Low-cost measurement of face mask efficacy for filtering expelled droplets during speech
Abstract
Mandates for mask use in public during the recent coronavirus disease 2019 (COVID-19) pandemic, worsened by global shortage of commercial supplies, have led to widespread use of homemade masks and mask alternatives. It is assumed that wearing such masks reduces the likelihood for an infected person to spread the disease, but many of these mask designs have not been tested in practice. We have demonstrated a simple optical measurement method to evaluate the efficacy of masks to reduce the transmission of respiratory droplets during regular speech. In proof-of-principle studies, we compared a variety of commonly available mask types and observed that some mask types approach the performance of standard surgical masks, while some mask alternatives, such as neck gaiters or bandanas, offer very little protection. Our measurement setup is inexpensive and can be built and operated by nonexperts, allowing for rapid evaluation of mask performance during speech, sneezing, or coughing.
INTRODUCTION
The global spread of coronavirus disease 2019 (COVID-19) in early 2020 has substantially increased the demand for face masks around the world while stimulating research about their efficacy. Here, we adapt a recently demonstrated optical imaging approach (1, 2) and highlight stark differences in the effectiveness of different masks and mask alternatives to suppress the spread of respiratory droplets during regular speech.
In general, the term “face mask” governs a wide range of protective equipment with the primary function of reducing the transmission of particles or droplets. The most common application in modern medicine is to provide protection to the wearer (e.g., first responders), but surgical face masks were originally introduced to protect surrounding persons from the wearer, such as protecting patients with open wounds against infectious agents from the surgical team (3) or the persons surrounding a tuberculosis patient from contracting the disease via airborne droplets (4). This latter role has been embraced by multiple governments and regulatory agencies (5), since patients with COVID-19 can be asymptomatic but contagious for many days (6). The premise of protection from infected persons wearing a mask is simple: Wearing a face mask will reduce the spread of respiratory droplets containing viruses. Recent studies suggest that wearing face masks reduces the spread of COVID-19 on a population level and consequently blunts the growth of the epidemic curve (7, 8). Still, determining mask efficacy is a complex topic that is still an active field of research [see, for example, (9)], made even more complicated because the infection pathways for COVID-19 are not yet fully understood and are complicated by many factors such as the route of transmission, correct fit and usage of masks, and environmental variables. From a public policy perspective, shortages in supply for surgical face masks and N95 respirators, as well as concerns about their side effects and the discomfort of prolonged use (10), have led to public use of a variety of solutions that are generally less restrictive (such as homemade cotton masks or bandanas) but usually of unknown efficacy. While some textiles used for mask fabrication have been characterized (11), the performance of actual masks in a practical setting needs to be considered. The work we report here describes a measurement method that can be used to improve evaluation to guide mask selection and purchase decisions.
A schematic and demonstration image are shown in Fig. 1. In brief, an operator wears a face mask and speaks into the direction of an expanded laser beam inside a dark enclosure. Droplets that propagate through the laser beam scatter light, which is recorded with a cell phone camera. A simple computer algorithm is used to count the droplets in the video. The required hardware for these measurements is commonly available; suitable lasers and optical components are accessible in hundreds of research laboratories or can be purchased for less than $200, and a standard cell phone camera can serve as a recording device. The experimental setup is simple and can easily be built and operated by nonexperts.
Fig. 1 Schematic of the experimental setup.
A laser beam is expanded vertically by a cylindrical lens and shined through slits in the enclosure. The camera is located at the back of the box, with a hole for the speaker in the front. The inset shows scattering for water particles from a spray bottle with the front of the box removed. Photo credit: Martin Fischer, Duke University.
Below, we describe the measurement method and demonstrate its capabilities for mask testing. In this application, we do not attempt a comprehensive survey of all possible mask designs or a systematic study of all use cases. We merely demonstrated our method on a variety of commonly available masks and mask alternatives with one speaker, and a subset of these masks were tested with four speakers. Even from these limited demonstration studies, important general characteristics can be extracted by performing a relative comparison between different face masks and their transmission of droplets.
