 mins per hr) ÷ (26,214 particles per liter X 28 liters per cubic ft). In a 9000 cubic ft classroom 1,100 cfm delivers 7.3 ACH.

Discussion

All purifiers are reported by teachers and staff to be in use daily during the school day, although purifier usage was not independently tracked through the school year on a daily basis (e.g. with a power consumption meter) it is possible some may not have been used consistently every day. However, the reduced filtration efficiency observed in the Figures 4, 5, and 6 between the new DIY Lennox MERV 16 measured at time of installation in October, 2022 (92% average) compared to measurements in October, 2022 of those installed in or around February, 2022 (77% average) is a reflection of these purifiers being used regularly if not daily since the latter were installed. Similarly the reduced filtration efficiency often below 95% and as low as 80% in some cases for many of the purifiers (HEPA A-C) in Figures 4-6 is also a reflection of “wear and tear” of frequent if not daily usage.

Three to six purifiers were needed in classrooms to meet California recommended 6 to 12 ACH. Deployment of HEPA and DIY at an elementary school in California [5]. The experience at a California K-5 elementary school demonstrates the feasibility of daily operation in 16 classrooms, library, auditorium, lunchroom, and hallway. Teachers reported noise generated by DIY purifiers (made of box fans) to be tolerable / acceptable for daily classroom use when run on low speed. Filters in both DIY and HEPA purifiers held up for over six months of daily use, and in many cases over one year.

Based on the measurement of reduced filtration efficiency after approximately six months of daily usage by teachers and staff, the filters in DIY air purifiers exhibited signs of “wear and tear” and are expected to last through one school year. Whereas the filters in HEPA purifiers may last longer. The HEPA models in use exhibited a wide variation in filtration efficiency at 0.3 μm among different units in the school ranging from almost 80% to well above 95%. This suggests some “wear and tear” on some of the filters which have been in use for one year or more, yet even at the lowest levels measured these efficiencies of HEPA purifiers remain useful to remove particulates in the room in a timely manner. If this trend continues and the filtration efficiencies continue to degrade at the same rate these filters perhaps may need to be replaced in a year or two.

The new Lennox MERV 16 used in DIY purifiers exhibited a much higher filtration efficiency at 0.3 μm (93% average) than DIY purifiers using the same model filters in operation since February, 2022 (77% average). Although the filtration efficiency was not recorded at the time of their installation in February, 2022 this difference likely reflects “wear and tear” on the filters through the course of approximately half a school year. It remains to be seen if these filtration efficiencies will hold up by the end of one year (February, 2023) or need to be replaced earlier. The 2”-4” Nordic Pure MERV 13-14 remain useful for filtering particles but the 2” MERV 13 similarly exhibited reduced filtration efficiency at 0.3 μm (50% average) compared to measurements in [3] (> 60%), some of which were below 50% and may need to be replaced sooner.

Among 35 DIY air purifiers deployed in February, 2022 (Lennox and Nordic Pure) the fan on one of them stopped working, and another started making unusual noises. Both fans had to be replaced. Purifiers that might need their fan or filter replaced if their airspeed is much lower than others of the same model or configuration. Variation in airspeed can be seen among purifiers of the same model or configuration in Figure 7.

As verified in one classroom of the school (Figure 8), tracking the indoor decay rate of ambient aerosol particles in the submicron range (0.3 μm) [19] can be used to validate the ACH of air purifiers (HEPA and DIY) separately from the ACH contribution from the HVAC system. In the 9000 cubic foot classroom estimated at 2260 cfm combined from 7 portable air cleaners, the exchange rate was measured at 18 ACH from air purifiers and 7 ACH from HVAC for a combined total of effectively 25 ACH. A high rate of particulate pollution measured outside at approximately 60,000 per liter was filtered down to approximately 25,000 per liter by the HVAC (“fresh” air) as measured at the vents and injected into the classroom at the rate of 7 ACH. The air purifiers further reduced this to approximately 10,000 per liter inside the classroom. During periods of extreme wildfire pollution, this measurement illustrates why enabling the HVAC can counteract the filtering effect of the portable air cleaners if the HVAC supplying “fresh” air ends up injecting high rates of carcinogenic wildfire particles into the classroom. With or without central HVAC, portable air cleaners (HEPA and DIY) exceeded the level of air exchange (6-12 ACH) recommended by California Department of Public Health demonstrating feasibility and unit economics of meeting these targets using portable air cleaners per classroom for $200 to $650 with DIY versus $600-$12,000 with HEPA. Aside from the models in use at the school, a wide range of HEPA and DIY air purifiers with low-noise generation are available to reach the 5-6 and 11-12 ACH targets as tested in Table 2 of [19] and some representative models are shown in Table 3 below. The HEPA models are available through retail channels such as Amazon, Walmart, etc. DIY purifiers use the build procedure described in [6] [7] with parts available through retailers such as Amazon, Walmart, etc. It is important to note each room is different so any combination of air purifiers may result in different ACH than predicted from CADR, based on manufacturing variations, placement of purifiers, mixing conditions, speeds at which the purifiers are run, wear and tear, etc. The test procedure described using ambient aerosols (Figure 8) to verify ACH contributed by portable air cleaners and HVAC separately can be the basis for ACH certification or validation in classrooms and other types of rooms without generating aerosol contaminants (e.g. salt water, smoke, tracers) which may be unsafe or disallowed.