At face value, mask filtration seems quite simple - particles try to pass through holes in the mask and are caught in a sieve-like fashion. Right? Not exactly. There is much more to how masks filter airborne particles than you might anticipate! In fact, they rarely filter like a sieve at all.
There are many misinterpretations as to how exactly masks filter. This leads to many misconceptions about how masks work and their capabilities. A lack of easily accessible resources and false information is primarily to blame. It's virtually impossible to understand the fundamentals behind masks and filtration without delving into academic papers.
We noticed this lack of accessible knowledge and wrote an article explaining how masks function and filter particles. On top of this, we also wanted to address the difference between two terms that are often mentioned - electrostatic filtration and mechanical filtration.
If these terms sound a bit confusing, don't worry! We will explain from the ground up how masks filter particles and the different mechanisms in play. If you want to dive deeper, all of the sources for this article are provided. However, you will be able to take away all of the essentials from this article alone.
How Does Mask Filtration Work?
It's essential to clarify the fundamentals before moving on. While this may seem very basic, filtration can differ slightly between its different applications. Therefore, let's first define filtration within the context of masks:
'Filtration is the process of separating solid particles present in a solid-fluid mixture to increase the purity of the air that passes through the material, which is called a filter.' (1)
Put simply, filtration in masks is all about removing solid particles from the air that passes through. These particles can be anything from PM2.5 to virions (viral particles). The air passing through the filter acts very similarly to liquid regarding how it navigates through the device.
As atmospheric air passes through a filter, it creates a streamline. Similarly to liquid, the airflows passing through the filter go from higher pressure to lower pressure areas. This pressure-based movement is similar to the flow of fluids.
As air passes through the filter in streamlines, it has to navigate through a 'web' or mesh containing thousands of micro or nanofibres. This mesh aims to 'catch' and adhere solid particles following the streamlines while still allowing air through. In this way, the wearer can receive air from which most solid particles have been removed.
How are Filters Made?
While there is a range of different methods for creating filters, they all rely on utilising meshes of fibres that are then layered to create what essentially becomes a maze for particles to pass through before reaching the wearer's lungs.
This concept is relatively straightforward. However, the issue is that there is a balance that filters must strike. While adding more layers of micro or nanofibres to a filter will increase the percentage of particles caught, it will also decrease the airflow. This means that an increase in filtration efficacy often leads to a decrease in the breathability of a filter.
Therefore, a filter needs to have sufficient pores (holes through which air can pass) and enough fibre density to filter harmful particles out before passing through the filter media.
Most filters that you will hear about, and those primarily discussed in this article, are melt-blown. Melt-blown filters are found in the vast majority of N95 respirators. However, they are also found in a range of other respirators, such as KN95 and KF94 devices.
Melt-blown filters are nonwoven, meaning that the fibres are bonded together using heat or chemical treatment. Where woven materials are designed in a perpendicular, overlapping fashion, nonwoven filters are web-like structures (2). This means that fibre placement is somewhat random, and generally, nonwoven structures are subpar to woven materials on the same scale.
The melt-blown nonwoven approach is so common because it is affordable and mass-producible. Melt-blown membranes typically have a fibre size of 0.5 to 10μm and are made up of fibres with a diameter of 1-2μm (3). Further, the typical melt-blown filter fabric has a pore size of 29.285μm (4). This last fact will come in important later one!
As we mentioned at the top of this article, masks do not usually filter by sieving. Only the largest particles will be filtered using a sieving mechanism. The more harmful particles such as PM2.5 and pathogens are far smaller and can easily fit through the pores in a filter.
Since these tiny particles can so easily fit through the pores, there have to be other filtration mechanisms in place to catch them. Remember, the average pore size on melt-blown filters is just under 30μm; however, PM2.5 is merely 2.5μm! Virions can be even smaller, with COVID-19 particles being around 100nm.
Rather than sieving particles following airstreams through the filter, these mechanisms all perform slightly differently to catch particles through depth filtration. Depth filtration is the system through which particles get caught as they attempt to navigate through the layers of the filter media.
There are four critical mechanisms through which fibrous filters prevent solid particles from reaching the wearer. These mechanisms are inherent in all microfibre and nanofibre filters and are the basis upon which fine-particle filtration is built. These are the four mechanisms:
Gravitational settling (large particles, 1μm - 10μm) - Occurs when large particles enter the filter but gradually fall as the forces of gravity and ballistic forces act on them. Once these particles fall, they will settle in the filter and remain. This only impacts the largest particles that enter the filter.
Inertial impaction (medium particles, > 0.2μm) - As particles pass through the filter, they follow air streamlines. However, larger particles, those with higher densities or those moving at a higher velocity, will have too much inertia and stray from the streamlines. Once strayed, the particle will likely collide with fibre and adhere to it. Inertial impaction impacts mid-sized particles and has minimal effect on nanoparticles.
Direct interception (medium/small particles, < 0.6μm) - Also impacting medium-sized particles, direct interception occurs when particles follow streamlines but pass too close to a fibre. Once in contact with the fibre, the particle will adhere to it. Direct interception impacts both medium and small particles.
