Center of Excellence for Protective Equipment and Materials (CEPEM), 1280 Main St. W., Hamilton, ON, Canada
Use the link below to share the full text version of this article with your friends and colleagues. learn more.
Public health agencies recommend that communities use masks to reduce the spread of airborne diseases such as COVID-19. When the mask acts as a high-efficiency filter, the spread of the virus will be reduced, so it is important to evaluate the particle filtration efficiency (PFE) of the mask. However, the high costs and long lead times associated with purchasing a turnkey PFE system or hiring an accredited laboratory hinder the testing of filter materials. There is clearly a need for a “customized” PFE test system; however, the various standards that prescribe PFE testing of (medical) masks (for example, ASTM International, NIOSH) vary greatly in the clarity of their protocols and guidelines. Here, the development of an “internal” PFE system and method for testing masks in the context of current medical mask standards is described. According to ASTM international standards, the system uses latex spheres (0.1 µm nominal size) aerosols and uses a laser particle analyzer to measure the particle concentration upstream and downstream of the mask material. Perform PFE measurements on various common fabrics and medical masks. The method described in this work meets the current standards of PFE testing, while providing flexibility to adapt to changing needs and filtering conditions.
Public health agencies recommend that the general population wear masks to limit the spread of COVID-19 and other droplet and aerosol-borne diseases. [1] The requirement to wear masks is effective in reducing transmission, and [2] indicates that untested community masks provide useful filtering. In fact, modeling studies have shown that the reduction in COVID-19 transmission is almost proportional to the combined product of mask effectiveness and adoption rate, and these and other population-based measures have a synergistic effect in reducing hospitalizations and deaths. [3]
The number of certified medical masks and respirators required by healthcare and other frontline workers has increased dramatically, posing challenges to existing manufacturing and supply chains, and causing new manufacturers to quickly test and certify new materials. Organizations such as ASTM International and the National Institute of Occupational Safety and Health (NIOSH) have developed standardized methods for testing medical masks; however, the details of these methods vary widely, and each organization has established its own performance standards.
Particulate filtration efficiency (PFE) is the most important characteristic of a mask because it is related to its ability to filter small particles such as aerosols. Medical masks must meet specific PFE targets[4-6] in order to be certified by regulatory agencies such as ASTM International or NIOSH. Surgical masks are certified by ASTM, and N95 respirators are certified by NIOSH, but both masks must pass specific PFE cut-off values. For example, N95 masks must achieve 95% filtration for aerosols composed of salt particles with a count average diameter of 0.075 µm, while ASTM 2100 L3 surgical masks must achieve 98% filtration for aerosols composed of latex balls with an average diameter of 0.1 µm Filter.
The first two options are expensive (>$1,000 per test sample, estimated to be >$150,000 for specified equipment), and during the COVID-19 pandemic, there are delays due to long delivery times and supply issues. The high cost of PFE testing and limited access rights—combined with a lack of coherent guidance on standardized performance evaluations—have led researchers to use a variety of customized testing systems, which are often based on one or more standards for certified medical masks.
The special mask material testing equipment found in the existing literature is usually similar to the above-mentioned NIOSH or ASTM F2100/F2299 standards. However, researchers have the opportunity to choose or change the design or operating parameters according to their preferences. For example, changes in sample surface velocity, air/aerosol flow rate, sample size (area), and aerosol particle composition have been used. Many recent studies have used customized equipment to evaluate mask materials. These equipment use sodium chloride aerosols and are close to NIOSH standards. For example, Rogak et al. (2020), Zangmeister et al. (2020), Drunic et al. (2020) and Joo et al. (2021) All constructed equipment will produce sodium chloride aerosol (various sizes), which is neutralized by electric charge, diluted with filtered air and sent to the material sample, where optical particle sizer, condensed particles of various Combined particle concentration measurement [9, 14-16] Konda et al. (2020) and Hao et al. (2020) A similar device was built, but the charge neutralizer was not included. [8, 17] In these studies, the air velocity in the sample varied between 1 and 90 L min-1 (sometimes to detect flow/velocity effects); however, the surface velocity was between 5.3 and 25 cm s-1 between. The sample size seems to vary between ≈3.4 and 59 cm2.