RESULTS
We tested 14 commonly available masks or mask alternatives, one patch of mask material, and a professionally fit-tested N95 mask (see Fig. 2 and Table 1 for details). For reference, we recorded control trials where the speaker wore no protective mask or covering. Each test was performed with the same protocol. The camera was used to record a video of approximately 40 s length to record droplets emitted while speaking. The first 10 s of the video serve as baseline. In the next 10 s, the mask wearer repeated the sentence “Stay healthy, people” five times (speech), after which the camera continued to record for an additional 20 s (observation). For each mask and for the control trial, this protocol was repeated 10 times. We used a computer algorithm (see Materials and Methods) to count the number of particles within each video.
Fig. 2 Pictures of face masks under investigation.
We tested 14 different face masks or mask alternatives and one mask material. Photo credit: Emma Fischer, Duke University. For photos showing the masks as actually worn, see fig. S8 (Supplementary Materials).
| Mask, name | Description |
|---|---|
| 1, “Surgical”* | Surgical mask, three layers |
| 2, “Valved N95” | N95 mask with exhalation valve |
| 3, “Knitted” | Knitted mask |
| 4, “PolyProp” | Two-layer polypropylene apron mask |
| 5, “Poly/cotton” | Cotton-polypropylene-cotton mask |
| 6, “MaxAT” | One-layer Maxima AT mask |
| 7, “Cotton2” | Two-layer cotton, pleated style mask |
| 8, “Cotton4” | Two-layer cotton, Olson style mask |
| 9, “Cotton3” | Two-layer cotton, pleated style mask |
| 10, “Cotton1” | One-layer cotton, pleated style mask |
| 11, “Neck Gaiter” | One-layer polyester/spandex, 0.022 g/cm2 |
| 12, “Bandana”* | Double-layer bandana, 0.014 g/cm2 |
| 13, “Cotton5”* | Two-layer cotton, pleated style mask |
| 14, “Fitted N95” | N95 mask, no exhalation valve, fitted |
| “Swath” | Swath of mask material, polypropylene |
| “None”* | Control experiment, no mask |
Table 1 Face masks under investigation.
This table lists the investigated face masks, mask alternatives, and mask material (masks are depicted in Fig. 2). Masks marked with an asterisk (*) were tested by four speakers; all others were tested by one speaker.
The results of our mask study is depicted in Fig. 3A, where we show the relative droplet count for each tested mask. Data displayed with solid dots represent the outcome of the same speaker testing all masks; the points and error bars represent the mean value and distribution SD, respectively, of the total droplet count normalized to the control trial (no mask). For this speaker’s control trial, the absolute droplet count was about 960. A graph with a corresponding logarithmic scale can be found in fig. S1. Data in Fig. 3A shown with a hollow circle represent an average over four different speakers wearing the same type of masks (surgical, cotton5, and bandana); the values and error bars represent the mean value and SD of the average relative droplet count from all four speakers. The additional speakers’ reference counts for the control trial (no mask) were about 200, with similar fractional variance to the main speaker (see fig. S2 for details).
Fig. 3 Droplet transmission through face masks.
(A) Relative droplet transmission through the corresponding mask. Each solid data point represents the mean and SD over 10 trials for the same mask, normalized to the control trial (no mask), and tested by one speaker. Hollow data points are the mean and SDs of the relative counts over four speakers. A plot with a logarithmic scale is shown in fig. S1. The numbers on the x-axis labels correspond to the mask numbers in Fig. 2 and Table 1. (B) The time evolution of the droplet count (left axis) is shown for representative examples, marked with the corresponding color in (A): no mask (green), bandana (red), cotton mask (orange), and surgical (blue, not visible on this scale). The cumulative droplet count for these cases is also shown (right axis). t, time.
We measured a droplet transmission fraction ranging from below 0.1% (fitted N95 mask) to 110% (neck gaiter, see discussion below) relative to the control trials. In Fig. 3B, the time evolution of detected droplets is shown for four representative examples (surgical, cotton5, bandana, and the control trial) tested by the first speaker—the data for all tested masks are shown in fig. S3. Solid curves indicate the droplet transmission rate over time. For the control trial (green curve), the five distinct peaks correspond to the five repetitions of the operator speaking. In the case of speaking through a mask, there is a physical barrier, which results in a reduction of transmitted droplets and a significant delay between speaking and detecting particles. In effect, the mask acts as a temporal low-pass filter, smoothens the droplet rate over time, and reduces the overall transmission. For the bandana (red curve), the droplet rate is merely reduced by a factor of 2, and the repetitions of the speech are still noticeable. The effect of the cotton mask (orange curve) is much stronger. The speech pattern is no longer recognizable, and most of the droplets, compared to the control trial, are removed. The curve for the surgical mask is not visible on this scale. The shaded areas for all curves display the cumulative particle count over time: The lower the curve, the more droplets are blocked by the mask. Figure 3B shows the droplet count for the four masks measured by one speaker; fig. S4 shows the data for all four speakers using identical masks.
