Filtration Efficiency Assessment of Decontaminated FFP2 Masks for Safe re-Use: Study Conducted as Part of the COVID-19 Response Plan at HASSAN II University Hospital in Morocco

Abstract

During COVID-19, healthcare workers were at risk of infection and needed protection. Unfortunately, crisis-related mask shortages are forcing hospitals to look for ways to reuse masks after decontamination. The study aimed to assess the effectiveness of decontaminated FFP2 masks using moist heat or hydrogen peroxide. It compared the filtration efficiency, chemical composition, and structural changes of these masks with new FFP2 masks. This evaluation was carried out through techniques like scanning electron microscopy, Fourier transform infrared spectroscopy, and thermogravimetric analysis. The finding indicates that the analysis of the chemical composition and structure of the filter media did not exhibit significant alterations or structural deformations. Remarkably, the breathing resistance tests verified that decontaminated FFP2 masks yielded consistent results, differing only by 10 Pa compared to the EN149 standard. Furthermore, the chemical analysis revealed that the filter media in decontaminated masks remained unaltered, with no detectable chemical derivatives found in their components.

1. Introduction

The outbreak of the novel coronavirus disease (COVID-19) pandemic has raised urgent concerns about distributing and using personal protective equipment (PPE) among healthcare workers responsible for patient care. In response to this crisis, the World Health Organization (WHO) has warned of increased demand and potential shortages of PPE. The rapid influx of infected patients and the global impact of the epidemic have placed enormous strains on supply chains, leading to the loss of critical equipment, such as ventilators, and the units needed for patient care, such as FFP2 single-use ventilators. This interruption in the supply chain also affects single-use PPE

FFP2 masks are essential in personal protective equipment in healthcare [1, 2, 3, 4]. These masks are designed to effectively filter particles in the 0.4–0.6 μm range with 94% efficiency. These are usually recommended for respiratory protection [2]. Traditionally, these masks are intended for single use to prevent possible microbial and viral contamination. However, due to the recent COVID-19 pandemic, there has been a significant shortage of these masks, and various decontamination methods are being investigated to assess their potential for reuse. As a result, a rapidly growing need exists to explore the potential for reusing these masks to alleviate mask shortages and optimally protect healthcare workers.

Several efforts have been made globally to address the shortage of FFP2 masks and provide compelling evidence of their potential for safe reuse. Various methods have been documented, including dry and moist heat, hydrogen peroxide vapor, ozone, and UV irradiation. Studies conducted in this area have introduced approaches to assess the filtration efficiency of masks after decontamination according to the established criteria [5, 6, 7]. Nonetheless, there are concerns that these processes may affect the filtration performance of masks, especially when working with hydrogen peroxide vapor [8].

However, implementing a reuse protocol presupposes the availability of a simple, safe reprocessing method adapted to the hospital environment and considering both logistics and human factors. These requirements, especially the availability and simplicity of the supply chain in hospitals in Morocco, led to the selection of reprocessing methods based on moist heat and vaporized hydrogen peroxide, which are most commonly used in the COVID-19 pandemic. This study investigates the structural and chemical changes of FFP2 masks after exposure to the chosen treatment method. Specifically, to mitigate environmental hazards, we tested solely on the new Duckbill brand FFP2 masks widely used in Moroccan hospitals. These masks have undergone up to four reprocessing cycles using each method. The primary purpose of this protocol is to analyze the filtration and mechanical properties of decontaminated FFP2 masks.

2. Materials and method

2.1 Filtering mechanisms

The FFP2 masks can filter out foreign particles by intercepting them within various layers of the mask, using physical mechanisms that can be either mechanical or electrostatic [9]. Particles typically traverse the mask in a laminar flow, allowing the airflow to bypass obstacles such as fibers easily. Mechanical entrapment of particles on fiber surfaces can occur when the inertia of the particles is sufficient to deviate from the aerodynamic trajectory or when suspended particles do not follow the laminar flow and interact with air molecules, resulting in a disordered, random trajectory. This vibratory motion increases the probability that the particles get into the fibers and remain suspended due to Van der Waals’s weak molecular interaction. Generally, this mechanism helps trap airborne particles with a diameter of around 0.1 μm, Figure 1 [10, 11].