On the contrary, there are few studies on the evaluation of mask materials through equipment using latex aerosol, which is close to the ASTM F2100/F2299 standard. For example, Bagheri et al. (2021), Shakya et al. (2016) and Lu et al. (2020) Constructed a device to produce polystyrene latex aerosol, which was diluted and sent to material samples, where various particle analyzers or scanning mobility particle size analyzers were used to measure particle concentration. [18-20] And Lu et al. A charge neutralizer was used downstream of their aerosol generator, and the authors of the other two studies did not. The air flow rate in the sample also changed slightly—but within the limits of the F2299 standard—from ≈7.3 to 19 L min-1. The air surface velocity studied by Bagheri et al. is 2 and 10 cm s–1 (within the standard range), respectively. And Lu et al., and Shakya et al. [18-20] In addition, the author and Shakya et al. tested latex spheres of various sizes (ie, overall, 20 nm to 2500 nm). And Lu et al. At least in some of their tests, they use the specified 100 nm (0.1 µm) particle size.
In this work, we describe the challenges we face in creating a PFE device that conforms to the existing ASTM F2100/F2299 standards as much as possible. Among the main popular standards (ie NIOSH and ASTM F2100/F2299), the ASTM standard provides greater flexibility in parameters (such as air flow rate) to study the filtering performance that may affect PFE in non-medical masks. However, as we As demonstrated, this flexibility provides an additional level of complexity in designing such equipment.
The chemicals were purchased from Sigma-Aldrich and used as is. Styrene monomer (≥99%) is purified through a glass column containing an alumina inhibitor remover, which is designed to remove tert-butylcatechol. Deionized water (≈0.037 µS cm–1) comes from the Sartorius Arium water purification system.
100% cotton plain weave (Muslin CT) with a nominal weight of 147 gm-2 comes from Veratex Lining Ltd., QC, and the bamboo/spandex blend comes from D. Zinman Textiles, QC. Other candidate mask materials come from local fabric retailers (Fabricland). These materials include two different 100% cotton woven fabrics (with different prints), one cotton/spandex knitted fabric, two cotton/polyester knitted fabrics (one “universal” and one “sweater fabric”) and A non-woven cotton/polypropylene blended cotton batting material. Table 1 shows a summary of known fabric properties. In order to benchmark the new equipment, certified medical masks were obtained from local hospitals, including ASTM 2100 Level 2 (L2) and Level 3 (L3; Halyard) certified medical masks and N95 respirators (3M).
A circular sample of approximately 85 mm diameter was cut from each material to be tested; no further modifications were made to the material (for example, washing). Clamp the fabric loop in the sample holder of the PFE device for testing. The actual diameter of the sample in contact with the air flow is 73 mm, and the remaining materials are used to tightly fix the sample. For the assembled mask, the side that touches the face is away from the aerosol of the supplied material.
Synthesis of monodisperse anionic polystyrene latex spheres by emulsion polymerization. According to the procedure described in the previous study, the reaction was carried out in a semi-batch mode of monomer starvation. [21, 22] Add deionized water (160 mL) to a 250 mL three-necked round bottom flask and place it in a stirring oil bath. The flask was then purged with nitrogen and inhibitor-free styrene monomer (2.1 mL) was added to the purged, stirred flask. After 10 minutes at 70 °C, add sodium lauryl sulfate (0.235 g) dissolved in deionized water (8 mL). After another 5 minutes, potassium persulfate (0.5 g) dissolved in deionized water (2 mL) was added. Over the next 5 hours, use a syringe pump to slowly inject additional inhibitor-free styrene (20 mL) into the flask at a rate of 66 µL min-1. After the styrene infusion was completed, the reaction proceeded for another 17 hours. Then the flask was opened and cooled to end the polymerization. The synthesized polystyrene latex emulsion was dialyzed against deionized water in a SnakeSkin dialysis tube (3500 Da molecular weight cut-off) for five days, and the deionized water was replaced every day. Remove the emulsion from the dialysis tube and store it in a refrigerator at 4°C until use.