Brownian Diffusion (small particles, < 0.2 μm) - The smallest particles are impacted by Brownian Motion, a phenomenon where nanoparticles move somewhat random and uncontrollably. These abnormal movements cause the particles to stray from streamlines. While wondering, these microscopic particles will often contact a fibre and get caught on the surface. (5)
These mechanisms work in conjunction to filter both larger and smaller particles. At some points, these mechanisms will simultaneously impact particles (such as for those under < 0.2μm where both interception and diffusion are active), while at other times a sole mechanism will be at work.
Through these four mechanisms, the majority of larger particles can be caught by the fibres of a filter. However, particles smaller than 100nm can penetrate nonwoven melt-blown filters. As such, a new mechanism had to be introduced that was effective against these ultra-fine particles.
This mechanism is electrostatic filtration. Created by Peter Tsai in 1992, electrostatic filtration is different from the aforementioned mechanical filtration mechanisms as it isn't inherent in fibrous filters. Instead, filter media must undergo a charging process to gain an electrostatic charge. So let's dive into how electrostatic filtration differs from mechanical filtration mechanisms.
Electrostatic filtration is very common in respirators such as N95's. In fact, Peter Tsai patented the N95 mask filter, which was created based on electrostatic filtration. Today, it is likely that all NIOSH certified respirators (N95, N99, N100) contain some electrostatic filter media (6).
While Brownian diffusion is effective against many smaller particles, mechanical filtration struggles against nanometer-sized particles. This is because, in typical N95 respirators, there is a pore size of almost 30μm. With pores this large, some particles can pass through the filter without being impacted by the mechanical filtration mechanisms. This is where electrostatic filtration becomes essential.
So how does electrostatic filtration work? When filters are created, they go through a charging process that gives them a static charge. These devices can then attract and intercept both charged and uncharged particles attempting to navigate through the filter.
Particles are attracted and caught via Coulombic forces (charged particles) and dielectrophoretic forces (neutral/uncharged particles). These forces can remove particles from the airstream that mechanical mechanisms in microfibre filters can't due to the large pore sizes.
Typically, electrostatic filtration is carried out by an added polypropylene layer. Polypropylene is an electret, meaning that it can hold a charge. Being dielectric, this charge will not create currents that flow through the material. Instead, electric polarisation occurs, and the charges are displaced minutely. They are charged enough, though, to attract ultra-fine particles passing through.
N95 masks have substantially less filtration efficacy without an electret layer, especially against nanoparticles. Many studies have characterised this struggle that shows charged (or recharged) electret filters significantly outperform their non-conductive counterparts. (7, 8)
Electrostatic filtration provides many benefits for the filtration capabilities of a mask or respirator. However, one big downside is that the electrostatic charge can be lost, especially in humid environments. For this reason, most N95 respirators are recommended for only one use.
A study tested an N95 respirator's filtration efficacy over a range of particle sizes at 10%, 30% and 70% relative humidity (RH). Across the tested particle size range of 15nm to 200nm, the respirator in 10% RH outperformed both the 30% and 70% RH respirators. At the MPPS, the 10%, 30% and 70% RH respirators had 2.73, 3.30, and 4.27% penetration rates, respectively.
This shows that humidity strongly impacts filters that rely heavily on electrostatic filtration. Over time, this impact can increase and lead to an even more significant performance difference between a respirator exposed to low RH and one exposed to high RH. (9)
KN95 Filtration efficiency after washing and recharging. Source.
Another interesting study was done with a KN95 mask that was initially measured to have 95% filtration. This mask was washed with detergent and then air-dried in a standard washing machine. After this process, the mask's filtration was tested again and recorded to be 75%. The mask was recharged and tested for the third time - this time, the filtration efficacy was measured at 95%. (10)
These findings show that while electrostatic filtration greatly increases the filtration properties of filter media, the filtration efficacy will drop over time as the media is exposed to water particles in humid environments. Further, these devices cannot be washed with water-based methods without losing significant filtration.
While recharging a mask will allow it to regain most of its filtration capabilities, most users will not have access to tools to do so. Therefore, consumers should be careful to replace electrostatic filters when the charge is lost. While it's hard to identify when the filtration capabilities drop, you should never wash an electrostatic filter. Further, it's worth replacing these filters after days with high humidity.
It is important to note that all of the research in this section has been carried out on N95 and KN95 certified respirators. These products are designed to be one-use and, as such, have a limited lifespan. However, some reusable masks have developed innovative methods to retain filtration even when relying on electret filters.
AirPop uses a multi-layer filter with an outer layer of non-woven fluid-resistant material, followed by a spun-bond mechanical pre-filter and an electrostatic melt-blown core. These layers act together to perform both mechanical and electrostatic filtration and have been been shown to retain > 97% filtration efficacy even after 10 was cycles using a 70% ethanol solution. (15)
Even after five wash cycles that immersed the Light SE Mask in a 1:50 bleach solution before being hand-washed, the mask was able to maintain > 98% filtration efficacy (16). Therefore, although the devices rely heavily on electrostatic filtration, they can maintain high filtration even after wash cycles that would normally degrade a mask or respirator significantly more.