We noticed that speaking through some masks (particularly the neck gaiter) seemed to disperse the largest droplets into a multitude of smaller droplets (see fig. S5), which explains the apparent increase in droplet count relative to no mask in that case. Considering that smaller particles are airborne longer than large droplets (larger droplets sink faster), the use of such a mask might be counterproductive. Furthermore, the performance of the valved N95 mask is likely affected by the exhalation valve, which opens for strong outwards airflow. While the valve does not compromise the protection of the wearer, it can decrease the protection of persons surrounding the wearer. In comparison, the performance of the fitted, non-valved N95 mask was far superior.
DISCUSSION
The experimental setup is very straightforward to implement, and the required hardware and software are ubiquitous or easily acquired. However, this simplicity does go along with some limitations that are discussed here, along with routes for possible improvements and future studies. Again, we want to note that the mask tests performed here (one speaker for all masks and four speakers for selected masks) should serve only as a demonstration. Intersubject variations are to be expected, for example, because of differences in physiology, mask fit, head position, speech pattern, and such.
A first limitation is that our experimental implementation samples only a small part of the enclosure, and hence, some droplets that are transmitted through the masks might not be registered in the laser beam. Similarly, the face of the speaker is positioned with respect to the speaker hole by aligning the forehead and chin to the box. The physiology of each speaker is different, resulting in variations of the position of the mouth relative to the light sheet. Hence, the droplet count reflects only a portion of all droplets, but as we perform the experiment with the same initial conditions for all masks, the relative performance of the masks can be compared. A speaker hole that is sealed around the face would prevent the undetected escape of particles and ease comparison between different speakers.
Second, the use of a cell phone camera poses certain limitations on detection sensitivity, i.e., the smallest recognizable droplet size. To estimate the sensitivity, we consider the light that is scattered by droplets passing through the laser beam. The amount of light scattered into the camera direction depends on the wavelength of light, the refractive index of the droplet, and its size (and shape). To estimate the light scattering of droplets into the camera as a function of their diameter, we used the Python package PyMieScatt (12), which is an implementation of the Lorenz-Mie theory [see (13) for a review]. The result is visualized in Fig. 4. Figure 4A shows an example of the scattering distribution for a 532-nm light scattered from a droplet of 5 μm diameter and a refractive index of water (n = 1.33). In this example, the particle size is substantially bigger than the wavelength of the light (the so-called Mie regime). Almost all the light is scattered into the forward direction (0°) and very little into the direction of the camera (indicated by the shaded green cone around 90°). For the given camera acceptance angle, we display in Fig. 4B the estimated number of photons per frame scattered into the cell phone camera aperture as a function of particle diameter. By illuminating the camera directly with an attenuated laser beam of known power, we determine the detection sensitivity. A minimum of about 75 photons (on a single camera pixel) or about 960 photons (spread over several pixels) per frame were required for the camera to detect a droplet (for details on the detection characterization, see the Supplementary Materials). Both detection thresholds are indicated by horizontal black lines and the red shaded area in Fig. 4B. The more conservative detection threshold corresponds to a minimum detectable droplet size of 0.5 μm. The main limitation is the low collection efficiency of our small camera aperture—we currently capture only 0.01% of the full solid angle. An increased collection efficiency is possible with a larger relay lens in front of the camera, but this would come at the cost of a reduced field of view.
Fig. 4 Light scattering properties.