Dynamic light scattering (DLS) was performed with Brookhaven 90Plus analyzer, the laser wavelength was 659 nm, and the detector angle was 90°. Use the built-in particle solution software (v2.6; Brookhaven Instruments Corporation) to analyze the data. The latex suspension is diluted with deionized water until the particle count is approximately 500 thousand counts per second (kcps). The particle size was determined to be 125 ± 3 nm, and the reported polydispersity was 0.289 ± 0.006.
A ZetaPlus zeta potential analyzer (Brookhaven Instruments Corp.) was used to obtain the measured value of the zeta potential in the phase analysis light scattering mode. The sample was prepared by adding an aliquot of latex to a 5 × 10-3m NaCl solution and diluting the latex suspension again to achieve a particle count of approximately 500 kcps. Five repeated measurements (each consisting of 30 runs) were performed, resulting in a zeta potential value of -55.1 ± 2.8 mV, where the error represents the standard deviation of the average value of the five repetitions. These measurements indicate that the particles are negatively charged and form a stable suspension. DLS and zeta potential data can be found in the supporting information tables S2 and S3.
We built the equipment in accordance with ASTM International standards, as described below and shown in Figure 1. The single-jet Blaustein atomization module (BLAM; CHTech) aerosol generator is used to produce aerosols containing latex balls. The filtered air stream (obtained through the GE Healthcare Whatman 0.3 µm HEPA-CAP and 0.2 µm POLYCAP TF filters in series) enters the aerosol generator at a pressure of 20 psi (6.9 kPa) and atomizes a portion of the 5 mg L-1 suspension The liquid is injected into the latex ball of the equipment through a syringe pump (KD Scientific Model 100). The aerosolized wet particles are dried by passing the air stream leaving the aerosol generator through a tubular heat exchanger. The heat exchanger consists of a 5/8” stainless steel tube wound with an 8-foot-long heating coil. The output is 216 W (BriskHeat). According to its adjustable dial, the heater output is set to 40% of the maximum value of the device (≈86 W); this produces an average outer wall temperature of 112 °C (standard deviation ≈1 °C), which is determined by a surface-mounted thermocouple (Taylor USA) measurement. Figure S4 in the supporting information summarizes the heater performance.
The dried atomized particles are then mixed with a larger volume of filtered air to achieve a total air flow rate of 28.3 L min-1 (that is, 1 cubic foot per minute). This value was chosen because it is the accurate flow rate of the laser particle analyzer instrument sampling downstream of the system. The air stream carrying the latex particles is sent to one of two identical vertical chambers (ie smooth-walled stainless steel tubes): a “control” chamber without mask material, or a circular-cut “sample” chamber-use detachable The sample holder is inserted outside the fabric. The inner diameter of the two chambers is 73 mm, which matches the inner diameter of the sample holder. The sample holder uses grooved rings and recessed bolts to tightly seal the mask material, and then insert the detachable bracket into the gap of the sample chamber, and seal it tightly in the device with rubber gaskets and clamps (Figure S2, support information).
The diameter of the fabric sample in contact with the airflow is 73 mm (area = 41.9 cm2); it is sealed in the sample chamber during the test. The airflow leaving the “control” or “sample” chamber is transferred to a laser particle analyzer (particle measurement system LASAIR III 110) to measure the number and concentration of latex particles. The particle analyzer specifies the lower and upper limits of particle concentration, respectively 2 × 10-4 and ≈34 particles per cubic foot (7 and ≈950 000 particles per cubic foot). For the measurement of latex particle concentration, the particle concentration is reported in a “box” with a lower limit and an upper limit of 0.10–0.15 µm, corresponding to the approximate size of singlet latex particles in the aerosol. However, other bin sizes can be used, and multiple bins can be evaluated at the same time, with a maximum particle size of 5 µm.
