Air filtration, as we all know, is the process of separating dispersed particles from air by means of porous media. Filter fibers are expected to separate and retain particles on or within the filter medium. An example of the use of air filtration is as a preventive measure to protect the respiratory systems of human occupants indoors as well as the HVAC equipment. This can be done through mechanical filtration to reduce contaminant concentrations, which vary depending upon the nature of the application and location. The success of any filtration system depends upon its ability to capture the right contaminants in the right quantity whilst presenting the lowest resistance to the airflow.
It is not only dust
Airborne particles are all around us — they come in many different forms, such as grit, dust, smoke, fumes, or mist, and we should not forget smog and fog. The types of dust are numerous: mineral dusts, such as those containing free crystalline silica (e.g. as quartz); coal and cement dusts; metallic dusts, such as lead, cadmium, nickel, and beryllium dusts; other chemical dusts, e.g., many bulk chemicals and pesticides; organic and vegetable dusts, such as flour, wood, cotton, tea dusts, and pollens; and biohazard dusts, such as viable particles, molds, and spores, as shown in Figure 1.
The air we breathe is full of microscopic particles, which can be health hazardous to and are thus considered as a specific type of air pollution. The size of these particles is in the order of several nanometers to several micrometers. Epidemiological studies have shown beyond doubt an association between increased urban air pollution and adverse health effects on susceptible sectors of the population, particularly the elderly who may have pre-existing respiratory or cardiovascular diseases [1,2]. Urban air contains particles whose size may be classed as coarse to ultrafine (<0.1 μm in diameter). Ultrafine particles contribute very little to the total mass in a sample of air, but they exist in very high numbers, which in episodic events can reach several hundred thousand/cm3 in urban air. Therefore, characterizing atmospheric pollutants is essential to the understanding of air filter performance and enhancing IAQ.
Not all dusts arise from human activity. High winds in desert regions and volcanic events are both natural phenomena that cause high airborne dust concentrations. So, weather conditions, the natural environment, and human activities can cause windblown, construction, or fugitive dust, which contributes to air pollution. High winds can raise large amounts of dust from areas of dry, loose, or disturbed soil.
An adult person typically breathes 17,000 liters of air daily, so a low concentration of airborne contaminant represents a large quantity of the contaminant entering the human body by inhalation. Every day, billions of particles are inhaled with the ambient air by every human being. Many of these particles are deposited in the respiratory tract; the deposition depends on the size, density, shape, charge, and surface properties of the particles and the breathing pattern of the individual. Therefore, the physical and chemical characterization of the contaminants are essential in understanding the health effect and selecting appropriate air filters to reduce their particle size distribution and concentration. From the toxicological point of view, all particles smaller than 10 μm in diameter have the potential of being biologically active in susceptible individuals.
Irritation of the nasal passages by dust and its soiling effects are matters of common observation, and some of the unpleasant manifestations, such as hay fever and asthma, are experienced by many. The case against dust and air pollution is not confined to their unpleasantness. Silicosis, caused by inhalation of minute particles of silica, and other forms of pneumoconiosis (caused by inhalation of dust, including, asbestos, coal, metallic particles, decaying organic matter from vegetation, or bird droppings) are serious pulmonary diseases. In short, many airborne particles are the source of disabling or fatal illnesses. There is no need to question the need for air cleaning.
Studies have suggested that drivers, passengers, and pedestrians are exposed to high concentrations of airborne pollutants emitted by vehicles [3,4]. In fact, nanosized particles, including those originating from combustion of different materials, have been linked to various health effects, such as fibrosis, chronic inflammatory lung disease, and cancer .
What has filtration done for us?
Air filtration is considered the most common method for cleaning air and used in many diverse applications; it encompasses multi-disciplinary fields, which add to the complexity of air filter performance assessment. Although the basic filtration principles seem straightforward, there is still a gap between theory and application. Ironically, we had assumed that the past 60 years of progress in diagnostics, vaccines, and supercomputing and in analytical tools, such as genomics, bioinformatics, and scanning electron microscopy, would render our preparedness intact to combat any outbreak. Our status quo today suggests that an intrinsic understanding of the dynamics of filter performance is imperative if we are serious about enhancing air quality. The novel SARS-CoV-2 virus requires not only new tools, but also novel attitudes, particularly as knowledge about the current COVID-19 coronavirus pandemic is still evolving.