Figure 1.

Functional mechanisms of FFP2 filter media.

When the fibers contain significant electrical charges, an electrostatic process can capture the particles, whatever their charge be [11]. During this process, the fibers can attract particles with inherent charges via Coulomb forces and neutral polar particles [e.g., small water droplets] via dielectrophoresis forces. These latter result from the interaction between polarized objects and electric gradients. The filtration efficiency of most FFP2 masks is achieved through a combination of mechanical and electrostatic influences. The assessment of the filtration efficiency of these masks encompasses a series of tests and parameters in line with the established industry standards.

2.2 Measurements of respiratory mask efficiency

Standards provide a set of measurable parameters for assessing the effectiveness of respirators. When examining the reusability of respirators, studies often focus on specific parameters such as breathing resistance, aerosolization tests, and leakage tests. In our research, we have expanded these parameters and have included additional measures commonly assessed in the manufacturing field for this specific type of filter. The parameters considered in this study are.

2.2.1 Measure 1: respiratory resistance

The ISO EN 149 standard incorporates a parameter that includes two essential measurements: inhalation and breathing resistance. This test takes place over 3 minutes and provides respiratory efficiency and resistance values by applying two separate flow rates. The first, simulating inhalation, was set at 95 l/min and 0.2 m/s, while the second, simulating expiration, was set at 160 l/min and 0.34 m/s.

2.2.2 Measure 2: the lifetime and respiratory comfort

This parameter is mainly related to the effectiveness of the respirator, which is designed to establish a secure seal between the respirator and the wearer’s face. If the seal is inadequate, the contaminated air will take the path of least resistance through the seal face leak. An improper seal will therefore affect the level of protection provided to the wearer. This can affect the product’s life, which is generally estimated to be between 4 and 8 hours [12].

2.2.3 Measure 3: the quality of distribution and the intermingling of fibers

The best-known manufacturing process for FFP2 masks’ nonwoven fabrics is the melting process that gives fine fibers. The fibers’ fineness is usually between 0.6 and 10 μm for the melt-blown and 1 to 50 μm for the spunbond. This parameter is essential to obtain a high filtering efficiency.

2.2.4 Measure 4: chemical structure

Typically, the layers and filter media of the masks under study are predominantly polypropylene. However, subjecting polypropylene to elevated temperatures can lead to material degradation and a decline in its mechanical properties. This study highlighted the formation of oxidation products such as hydroperoxides and various carbonyl products during its degradation. They can influence the filtering performance of our masks, which results in the appearance of derivatives of chemical compounds. In this study, we have proceeded to chemical characterization to explore the changes in the molecular structure of the filter media. It is important to note that some common thermoplastics, such as polypropylene (PP), have a low thermal degradation plateau [13, 14, 15].

2.2.5 Measure 5: moisture saturation

Moisture accumulation reduces the filtering efficiency of FFP2 masks. Therefore, each type of mask has a maximum usage time that depends on the ability of its materials to absorb moisture [16]. In this work, we have considered this parameter, which can affect the filtration efficiency of our masks, mainly as we use a decontamination protocol based on moist heat [13, 14].

2.3 Testing protocol

2.3.1 FFP2 mask decontamination protocols

The masks were split into two groups for testing. The first group underwent moist heat sterilization, while the second group was treated with hydrogen peroxide containing silver ions. Specifically, the first group was subjected to an autoclave cycle at 121°C for 20 minutes, followed by a drying period of 15 minutes. To ensure the sterilization performance validation, we have used multiparametric tests and Bowie–Dick penetration tests as required by international standards and guidelines (SO 17665). The second sample of masks was exposed to a decontamination cycle using hydrogen peroxide doped with 12% silver ions at a rate of 6 ml per m³ for 10 minutes and a contact time of one hour on a 27 m³ enclosure. For this purpose, we have used a rotating nozzle micro-nebulizer to generate particles smaller than 5 μm to ensure smooth penetration.

