Electrospun filter media for protection against PM2.5

Fig 1. Haze shrouded city

Air pollution in a city is often gauged by measuring the amount of PM_2.5(Particulate matter, 2.5 µm) in the environment as particles of this size is believed to pose the greatest health risks. Such particles come from various sources such as car exhaust, burning, cooking and power plants. Having a filter barrier is the most effective form of protection since it is not possible to eliminate the source of PM2.5. Electrospun nanofiber coating on filtration substrate has been known to significantly increase its filtration performance and has been used commercially in filter medium for both industrial application and personal use. However, most investigation are based on particle sizes that are less than 1 µm. With increasing awareness of air pollution in cities, the effectiveness of electrospun nanofibers in the removal of PM2.5 require greater investigations.

An early investigation on electrospun fibers targeted at PM2.5 removal was performed by Triped et al (2009) using silk fibroin. With a fiber diameter of 10 µm, its filtration efficiency for PM2.5 was found to be 39% at an air-flow rate of 5 L/min. Such low performance of the electrospun fiber media is probably due to its fiber size. Several studies on air filtration performance of electrospun fibers have shown that fiber size of less than 300 nm is necessary for improved filtration performance [Li et al 2006]. Using electrospun polyacrylonitrile fibers, Liu et al (2005) showed that as the fiber diameter increases from 200 nm to 1 µm, the removal efficiencies decreased significantly from 98% to 48%. At relatively large membrane thickness the filtration efficiency of PM2.5 are generally not influenced by the material used with the exception of polystyrene which exhibits low removal efficiency of about 40% [Liu et al 2005]. Given that polystyrene is a common filtration material, this result will require further investigation.

Fig. 2 Permeability and intercept rate comprehensive comparison [Li and Gong. Journal of Chemistry, vol. 2015, Article ID 460392, 2015. This work is licensed under a Creative Commons Attribution 3.0 Unported License.].

Li and Gong (2015) did a study comparing the filtration efficiency and pressure drop of electrospun polysulfone fibers against several commercially available face masks for PM2.5 protection (See figure 2). Typical air pressure drop in commercially available face mask is between Δ0.5 to Δ0.7. Pressure drop for electrospun fiber deposition at 15 minutes is about Δ0.41 and at 30 minutes is about Δ0.87. Rejection rate is 91% for the former and 97% for the latter. Thus it can be anticipated that a fiber deposition duration of between 15 to 30 minutes will provide a nanofiber coating that gives a balance between pressure drop and rejection.

At low electrospun fiber membrane thickness, preliminary investigation suggested that its performance is dependent on the material used. Polyvinylpyrrolidone and polyvinyl alcohol exhibited deteriorating removal efficiency with reduced electrospun fiber membrane thickness while polyacrylonitrile (PAN) maintains its particle removal efficiency. Thickness of the membrane has an effect on its transparency and a thin electrospun PAN filter membrane was able to show removal efficiency of more than 95% while allowing 90% transparency [Liu et al 2015]. The reason for the material, thickness and removal efficiency is not apparent although it may be due to the strength and rigidity of the fibers. Comparison with commercial filtration membranes showed much better quality factor in transparent PAN electrospun membrane [Liu et al 2015]. A detailed study on the effect of electrospun PAN fiber diameter and its distribution on the slip-flow effect and pressure drop were reported by Zhao et al (2016). Fiber diameters between 60-100 nm was found to be most effective in facilitating the slip flow effect which coincides with the mean free path of air molecules at 65.3 nm. With optimum slip flow, the pressure drop would be the least. When electrospun PAN fiber diameter drops below 60 nm, the pressure drop starts to increase and this has been attributed to smaller pore size due to closer packing between smaller diameter nanofibers and this interferes with air flow. Their studies suggests that a pore size greater than 3.5 µm is needed for slip flow. With optimum electrospun PAN membrane parameters, they were able to achieve air flow resistance of 29.5 Pa, a PM_2.5 filtration efficiency of 99.09% with transmittance of 77%.

Fig. 3 Personal facemask

Given the interests in PM2.5 filtration and growing evidence that electrospun fibers are very effective in this application, research has moved into improving its effectiveness. Jing et al (2016) showed that it is possible to improve the filtration performance of electrospun PAN fibers by doping the spinning solution with ionic liquid diethylammonium dihydrogen phosphate (DEAP). The resultant DEAP/PAN nanofibers exhibits significantly better PM2.5 filtration performance than neat PAN nanofibers. This improvement was attributed to higher surface roughness, hydrophilicity, and dipole moment.