Our modest filter selection
Most residential filtration relies on single-stage filtration, which is usually 1 or 2 inches thick, a disposal filter, or a washable metallic filter. When HEPA filters are used, they have to be preceded to a minimum of three filtration stages, namely, fresh air, prefiltration, and fine filtration. In recent decades, more emphasis has been placed on preventive maintenance (PM) filtration with medium efficiency and minor attention given to bioaerosols and gaseous filtration. Future air filtration plans would need to investigate new technologies to capture bioaerosols and gaseous contaminants.
Common filtration mechanisms
The performance characteristics of a filter are concerned with the efficiency it can provide at the expense of pressure drop. The overall efficiency of a filter is based on the combination of the dominant collection mechanisms for a given particle size range. Therefore, the particle size is of paramount importance in determining the overall filter efficiency. A particle is generally imagined to be spherical, and its diameter is usually used to describe its size. Unfortunately, there are several ways of defining particle size, particularly for those of irregular shapes as shown in Figure 2.
Straining: Straining occurs when the filter pores strains the particles due to particle size, leaving the particles lying predominantly on the surface of the filter medium. This mechanism does not play a major role in depth filtration.
Impaction: The impaction mechanism occurs when a particle cannot navigate through the inhomogeneous filter structure and does not follow the gas streamlines due to its inertia (size and mass) and hence, impacts the filter fiber and leaves the airstream. The impaction mechanism depends on the particle size, density of the dust, depth of the filter, and the velocity of the airflow. This mechanism is predominant for particles that have a high density or which have a diameter greater than 1 µm. Its influence increases with the size of the particle.
Interception: Interception occurs when a particle is intercepted by the fiber and adheres to it due to Van der Waals forces. Particles captured by interception do not have enough inertia to travel in a straight line to be filtered by impaction. Several studies have suggested the relative independence of the interception mechanism of the gas velocity [6; 7]. However, Dorman  stated that the interception mechanism is independent of the gas velocity, except in so far as the flow pattern changes with velocity. Jaroszczjk et al , on the other hand, highlighted the role of particle size and its velocity in the filtration processes and suggested that it may be determined experimentally by using actual filter media with careful selection of test sample and contaminants.
In gas filtration, it is common to assume that each mechanism acts independently of others. For particles of 1 µm and smaller, complex interactions between Brownian diffusion and inertial effects results in the so-called most penetrating particle size (MPPS). The MPPS is defined as the particle size at which the filter efficiency is the lowest at a given flow rate .
Diffusion occurs when ultra-fine particles (particles with a diameter, dp, <1µm) are bombarded by air molecules to adopt a random (Brownian) motion and are eventually captured by a filter fiber. To enhance the diffusion mechanism, the time of the air must be extended to increase the likelihood of particle-fiber attraction. In other words, subjecting the filter to a lower media velocity will have a pronounced effect on the filter efficiency if the particle size distribution is < 1µm. Introducing a smaller fiber size distribution relative to the particle size distribution can enhance the diffusional efficiency of the filter.
For PM1 (particles < 1µm in size), researchers use a simple way to account for the influence of interception and diffusion mechanism by combining the two efficiencies. Such addition is based on the assumption that only one mechanism is predominant at the time and the contribution of the other mechanism is negligible. Although the assumption does not represent, or rather overlooks, the fact that near the MPPS both mechanisms can be dominant and compete for the same particle, hence leading to the particle size, which becomes the MPPS, the results have provided a reasonable prediction of overall efficiency for sub 1µm particles.
Filtration is the separation and retention of particles from the airstream by means of porous medium. The essence of filtration lies in the particle capture by filter fibers within the filter depth and/or at its surface. However, when particles begin to collect other particles, filtration performance becomes more complex to predict. The first step in the general understanding of the air filtration process is distinguishing between the two types of filtration processes, namely depth and surface filtration. Fabric and membrane filters are considered surface filters, while fibrous and granular filters fall into the depth filtration category.
In surface filtration, large particles are expected to be deposited on the surface of the filter medium by a sieving (straining) mechanism so as to form the so-called dust cake (Figure 4). Thus, most of the filtration action takes place on the filter surface by means of dust-cake formation. In surface straining, any particle that is larger in size than the pores of the medium deposits on the surface and stays there until it is removed by a regeneration technique such as pulsing system. Although the aim is to have surface deposition of particles, some particles, smaller in size than the pores, can penetrate the filter medium. A dust cake can be formed by a combination of two primary mechanisms: bridging and complete blocking. Bridging takes place when particles smaller than the pore sizes in the filter medium form a cake. This occurs when the particles are at a higher concentration at the feed. On the other hand, complete blocking is a sieving process that occurs when particles are larger than the pore sizes.