(A) Angle distribution (scattering phase function) for light scattered by a water droplet of 5 μm diameter for illumination with green laser light. Note the logarithmic radial scale. 0° is the forward direction, and 180° is the backward direction. The camera records at around 90°, indicated by the green segment (not to scale). (B) Calculated number of photons recorded by the camera in one frame as a function of the droplet diameter. The red shaded area and the two solid lines indicate the detection thresholds of the camera. For ideal conditions (all photons impinge on a single pixel), the camera requires at least about 75 photons per frame corresponding to a droplet diameter of 0.1 μm; for photons distributed over multiple pixels, the threshold is around 960 photons and corresponds to a diameter of 0.5 μm.
Third, the use of a single cell phone camera also limits the achievable size resolution (currently 120 μm per pixel), given the large field of view that is required to image as many droplets as possible. This makes it unfeasible to directly measure the size of small (aerosol) droplets in our setup. However, while we cannot measure the size of droplets at or below the pixel resolution, we can still detect and count the smaller droplets, down to the sensitivity limit described above. For very large particles, the limited dynamic range of the camera also poses a challenge for determining the size, since pixels easily saturate and hence distort the shape of the recorded droplet. We want to point out that neither the limited pixel resolution nor the saturation affect the particle counts presented in Fig. 3. Choosing a higher quality camera and a smaller field of view, combined with a funnel setup to guide droplets toward the imaging area, would reduce the minimum observable size, so would approaches that use camera arrays to improve resolution without sacrificing sensitivity or field of view (14). Keeping in mind these sizing limitations, we can still estimate the size distribution for the larger droplets (see fig. S5 for a qualitative size plot), which presents some interesting observations such as the neck gaiter performance mentioned earlier.
We should point out that our experiments differ in several ways from the traditional methods for mask validation, such as filtration efficiency of latex particles. As is apparent from the neck gaiter study, liquid filtration (and subsequent particle size reduction) is more relevant than solid filtration. In addition, our method could inform attempts to improve training on proper mask use and help validate approaches to make existing masks reusable.
In summary, our measurements provide a quick and cost-effective way to estimate the efficacy of masks for retaining droplets emitted during speech for droplet sizes larger than 0.5 μm. Our proof-of-principle experiments only involved a small number of speakers, but our setup can serve as a base for future studies with a larger cohort of speakers and checks of mask performance under a variety of conditions that affect the droplet emission rate, like different speakers, volume of speech (15), speech patterns (16), and other effects. This method can also test masks under other conditions, like coughing or sneezing. Improvements to the setup can increase sensitivity, yet testing efficiency during regular breathing will likely require complementing measurements with a conventional particle sizer. A further area of interest is the comparison of mask performance between solid particles and droplets, motivated by the observed liquid droplet breakup in the neck gaiter and mask saturation by droplets, necessitating exchange in regular clinical practice.
MATERIALS AND METHODS
The optical setup we used was recently used to demonstrate expulsion of liquid droplets during speech and for characterization of droplet residence times in air (1, 2). A schematic of the setup is shown in Fig. 1. In short, a light sheet was shined through an enclosure, where light scattering from particles traversing the light sheet was detected with the camera. To form the light sheet, a cylindrical lens transformed a green laser beam into an elliptical profile, which was directed through the enclosure. The laser source was a scientific pump laser (Millennia, Spectra-Physics; power, 2 W; wavelength, 532 nm), but suitable green lasers of similar powers are available for less than U.S. $100; the scientific lasers have better specifications (higher beam pointing and intensity stability, better beam profile), but these advantages are irrelevant in this application. The light sheet at the center of the enclosure had a thickness of 4.4 mm and a vertical size of 78 mm (Gaussian 1/e2 intensity beam widths). The enclosure (length by width by height: 30 cm by 30 cm by 35 cm) was constructed out of (or lined with) black material to minimize stray light. The sides of the box had slits for entry and exit of the light sheet. The front of the box had an 18-cm-diameter hole for the speaker, large enough for a person wearing a mask to speak into the box but small enough to prevent the face (or mask) from reaching the light sheet. To clear droplets from the box between experiments, laminar air from a high efficiency particulate air (HEPA) filter was continuously fed into the box from above through a duct with a cross section of 25 cm by 25 cm. The supplied air was being expelled through the light sheet slits and the speaker hole. A slight positive pressure in the box cleared droplets and prevented dust from entering into the box from outside. On the back of the box, a cell phone (Samsung Galaxy S9) was mounted at a distance of 20 cm from the light sheet. Using the Android application “open camera,” the frame size was set to 1920 × 1080 pixels, the focal distance was set to 20 cm, the exposure time was set to 1/50 s, and the frame rate was set to 30/s. At this focal distance, each camera pixel recorded an area of 120 μm by 120 μm at the position of the light sheet.