The equipment also includes other equipment, such as equipment for flushing the chamber and particle analyzer with clean filtered air, as well as necessary valves and instruments (Figure 1). The complete piping and instrumentation diagrams are shown in Figure S1 and Table S1 of the supporting information.
During the experiment, the latex suspension was injected into the aerosol generator at a flow rate of ≈60 to 100 µL min-1 to maintain a stable particle output, approximately 14-25 particles per cubic centimeter (400 000-per cubic centimeter) 700 000 particles). Feet) in a bin with a size of 0.10–0.15 µm. This flow rate range is required because of the observed changes in the concentration of latex particles downstream of the aerosol generator, which may be attributed to changes in the amount of latex suspension captured by the liquid trap of the aerosol generator.
In order to measure the PFE of a given fabric sample, the latex particle aerosol is first transferred through the control room and then directed to the particle analyzer. Continuously measure the concentration of three particles in rapid succession, each lasting one minute. The particle analyzer reports the time average concentration of particles during the analysis, that is, the average concentration of particles in one minute (28.3 L) of the sample. After taking these baseline measurements to establish a stable particle count and gas flow rate, the aerosol is transferred to the sample chamber. Once the system reaches equilibrium (usually 60-90 seconds), another three consecutive one-minute measurements are taken in rapid succession. These sample measurements represent the concentration of particles passing through the fabric sample. Subsequently, by splitting the aerosol flow back to the control room, another three particle concentration measurements were taken from the control room to verify that the upstream particle concentration did not change substantially during the entire sample evaluation process. Since the design of the two chambers is the same—except that the sample chamber can accommodate the sample holder—the flow conditions in the chamber can be considered the same, so the concentration of particles in the gas leaving the control chamber and the sample chamber can be compared.
In order to maintain the life of the particle analyzer instrument and remove the aerosol particles in the system between each test, use a HEPA filtered air jet to clean the particle analyzer after each measurement, and clean the sample chamber before changing samples. Please refer to Figure S1 in the support information for a schematic diagram of the air flushing system on the PFE device.
This calculation represents a single “repeated” PFE measurement for a single material sample and is equivalent to the PFE calculation in ASTM F2299 (Equation (2)).
The materials outlined in §2.1 were challenged with latex aerosols using the PFE equipment described in §2.3 to determine their suitability as mask materials. Figure 2 shows the readings obtained from the particle concentration analyzer, and the PFE values of sweater fabrics and batting materials are measured at the same time. Three sample analyses were performed for a total of two materials and six repetitions. Obviously, the first reading in a set of three readings (shaded with a lighter color) is usually different from the other two readings. For example, the first reading differs from the average of the other two readings in the 12-15 triples in Figure 2 by more than 5%. This observation is related to the balance of aerosol-containing air flowing through the particle analyzer. As discussed in Materials and Methods, the equilibrium readings (second and third control and sample readings) were used to calculate the PFE in dark blue and red shades in Figure 2, respectively. Overall, the average PFE value of the three replicates is 78% ± 2% for sweater fabric and 74% ± 2% for cotton batting material.