Depth filtration relies on capturing the particles within the filter medium. This method of non-cake filtration requires an understanding of the media properties where most of the filtration action occurs. The depth filtration process causes fibrous filters to have the same or better degree of purification and minimal resistance as that of surface filters. In this process, glass fiber media is used extensively due to its low pressure drop and better performance at high temperatures. The structure of glass fiber media is delicate and may not be regenerated either by air and/or water. Attempting to regenerate a fibrous filter by ejecting the dust out of it will lead to the destruction of the structure of the media, and the dust particles may not be removed completely. The structure of fibrous filters is illustrated in Figure 5 in its stationary stage, which is defined as the stage the filter experiences a negligible pressure drop rise due to particle deposition. On the other hand, Figure 6 illustrates the progression of particle loading challenging fibrous media from stationary phase to nonstationary phase and eventually to dust cake formation. The nonstationary filtration occurs when particles deposited around the filter fibers alter the pore structure of the filter media and cause a pronounced rise in the resistance to airflow through the filter and, hence, in the filter’s pressure drop. An illustration of particles deposition on filter media by depth and surface filtrations is shown in Figure 7.
Parameters such as particle concentration and particle size distribution (compared to the pore size distribution) can lead to their surface deposition. Eventually, the filter porosity undergoes a substantial change leading to premature filter clogging and shorter service life. Surface deposition occurrence on a depth filter does not warrant the full utilization of filter depth/thickness as shown in Figure 8. It could also reduce the permeability of the filter and force it to depart from the depth deposition stage much earlier than predicted causing a steep rise in pressure drop.
HVAC and depth filtration
Depth filtration requires detailed knowledge of the media properties, such as packing density, porosity, and thickness, amongst other parameters. While a sieving mechanism does not play a significant role in depth filtration, increasing the filter media thickness enhances the efficiency at the expense of increasing the pressure drop. Depth filters are mostly disposable, as the energy required to regenerate them is excessive. Attempting to regenerate a disposable filter may not remove the entire captured particles totally and may not maintain the original pore size distribution of the filer media. Such regeneration attempts, whether by air or water, can lead to efficiency degradation and may induce other issues, such as micro-organism growth, if returned to the air-handling units.
Air filter selection
Research has revealed that the actual performance of the air filters installed in air-handling units tends to deviate from the performance predicted by laboratory test reports [11,12]. Various atmospheric airborne contaminants, moisture, and temperature can alter the air filter performance, and, therefore, their physical and chemical characteristics can prove invaluable for their appropriate selection.
There is a lot of emphasis on using HEPA filters currently to combat the SAR-CoV-2 virus. Numerous recommendations demand a higher class of HEPA filter to enhance the air quality. However, it is important to highlight that appropriate air filter selection is equally important in providing the operating conditions needed to perform as designed. These parameters include face velocity, low particle concentration, low particle size distribution, and appropriate temperature and humidity levels. Therefore, appropriate filter selection, operation, and maintenance are of paramount importance. Other challenges confronting filter performance include upstream and downstream flow nonuniformity of filter pleats filters due to various media properties, cartridge design, and filter installation. These parameters can influence the efficiency of the filter and its sustainable performance. Therefore, obtaining reliable velocity distribution information is important to optimize filter performance.
Prior to finalizing the specifications of HEPA or UPLA filters, a detailed account of the filter media properties used in their manufacturing must be forthcoming and readily available for consultants to examine and approve and for manufacturers to supply accordingly. Appropriate filter selection should be delegated to filtration experts who can be held responsible for their selection and the air quality we eventually breathe. Innovative filter and HVAC solutions must be improved to provide preventive measures to combat COVID-19.
Raising the bar of the filter efficiency
It is of paramount importance to ensure that air filters are operated at designed conditions as indicated by their associated test reports. Otherwise, performance deviation is usually inevitable. Gas velocity is a critical operational parameter influencing the filter efficiency. The investigation of airflow rate variations can facilitate the investigation of gas velocity on the filter performance. An experimental study of high efficiency air filters, shown in Figure 9, highlights the following observation [11, 12]:
- The filter efficiency decreases as particle size increases until the MPPS is reached. Beyond the MPPS, the efficiency increases with the increase of particle size for a given flow rate.