Electrostatic vs Mechanical Filtration
In the 90s, filter manufacturers had two choices. The first was to create ticker or denser filters that relied solely on mechanical filtration. These filters were capable of filtering a high percentage of particles, but in doing so, they sacrificed breathability. The second choice was for manufacturers to opt for electrostatic filters, which could be significantly more breathable while also achieving a similar filtration efficacy (6).
This led to electrostatic filtration being widely adopted. Decreasing the breathability of masks and respirators even further would make them even less comfortable to wear. On the other hand, an electrostatic charge capable of significantly increasing the filtration of a device could be added with no downside to the wearer.
On top of this, electrostatic filtration can capture ultra-fine nanoparticles that diffusion and impaction can't be relied upon to filter in microfibre devices. In a recent study, an N95 had a filtration efficacy of ≥ 97.52% while electrostatically charged. However, when this charge was removed, the effectiveness dropped to < 50% for particle penetration at the MPPS. This massive difference shows how heavily these respirators rely on electrostatic filtration (11).
While electrostatic greatly increases the filtration efficacy of masks, especially when it comes to smaller particles, it also has downsides. The biggest downside is that the charge of the filter can be decreased or lost altogether when it comes in contact with water. What makes this a big issue is that even water vapour, as found in high humidity environments, can decrease efficacy over time.
Whereas mechanical filtration mechanisms will remain active until the filter becomes deformed, electrostatic filtration can lose efficacy after a few hours of wear in environments with high RH. If washed, nonwoven melt-blown filters will degrade in performance even faster.
This is highly reliant on the quality of the filter media and the filter design, however, and some electret filters are able to remain effective for even up to 50 hours (15). These filters tend to use higher quality materials that are more conductive and contain smaller fibres in a consistent layout capable of attracting more particles.
Another interesting fact about electret filters is that the most penetrating particle size is lower. The most penetrating particle size is typically around 300nm for filters relying solely on mechanical filtration. However, the MPPS can be far lower for electrostatic filters - some studies have shown it to be around 50nm (12).
To summarise everything, mechanical filtration mechanisms are inherent in fibrous filters and are always active. However, their performance can degrade if the fibres in the filter get damaged. Mechanical filtration is typically effective on melt-blown filters down to around 100nm. At this point, many particles are too small to be captured by fibres, and electrostatic filtration comes into play.
Therefore, electrostatic filtration is a very beneficial addition when it comes to masks and respirators utilising melt-blown filters. However, it can also wear off, which is why masks that rely heavily on this filtration method should be replaced regularly.
The Future of Filters
Fibre size in melt blown and nanofibre filters. Source.
Before closing this article, there is one topic that we want to discuss. While the melt-blown filters in use on most masks and respirators currently require electrostatic filtration, modern innovations and technology have led other filtration methods to become capable.
In this article, we focused heavily on microfibre filters. Studies have found melt-blown N95 filters generally contain fibres with a diameter of 1-2μm (3) a pore size of 29.285μm (4). However, modern manufacturing can shrink these fibres down to a nanoscale with diameters of 100nm - 500nm (3). This has some exciting implications.
Since nanofibres are significantly thinner, they can be placed more densely, reducing pore sizes. In addition, since the fibres are smaller and pores are more plentiful (albeit smaller), breathability can be retained better than on a microfibre filter. Due to this, nanofibre filters don't need to rely on electrostatic filtration and therefore don't suffer the same downsides.
Nanofibre filters relying solely on mechanical filtration mechanisms have been shown to provide and retain filtration efficacy far better than microfibre counterparts. A study tested meltdown and nanofibre filters after ethanol cleaning cycles. The melt-blown filter dropped to 64% filtration after one cycle, but the nanofibre filter retained 97-99% filtration (13).
Similar findings were found in another study where nanofibre and melt-blown filtration were compared after several water dipping cycles. After four cycles, the melt-blown filter had dropped to 79% filtration efficacy, whereas the nanofibre filter retained > 99% filtration (14).
We wanted to mention these results because they demonstrate how effective mechanical filtration mechanisms can be with the right application. While electrostatic filtration is vital in the microfibre filters that we currently use regularly, mechanical filtration mechanisms can be equally as effective or even more so in masks utilising smaller fibres.
These filters are still a relatively new innovation, and they are significantly more expensive to produce. However, in the future, we may see a switch from electrostatic filtration being the predominant filtration mechanism to filters relying purely on mechanical mechanisms.
Further, many nanofibre filter medias perform significantly worse than this ‘best case scenario’. While these filters can, in theory, provide extremely effective filtration relying solely on mechanical mechanisms, the technology is still very new and many manufacturers have not yet been able to reach nanofibres of high enough quality to retain such effectiveness.