2.3.2 Samples studied

The samples analyzed are FFP2 (Duckbill)-type masks. The composition of the layers and the filter media is generally based on polypropylene. These masks have been exposed to moist heat sterilization and hydrogen peroxide doped with silver ions. The parts of the samples studied are the inner and outer layers (spunbond) of the mask and the intermediate filter media (melt-blown). The protocol used takes into account a maximum of four reuses of the FFP2 mask. Indeed, the protocol consists of performing decontamination tests on 20 new masks to protect the manipulators during the study. This protocol also includes the analysis of a control sample of FFP2 masks without any exposure to any reprocessing procedure.

2.3.3 Analysis methods

2.3.3.1 Respiratory resistance test

Respiratory resistance testing consists of physical tests to measure pressure according to the standard ISO EN 149 test. This test consists of measuring the resistance to airflow through the whole respirator at various flow rates designed to represent the peak inhalation and exhalation rates of a wearer during work of low to moderate intensity [including peak flow rates of 95 lpm, minute volume of 30 lpm, and 160 lpm, minute volume of 50 lpm]. The maximum permitted breathing resistance is 3.0 mbar. We used a motorized differential pressure sensor adapted to the test forces and standard accessories for fixing and holding FFP2 masks on the test device to perform the test.

2.3.3.2 Scanning electron microscopy (SEM)

The morphological properties, distribution quality, and fibers’ intermingling have been studied by Quanta 200 FEI equipped with EDX probe with software Genesis 2000i and 15 kV accelerating voltage. The samples were mounted on adhesive tape and a water-coated nozzle to improve electrical conductivity. The samples included in this type of analysis are the filter media (melt-blown) and the external and internal layers (spunbond) of the FFP2 masks that are the object of this study.

2.3.3.3 Fourier transforms infrared spectroscopy (FTIR)

The chemical structure changes of the reprocessed masks are characterized by examining the FTIR spectra. These spectra were obtained using Spectrometer FT-IR Nicolet iS50, piloted, and recorded in the frequency range between 450 and 4000 cm⁻¹ using a resolution of 4 cm⁻¹ with a sampling frequency of 16 scans. The samples included in this type of analysis are the filter media of the masks (melt-blown).

2.3.3.4 Thermogravimetric analysis (TGA)

To observe possible moisture saturation in the layers of the FFP2 masks, the thermogravimetric behavior of our samples was studied. This was done on a Q500 TGA device. The samples of approximately 10 mg were heated from 25 to 500°C at a heating rate of 10°C/min under 20 ml/min of nitrogen current. The samples examined for this type of analysis are the filter media of the masks FFP2 (melt-blown).

2.3.4 Analysis interpretation

All of the above tests were carried out on decontaminated and control samples. Using an empirical analytical method, the data acquired were evaluated to study the impact of the decontamination procedure on the filtering efficiency and filter media material (melt-blown, spunbond) of the FFP2 masks.

3. Results

3.1 Respiratory resistance

Respiratory resistance tests (inhalation and exhalation) were carried out on our samples after they had undergone various decontamination cycles to study the filtration efficiency of our masks after exposure to reprocessing processes. The test includes a motorized differential pressure sensor adapted to test forces and standard accessories for fixing and holding FFP2 masks on the test device. We have reported the results in Table 1.

Masks sample	Expiratory resistance [Pa]; 160 l/min–0.34 m/s	Inspiratory resistance [Pa]; 95 l/min–0.2 m/s
Witness FFP2	340	350
H2O2/Ag^+ FFP2	340	340
One steam cycle FFP2	360	320
Two steam cycle FFP2	330	330
Three steam cycle FFP2	330	330
Four steam cycle FFP2	330	330

Table 1.

Respiratory resistance results according to EN149 standard.

These findings show that the breathing resistance of the new FFP2 mask is 340 Pa on exhalation and 350 Pa on inhalation. The deviation from the EN149 standard of −/+ 50 Pa may be due to faulty storage or exposure to humidity or radiation [17, 18]. As these types of masks are electret type of filters, they must be stored under the manufacturer’s recommended conditions. We have already studied the effects of mask storage. It has been reported that for a stock of over 10 years for certain masks, 10% no longer meet the efficiency criteria and are below the standard [19]. However, with this defect, the filtration efficiency is relatively high or consistently above 90%.