To benchmark the performance of the system, ASTM 2100 certified medical masks (L2, L3) and NIOSH respirators (N95) were also evaluated. The ASTM F2100 standard sets the sub-micron particle filtration efficiency of 0.1 µm particles of level 2 and level 3 masks to be ≥ 95% and ≥ 98%, respectively. [5] Similarly, NIOSH-certified N95 respirators must show a filtration efficiency of ≥95% for atomized NaCl nanoparticles with an average diameter of 0.075 µm. [24] Rengasamy et al. According to reports, similar N95 masks show a PFE value of 99.84%–99.98%, [25] Zangmeister et al. According to reports, their N95 produces a minimum filtration efficiency of greater than 99.9%, [14] while Joo et al. According to reports, 3M N95 masks produced 99% of PFE (300 nm particles), [16] and Hao et al. The reported N95 PFE (300 nm particles) is 94.4%. [17] For the two N95 masks challenged by Shakya et al. with 0.1 µm latex balls, the PFE dropped roughly between 80% and 100%. [19] When Lu et al. Using latex balls of the same size to evaluate N95 masks, the average PFE is reported to be 93.8%. [20] The results obtained using the equipment described in this work show that the PFE of the N95 mask is 99.2 ± 0.1%, which is in good agreement with most previous studies.
Surgical masks have also been tested in several studies. The surgical masks of Hao et al. showed a PFE (300 nm particles) of 73.4%, [17] while the three surgical masks tested by Drewnick et al. The PFE produced ranges from approximately 60% to almost 100%. [15] (The latter mask may be a certified model.) However, Zangmeister et al. According to reports, the minimum filtration efficiency of the two surgical masks tested is only slightly higher than 30%, [14] far lower than the surgical masks tested in this study. Similarly, the “blue surgical mask” tested by Joo et al. Prove that PFE (300 nm particles) is only 22%. [16] Shakya et al. reported that the PFE of surgical masks (using 0.1 µm latex particles) decreased roughly by 60-80%. [19] Using latex balls of the same size, Lu et al.’s surgical mask produced an average PFE result of 80.2%. [20] In comparison, the PFE of our L2 mask is 94.2 ± 0.6%, and the PFE of the L3 mask is 94.9 ± 0.3%. Although these PFEs surpass many PFEs in the literature, we must note that there is almost no certification level mentioned in the previous research, and our surgical masks have obtained level 2 and level 3 certification.
In the same way that the candidate mask materials in Figure 2 were analyzed, three tests were performed on the other six materials to determine their suitability in the mask and demonstrate the operation of the PFE device. Figure 3 plots the PFE values of all tested materials and compares them with the PFE values obtained by evaluating certified L3 and N95 mask materials. From the 11 masks/candidate mask materials selected for this work, a wide range of PFE performance can be clearly seen, ranging from ≈10% to close to 100%, consistent with other studies, [8, 9, 15] and industry descriptors There is no clear relationship between PFE and PFE. For example, materials with similar composition (two 100% cotton samples and cotton muslin) exhibit very different PFE values (14%, 54%, and 13%, respectively). But it is essential that low performance (for example, 100% cotton A; PFE ≈ 14%), medium performance (for example, 70%/30% cotton/polyester blend; PFE ≈ 49%) and high performance (for example, sweater Fabric; PFE ≈ 78%) The fabric can be clearly identified using the PFE equipment described in this work. Especially sweater fabrics and cotton batting materials performed very well, with PFEs ranging from 70% to 80%. Such high-performance materials can be identified and analyzed in more detail to understand the characteristics that contribute to their high filtration performance. However, we want to remind that because the PFE results of materials with similar industry descriptions (ie cotton materials) are very different, these data do not indicate which materials are widely useful for cloth masks, and we do not intend to infer the properties-material categories. The performance relationship. We provide specific examples to demonstrate calibration, show that the measurement covers the entire range of possible filtration efficiency, and give the size of the measurement error.