For each trial, the camera recorded scattered light from particles in the laser beam before the speech (~10 s), during speech (~10 s), and for a period of droplet clearing (~20 s). The speech consisted of five repetitions of the phrase “Stay healthy, people,” spoken by a male test person with a strong voice but without shouting. Each trial was repeated 10 times, and the speaker drank a sip of water in between to avoid dehydration. Furthermore, for the masks that showed substantial amounts of detected particles (knitted, cotton, neck gaiter, and bandana), we conducted additional tests by repeatedly puffing air from a bulb through the masks, rather than speech from an experimenter. These control trials with air puffs confirmed that we recorded droplets emitted by the speaker, not dust from the masks.
The goal of the analysis is to compare the efficacy of different masks by estimating the total transmitted droplet count. Toward this end, we need to identify droplets in the video and discriminate between droplets and background or noise. For convenience, analysis of the videos was performed with “Mathematica” (Wolfram Research), but use of a commercial package does not pose any general restriction, since almost every high-level programming language (e.g., Python) offers the same functionality. From all videos, we removed a weak background that originated from the light sheet itself and from stray light and diffuse reflections from the experimenter’s face. We then binarized all frames with a common threshold that discriminates between scattered light from droplets and background signal and/or noise. Then, a feature detection algorithm is applied to each frame, which returns the center-of-mass positions and major axis and minor axis length of the best-fit ellipse for every droplet. Note that the major and minor axes returned by the algorithm are not a direct measure of the droplet size but a measurement of the amount of light scattered by the particle into the camera aperture (binary diameter). Furthermore, the major axis length is increased owing to particle motion during the camera exposure time. Because of the small dynamic range of the camera (8-bit), most droplets saturate the camera. However, the axis lengths returned by the algorithm can still be used for a qualitative droplet size estimation: A bigger droplet scatters more light than a smaller droplet. This insight is important to interpret the result of the neck gaiter. The neck gaiter has a larger transmission (110%; see Fig. 3A) than the control trial. We attribute this increase to the neck gaiter dispersing larger droplets into several smaller droplets, therefore increasing the droplet count. The histogram of the binary diameter for the neck gaiter supports this theory (see fig. S5).
If a droplet passes through the light sheet in a time shorter than the inverse frame rate, it will appear only in a single video frame. However, if the droplet spends more time in the light sheet, the droplet will appear in multiple frames. To avoid double counting droplets in consecutive frames, we use a basic algorithm to distinguish between single-frame particles and multiframe trajectories. The algorithm compares the distance between droplets in consecutive frames and assigns two droplets to a trajectory if their distance is smaller than a threshold value or counts them as individual droplets if their distance is larger than the threshold. The threshold value was empirically chosen to be 40 pixels. An example result of the algorithm is shown in fig. S6, which shows a projection of 10 consecutive frames. Every droplet recognized by the algorithm is highlighted by an ellipsoid, labeled with the frame number. Droplets that belong to the same trajectory are highlighted in the same color.
Acknowledgments
We thank Mathias Fischer for providing the sketch in Fig. 1 and S. Eriksson and J. Lindale for valuable discussions. Funding: This project has been made possible in part by grant number 2019-198099 from the Chan Zuckerberg Initiative DAF, an advised fund of the Silicon Valley Community Foundation, and internal funding from Duke University through the Advanced Light Imaging and Spectroscopy (ALIS) facility. Author contributions: M.C.F. and E.P.F. performed the experiments, D.G. performed the data analysis, I.H. and E.W. procured the masks, and W.S.W. provided expertise. M.C.F supervised the project. All authors were involved in data interpretation and manuscript preparation. Competing interests: A U.S. provisional patent application has been filed by Duke University on 12 June 2020. The authors of the current manuscript are identical to the inventors on the patent application. The patent information is as follows: Title: “Optical Method to Test Efficacy of Face Masks”; inventors: M.C.F., E.P.F., D.G., W.S.W., I.H., and E.W.; application number: 63/038331. The authors declare that they have no other competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors. All raw movie files are freely available at the Duke Research Data Repository at https://doi.org/10.7924/r4ww7dx6q.