For FFP2 masks that underwent a single cycle of hydrogen peroxide and silver ion disinfection or a steam heat autoclave cycle, performing respiratory resistance has remained constant with a slight difference of 10 Pa in comparison with the witness FFP2, a slight difference of 0.02% at inspiration.

FFP2 masks that underwent two to four cycles of disinfection by steam autoclaving showed a decrease in the order of 20 Pa on inspiration and expiration compared to the witness FFP2 masks. The uncertainty on this means of experimentation adopted in our study is +/− 0.01, which leads us to conclude that the proposed reprocessing techniques are acceptable. Indeed, the results obtained on the respiratory resistance test allow the possible reuse of three FFP2 masks according to the EN149 standard [20].

3.1.1 Scanning electron microscopy (SEM)

The quality of distribution and the intermingling of the fibers on the surface is evaluated using Quanta 200 FEI equipped with an EDAX probe and Genesis 2000i software with an acceleration of 15 kV voltage. Sectional views are taken to observe the organization of the fibers on the thickness and the different fiber mixtures in the filtering media (melt-blown) and the external layers of the FFP2 masks (spunbond).

The magnification ranged from x200 to x400. The resolution of the images received is 2048 x 1536 pixels, always the pixel as the conversion unit for measuring the fiber diameter. Many types of FFP2 respirator masks are currently commercialized. These are composed according to a standard manufacturing process of two polypropylene spunbond protective layers and two or more melt-blown filtering layers. Each of these layers will be studied separately.

We have grouped the fiber thickness and diameter data in Table 2 and Figure 2 (SEM). They detail the various operating procedures explained in the method section. The aim here is to get an idea of the structural characteristics of the layers studied and to point out any structural changes in the SEM of the exposed masks compared with the FFP2 control mask.

Masks sample	Layers studied	Thickness [mm]	Fiber diameter [μm]
Witness FFP2	Melt-blown	[20]	[3–5]
	Spunbond	[20]	[15–20]
H2O2/Ag^+ FFP2	Melt-blown	[20]	[3–5]
	Spunbond	[20]	[15–20]
One steam cycle FFP2	Melt-blown	[22]	[4–5]
	Spunbond	[20]	[17–20]
Two steam cycle FFP2	Melt-blown	[22.5]	[4–5]
	Spunbond	[21]	[18–20]
Three steam cycle FFP2	Melt-blown	[22.7]	[4.5–5]
	Spunbond	[22]	[18–20]
Four steam cycle FFP2	Melt-blown	[22.7]	[4.5–5]
	Spunbond	[22.5]	[18.520]

Table 2.

Structural characteristics of FFP2 mask monolayers per reprocessing cycle.

Figure 2.

SEM images of the melt-blown and spunbond composing FFP2 masks per reprocessing cycle.

Melt-blown layers are less compact than spunbond layers in the FFP2 mask manufacturing process, yielding the smaller fibers required for FFP2 mask filtration efficiency [21, 22]. This section observes a change in fiber diameters, especially since the reprocessing method adopted is steam heat. The latter can also influence the compactness of the filter media, which can become clogged due to dryness during autoclaving of FFP2 masks and, subsequently, an increase in pressure drop and the permeance of melt-blown layers [23, 24, 25].

Fiber diameter control seems more challenging to judge from Figure 2, which shows the fiber diameter distribution of the melt-blown and spunbond samples of our FFP2 masks in this study. It offers a wide disparity in fiber diameters within the same sample, particularly for filter media, a feature of melt-blown products [24].

In contrast to the samples exposed to steam heat autoclaving, which increased in diameter from 0.5 to 3.5 μm in the spunbond layers and from 2 to 2.5 mm in thickness, our results reveal that the diameter of the fibers in samples treated with hydrogen peroxide and silver ions remains constant compared to that of the control samples, which can be explained by the saturation of moisture in the samples studied Figure 2 and Table 2.

Results obtained in this way lead us to conclude that our masks have changed little structurally, and the parameters obtained in the respiratory resistance tests confirm this hypothesis since they have decreased, especially for FFP2 masks subjected to steam heat treatment cycles.