We obtained these PFE results to prove that our equipment has a wide range of measurement capabilities, low error, and compared with data obtained in the literature. For example, Zangmeister et al. The PFE results of several woven cotton fabrics (eg “Cotton 1-11″) (89 to 812 threads per inch) are reported. In 9 of the 11 materials, the “minimum filtration efficiency” ranges from 0% to 25%; the PFE of the other two materials is about 32%. [14] Similarly, Konda et al. The PFE data of two cotton fabrics (80 and 600 TPI; 153 and 152 gm-2) are reported. The PFE ranges from 7% to 36% and 65% to 85%, respectively. In the study of Drewnick et al., in single-layer cotton fabrics (ie cotton, cotton knit, moleton; 139–265 TPI; 80–140 gm–2), the range of material PFE is about 10% to 30%. In the study of Joo et al., their 100% cotton material has a PFE of 8% (300 nm particles). Bagheri et al. used polystyrene latex particles of 0.3 to 0.5 µm. The PFE of six cotton materials (120-200 TPI; 136-237 gm-2) was measured, ranging from 0% to 20%. [18] Therefore, most of these materials are in good agreement with the PFE results of our three cotton fabrics (ie Veratex Muslin CT, Fabric Store Cottons A and B), and their average filtration efficiency is 13%, 14% and respectively. 54%. These results indicate that there are large differences between cotton materials and that the material properties that lead to high PFE (ie Konda et al.’s 600 TPI cotton; our cotton B) are poorly understood.
When making these comparisons, we do admit that it is difficult to find materials tested in the literature that have the same characteristics (ie, material composition, weaving and knitting, TPI, weight, etc.) with the materials tested in this study, and therefore cannot be directly compared. In addition, the differences in the instruments used by the authors and the lack of standardization make it difficult to make good comparisons. Nevertheless, it is clear that the performance/performance relationship of ordinary fabrics is not well understood. The materials will be further tested with standardized, flexible and reliable equipment (such as the equipment described in this work) to determine these relationships.
Although there is a total statistical error (0-5%) between a single replicate (0-4%) and the samples analyzed in triplicate, the equipment proposed in this work proved to be an effective tool for testing PFE of various materials. Ordinary fabrics to certifiable medical masks. It is worth noting that among the 11 materials tested for Figure 3, the propagation error σprop exceeds the standard deviation between the PFE measurements of a single sample, that is, the σsd of 9 out of 11 materials; these two exceptions occur in Very high PFE value (ie L2 and L3 mask). Although the results presented by Rengasamy et al. Showing that the difference between repeated samples is small (ie, five repetitions <0.29%), [25] they studied materials with high known filtering properties designed specifically for mask manufacturing: the material itself may be more uniform, and the test is also This area of the PFE range may be more consistent. Overall, the results obtained using our equipment are consistent with the PFE data and certification standards obtained by other researchers.
Although PFE is an important indicator to measure the performance of a mask, at this point we must remind readers that a comprehensive analysis of future mask materials must consider other factors, that is, material permeability (that is, through pressure drop or differential pressure test). There are regulations in ASTM F2100 and F3502. Acceptable breathability is essential for the comfort of the wearer and preventing leakage of the mask edge during breathing. Since the PFE and air permeability of many common materials are usually inversely proportional, the pressure drop measurement should be performed together with the PFE measurement to more fully evaluate the performance of the mask material.
We recommend that guidelines for constructing PFE equipment in accordance with ASTM F2299 are essential for continuous improvement of standards, generation of research data that can be compared between research laboratories, and enhancement of aerosol filtration. Only rely on the NIOSH (or F3502) standard, which specifies a single device (TSI 8130A) and restricts researchers from purchasing turnkey devices (for example, TSI systems). Reliance on standardized systems such as TSI 8130A is important for current standard certification, but it limits the development of masks, respirators, and other aerosol filtration technologies that run counter to research progress. It is worth noting that the NIOSH standard was developed as a method for testing respirators under the harsh conditions expected when this equipment is needed, but in contrast, surgical masks are tested by ASTM F2100/F2299 methods . The shape and style of community masks are more like surgical masks, which does not mean that they have excellent filtration efficiency performance like N95. If surgical masks are still evaluated in accordance with ASTM F2100/F2299, ordinary fabrics should be analyzed using a method closer to ASTM F2100/F2299. In addition, ASTM F2299 allows for additional flexibility in different parameters (such as air flow rate and surface velocity in filtration efficiency studies), which may make it an approximate superior standard in a research environment.
Post time: Aug-30-2021