Currently it is unclear what are the factors influencing the filtration efficiency of the membrane for PM_2.5 particles. In a study by Kim et al (2019) comparing the filtration performance of electrospun polyvinyl alcohol (PVA), polyvinylidene fluoride (PVDF) and polyacrylonitrile (PAN) membrane showed that the material used does have an influence on performance. Electrospun PVA has the smallest diameter at 210 nm but it has the lowest PM_2.5 removal efficiency at 71%. Electrospun PAN and PVDF has larger diameters at 500 nm and 590 nm respectively, showed better PM_2.5 removal efficiency. For single and small molecules, dipole of PAN is higher than PVA and PCDF However, for larger molecules, dipole of PAN is the lowest. In electrospinning, the polymer needs to be of higher molecular weight, therefore, dipole moment is probably not a critical factor affecting filter performance. In general, filtration performance increases as the film thickness increases. Kim et al (2019) showed that as the thickness of electrospun PAN membrane increases, so is its filtration performance. When compared with electrospun PVDF and PVA, PAN with a thickness of 40 µm has a higher PM_2.5 filtration efficiency (93%) compared to slightly thicker PVDF and PVA electrospun membranes which has filtration efficiency of 83% and 71% respectively. The PAN electrospun membrane filtration efficiency is close to that of commercial semi-HEPA filter (96%) and quality factor although the commercial semi-HEPA filter is much thicker at 650 µm.

An interesting version of PM2.5 capturing was demonstrated by Zhang et al (2017) using in situ electrospinning of chitosan. Instead of using the end product of electrospinning for filtering, electrospinning was carried out in a "polluted" chamber. A removal rate of up to 3.7 µg m-3s-1 was recorded. The excellent PM2.5 capturing using this method was attributed to strong polarity of chitosan, electrostatic adsorption due to the presence of surface charges on the electrospinning jet and surface adhesion of the electrospun nanofibers.

For practical application of air filters, the effect of humidity and wetting on the PM_2.5 filtering performance needs to be investigated. Kim et al (2019) compared the filtration performance of electrospun polyvinyl alcohol (PVA), polyvinylidene fluoride (PVDF), polyacrylonitrile (PAN) membrane and commercial semi-HEPA microfiber filters. Electrospun PVA membrane was found to be unsuitable for application where there is a risk of water contact as it dissolves in it. PAN-NF membrane has the lowest water contact angle (WCA) at 11°. PVDF-NF and semi-HEPA microfiber filters has WCA of 133° and 124° respectively. Wetting caused a drop in air filtration rate in all the filters due to increase in pressure drop as their pores were blocked by water. Nanofiber filters has a greater initial drop in performance compared to microfiber filters as the nanofibers absorbed more water. However, recovery of PAN-NF filter were faster than PVDF-NF and semi-HEPA microfiber filters. Note that water gets absorbed quickly into PAN-NF filters due to the low water contact angle but remained on the surface of hydrophobic PVDF-NF and semi-HEPA. Hydrophilic PAN-NF may have helped to draw the water away from the pores and into the fiber interior. The low thickness of PAN-NF membrane also helps to encourage airflow through it and remove the moisture within. To retain adsorption of ultrafine particulate matters (PM) when moist, Choi et al (2021) introduced functional groups with permanent charges to the electrospun membrane. The electrospun poly(butylene succinate) (PBS) membrane was coated with chitosan nanowhiskers (CsWs) chitosan which contains cationic sites and polar amide groups that is able to attract polar ultrafine PM (e.g., SO₄^2- and NO^3-) even under humid conditions. By varying the concentration of PBS solution, Choi et al (2021) was able to electrospun both PBS micro and nanofibers. electrospin individual membranes made of microfiber and nanofiber respectively. CsWs were loaded onto the membrane by dip coating in a CsWs suspension. The presence of CsWs significantly increases the filtration performance of nanofiber membrane and microfiber membrane with little increase in pressure drop. By combining a layer of nanofiber membrane and microfiber membrane with CsWs coating, they were able to achieve a PM_1.0 removal efficiency of 97.5% and PM_2.5 removal efficiency of 98.3%. The quality factor of the integrated membrane is also greater than that of individual CsWs coated nanofiber and microfiber membranes. The maximum pressure drop was found to be 59 Pa (comfortable for human breathing) with negligible loss of PM removal efficiency when completely wet. The superior removal efficiency can be attributed to a combination of physical size exclusion and electrostatic adsorption coming from the permanent CsW charges.