3.1.2 Fourier transform infrared spectroscopy (FTIR)

IR absorption spectroscopy aims to determine the difference in chemical structure between reprocessed and control samples, with the protocol characterizing only the melt-blown filter media, by examining changes in molecular structure. Below are the spectrograms illustrating the chemical structure of the samples tested Figure 3.

Figure 3.

FTIR spectra of the melt-blown layers of FFP2 masks per reprocessing cycle.

All observed predominant bands are characteristic isotactic vibrational bands (iPP) [26, 27]. The FTIR spectrum of our samples shows a strong C-H stretching vibration band of the alkane from 2841 to 2955 cm⁻¹, and the C-H bend represents the peak at 1380 and 1454 cm⁻¹.

The emergence of a band at 1725 cm⁻¹ occurred in all samples, and this signal in the spectra reflects a carbonyl group (C = O) derived from the polyester group [27]. This suggests that the structure of the samples analyzed also contains a PET thermoplastic polymer. Tests confirmed the use of a mixture of PP and PET polymers in the melt-blown layers of FFP2 masks as protection against harmful bioaerosol particles [26].

The spectra obtained from the samples shown in Figure 3 were similar; no new peaks were detected, meaning there were no reactions or interactions between the materials during reprocessing our FFP2 masks compared with the control samples.

3.1.3 Thermogravimetric analysis (TGA)

The ATG purpose here is to observe any moisture saturation in the layers of FFP2 masks, mainly as we use a moist heat reprocessing method described above. On the one hand, this parameter can influence the performance of the filter media and, on the other, cause bacterial proliferation during storage in unfavorable environments [10, 28]. This thermogravimetric analysis may also show a cross-linking or decomposition reaction of polypropylene at low temperatures and a degradation of the filtering capacity of our masks exposed to a moist heat reprocessing method.

ATG curves for FFP2 H2O2/Ag+, FFP2 single-cycle, and FFP2 three-cycle steam autoclave showed degradation at 315°C compared to FFP2 Witness and FFP2 four-cycle and FFP2 two-cycle steam autoclave samples, which showed degradation at 360°C.

The impact of moisture saturation on our reprocessed samples was observed for FFP2 masks that were recycled two to four times, lengthening degradation over time compared to the FFP2 control. The breakdown temperature of FFP2 H2O2/Ag + is higher (320°C) than FFP2 Witness, although polypropylene is resistant to hydrogen peroxide at room temperature [29].

We attribute the weight loss in our samples to the depolymerization of the PP structure [30]. The thermogravimetric analysis results confirm that our FFP2 masks give specific stability to the filter layers with moist heat reprocessing with a sterilization plate not exceeding 125°C. At this temperature value, our TGA curves confirm this stability in Figure 4.

Figure 4.

Thermogravimetric analysis curves of the layers composing FFP2 masks per reprocessing cycle.

This result indicates that our mask reprocessing methods do not induce measurable chemical reactions such as cross-linking or polypropylene chain splitting (Table 3).

Masks sample	Respiratory resistance results	Structural characteristics (SEM)		Chemical structure; FTIR	Moisture saturation; TGA	Reuse Interpretation
		Thickness [mm]	Fiber diameter [μm]
H2O2/Ag^+ FFP2	ER/340	[20]	[3–5]	undetected chemical residues	unobserved filter disability	Compliant
	IR/340	[20]	[15–20]
One steam cycle FFP2	ER/360	[22]	[4–5]	undetected chemical residues	unobserved filter disability	Compliant
	IR/320	[20]	[17–20]
Two steam cycle FFP2	ER/330	[22.5]	[4–5]	undetected chemical residues	unobserved filter disability	Compliant
	IR/330	[21]	[18–20]
Three steam cycle FFP2	ER/330	[22.7]	[4.5–5]	undetected chemical residues	unobserved filter disability	Compliant
	IR/330	[22]	[18–20]
Four steam cycle FFP2	ER/330	[22.7]	[4.5–5]	undetected chemical residues	unobserved filter disability	Compliant
	IR/330	[22.5]	[18.5–20]

Table 3.