- The fractional efficiency generally decreases with the increase of gas velocity where the interception and diffusion mechanisms are dominant.
- The interception effect is relatively independent of the gas velocity variations.
- The MPPS shifts toward smaller particle size as the filter face velocity increases. Consequently, the fractional efficiencies of the filter exhibit fluctuations when compared with lower flow rates and their corresponding MPPS. It can be noticed that the MPPS decreases with the increase of flow rate.
Increasing the gas velocity shortens the particle dwell time inside the filtration medium. Consequently, it reduces the possibility for a particle-to-fiber contact, and, therefore, the efficiency may be compromised. Initial fractional efficiency and its corresponding MPPS can fluctuate as flow rate increases due to particles bounce and/or their detachment after they had been attached to the fiber surface. When the velocity is increased, particles can re-entrain into the airstream and therefore lower the overall efficiency. In highly efficient filters, this could mean the filter under test may fail to classify as EPA, HEPA or ULPA, if subjected to higher velocity than that rated in the associated test report. Conversely, the same filter could achieve higher efficiency than the rated one if subjected to a lower flow rate.
The preventive approach
While it is widely accepted, that SARS-2 is transmitted person to person and is not airborne over longer distances, it can, however, deposit onto surfaces due to people breathing, sneezing, and so on. Hence, it could get into HVAC systems, especially when air is extracted from populated rooms. We often regard HVAC systems as the usual suspect when it comes to poor air quality, as they are the air transporter. Moreover, the public should be mindful that practitioners and researchers are updating their knowledge and best practices in this field constantly. New research and experiences can broaden our understanding and provide tools and conditions to promote professional practices and necessary medical treatment in combating COVID-19. Furthermore, we ought to take all precautions and preventive measures pertaining to such an assumption. On the other hand, our HVAC and filtration systems must be ready and capable of confronting a wide array of pollutants, whether they are particulate, gaseous, and/or bioaerosol.
 Chen, C., Zhao, B., 2011. Review of relationship between indoor and outdoor particles: I/O ratio, infiltration factor and penetration factor. Atmos. Environ. 45, 275e288.
 EPA, 2005. Review of the national ambient air quality standards for particulate matter: policy assessment of scientific and technical information. OAQPS Staff Paper.
 Zak, Magdalena & Melaniuk-Wolny, Edyta & Widziewicz, Kamila. (2016). The exposure of pedestrians, drivers and road transport passengers to nitrogen dioxide. Atmospheric Pollution Research. 10.1016/j.apr.2016.10.011.
 Leavey A, Reed N, Patel S, Bradley K, Kulkarni P, Biswas P. Comparing on-road real-time simultaneous in-cabin and outdoor particulate and gaseous concentrations for a range of ventilation scenarios. Atmos Environ (1994). 2017;166:130-141. doi:10.1016/j.atmosenv.2017.07.016
 Byrne JD, Baugh JA. The significance of nanoparticles in particle-induced pulmonary fibrosis. Mcgill J Med. 2008;11(1):43-50.
 Brown R. C., Wake D. 1999. “Loading filters with monodisperse aerosols: macroscopic treatment”. J. Aerosol Sci., 30(2), 227-234.
 Lee K.W. and Mukund R., 2001. “Filter collection”, in “Aerosol Measurement: Principles, Techniques, and Applications” 2nd edition, Ed. P.A.Baron and K.Wileke, 197-228.
 Dorman R. G. 1964. “Theory of fibrous filtration”, in High efficiency air filtration”, Ed. White P.A.F. and Smith S.E., Butter Worths, London.
 Jaroszczjk T., Fallon S.L. and B.A. Pardue, 2002. “Analysis of engine air cleaner efficiency for different size dust distributions”, Fluid/Particle Separation, 14(2). 75-88.
 Tarleton E.S. and Wakeman R.J. 2008. “Dictionary of Filtration and Separation” Filtration Solutions, Exeter.
 Al-Attar I.S., 2011. “The effect of pleating density and dust type on performance of absolute fibrous filters”, doctoral diss., Loughborough University Institutional Repository.
 Al-Attar I.S., Wakeman, R.J., Tarleton, E.S., and Husain A., 2009. The effect of pleat count and air velocity on the initial pressure drop and fractional efficiency of HEPA filters, Filtration Journal, 10 (3), 200-206.
 DIN EN 1822-1 E:2011 High efficiency air filters (EPA, HEPA and ULPA) Part 1: Classification, performance testing, marking.