Reuse compliance according to results obtained of FFP2 mask monolayers per reprocessing cycle.

ER: Expiratory resistance, IR: Inspiratory resistance.

4. Discussion

Within the guidelines for personal protection in healthcare establishments, particularly regarding respiratory protection, the use of FFP2 masks and similar mask categories is traditionally recommended for single use. However, the scarcity of personal protective equipment (PPE) during this pandemic and the need to maintain a safe working environment for healthcare professionals have highlighted the need to look for innovative solutions to PPE shortages.

The reprocessing protocols proposed for FFP2 masks or other respiratory protection still pose problems and technical difficulties linked to logistical conditions and to the limits of risk management in the collection and reprocessing chain [28]. Based on data from the literature and existing resources, it adopted two protocols for reprocessing FFP2 masks in this study. The first involved hydrogen peroxide doped with silver ions, while the second used moist heat, described as a widely recognized technique in hospital sterile processing systems [28].

The work undertaken enabled us to assess the feasibility of reprocessing and reusing FFP2 masks, to manage a shortage. The results obtained during the first reprocessing, based on hydrogen peroxide and silver ions, enabled us to conclude that there was no influence on performing the masks tested. Characterization tests confirm the chemical and structural stability of our samples. The presence of silver ions at low doses seems insignificant to cause an effect on the electrostatic filtration provided by the electrical charge of the cast iron (“electret”) used in this type of mask. Another study using the same reprocessing method concluded that the mechanical integrity and performance of the FFP2 are maintained after exposure to 10 or 20 cycles of hydrogen peroxide [5].

The problem of SARS-CoV inactivation has also been widely documented; indeed, there is a vast scientific record of research into the inactivation of viruses and bacterial spores using hydrogen peroxide [31, 32]. And recently, a report submitted to the US FDA by a Columbus-based company, Battelle, has showed that vaporized hydrogen peroxide (HPV) can safely sterilize N-95 FFRs. In its study, the company used spores of the bacterium Geobacillus Stearothermophilus, as a biological indicator to contaminate N-95 FFRs (contaminated by either liquid droplets or exposure to an aerosol) [5].

Regarding the second method of reprocessing FFP2 masks using moist heat, the results appear reassuring for four reuses. Indeed, no structural changes or deformation of the integrity of our samples on a 121°C sterilization plate was detected. A study carried out under the same conditions confirms these results: an Staphylococcus epidermidis permeability test and filtration tests showed no significant difference before and after five cycles of moist heat decontamination [14, 33]. Another study used moist heat decontamination in real-world testing and reported a moderate influence on integrating the tested masks at 60°C and 80% relative humidity [34]. In comparison, another study revealed that decontamination tests by autoclaving at 134°C in Dutch hospitals resulted in mask deformation and loss of elastic strength [35].

The limitation of this work is that we did not validate the removal of the SARS-CoV-2 virus by autoclaving but relied on multiparameter chemical indicators to show the performance of the sterilization load on already biologically valid autoclaves. However, a recent study has indicated that a sterilization cycle, such as ours (15 minutes at 121°C), removed the SARS-CoV-2 [35].

5. Conclusion

We have extensively tested the reuse of respirator masks during this pandemic period. Various procedures have been proposed, and given the variability of the results obtained, it is likely that the success of decontamination is closely linked to mask models and their intrinsic structure. The results of our experiments show that the proposed reprocessing techniques, based on moist heat or hydrogen peroxide, did not influence the functionality of the Duckbill FFP2 masks tested.

According to our results, in the event of an acute shortage of FFP2 masks, these simple processes, which are available in hospitals, economical, and quick to implement, can guarantee the safe reuse of the type of FPP2 masks tested.

Acknowledgments

We are grateful to all who participated in the success of this work from UEMF and CIF, USMBA of Fez, Morocco, which demonstrates a high level of responsiveness during the SARS-CoV-2 pandemic. We also thank the volunteers who participated in this project.

Conflict of interest

The author[s] declared no potential conflicts of interest concerning this article’s research, authorship, and/or publication.