Supplementary Material
REFERENCES AND NOTES
1
V. Stadnytskyi, C. E. Bax, A. Bax, P. Anfinrud, The airborne lifetime of small speech droplets and their potential importance in SARS-CoV-2 transmission. Proc. Natl. Acad. Sci. U.S.A. 117, 11875–11877 (2020).
2
P. Anfinrud, V. Stadnytskyi, C. E. Bax, A. Bax, Visualizing speech-generated oral fluid droplets with laser light scattering. N. Engl. J. Med. 382, 2061–2063 (2020).
3
N. L. Belkin, The evolution of the surgical mask: Filtering efficiency versus effectiveness. Infect. Control Hosp. Epidemiol. 18, 49–57 (1997).
4
A. S. Dharmadhikari, M. Mphahlele, A. Stoltz, K. Venter, R. Mathebula, T. Masotla, W. Lubbe, M. Pagano, M. First, P. A. Jensen, M. van der Walt, E. A. Nardell, Surgical face masks worn by patients with multidrug-resistant tuberculosis: Impact on infectivity of air on a hospital ward. Am. J. Respir. Crit. Care Med. 185, 1104–1109 (2012).
5
Advice on the use of masks in the context of COVID-19, WHO Reference Number: WHO/2019-nCov/IPC_Masks/2020.4 (World Health Organization, 2020).
6
M. Klompas, C. A. Morris, J. Sinclair, M. Pearson, E. S. Shenoy, Universal masking in hospitals in the Covid-19 era. N. Engl. J. Med. 382, e63 (2020).
7
D. K. Chu, E. A. Akl, S. Duda, K. Solo, S. Yaacoub, H. J. Schünemann; COVID-19 Systematic Urgent Review Group Effort (SURGE) study authors, Physical distancing, face masks, and eye protection to prevent person-to-person transmission of SARS-CoV-2 and COVID-19: A systematic review and meta-analysis. Lancet 395, 1973–1987 (2020).
8
N. H. L. Leung, D. K. W. Chu, E. Y. C. Shiu, K.-H. Chan, J. J. Mc Devitt, B. J. P. Hau, H.-L. Yen, Y. Li, D. K. M. Ip, J. S. M. Peiris, W.-H. Seto, G. M. Leung, D. K. Milton, B. J. Cowling, Respiratory virus shedding in exhaled breath and efficacy of face masks. Nat. Med. 26, 676–680 (2020).
9
D. Bunyan, L. Ritchie, D. Jenkins, J. E. Coia, Respiratory and facial protection: A critical review of recent literature. J. Hosp. Infect. 85, 165–169 (2013).
10
J. J. Y. Ong, C. Bharatendu, Y. Goh, J. Z. Y. Tang, K. W. X. Sooi, Y. L. Tan, B. Y. Q. Tan, H.-L. Teoh, S. T. Ong, D. M. Allen, V. K. Sharma, Headaches associated with personal protective equipment—A cross-sectional study among frontline healthcare workers during COVID-19. Headache 60, 864–877 (2020).
11
A. Konda, A. Prakash, G. A. Moss, M. Schmoldt, G. D. Grant, S. Guha, Aerosol filtration efficiency of common fabrics used in respiratory cloth masks. ACS Nano 14, 6339–6347 (2020).
12
B. J. Sumlin, W. R. Heinson, R. K. Chakrabarty, Retrieving the aerosol complex refractive index using PyMieScatt: A Mie computational package with visualization capabilities. J. Quant. Spectrosc. Radiat. Transfer 205, 127–134 (2018).
13
G. Gouesbet, G. Gréhan, Generalized Lorenz-Mie Theories (Springer International Publishing, ed. 2, 2017).
14
L. Zhong, P. Konda, M. Harfouche, R. Horstmeyer, Imaging and Applied Optics (Optical Society of America, 2019).
15
S. Asadi, A. S. Wexler, C. D. Cappa, S. Barreda, N. M. Bouvier, W. D. Ristenpart, Aerosol emission and superemission during human speech increase with voice loudness. Sci. Rep. 9, 2348 (2019).
16
S. Asadi, A. S. Wexler, C. D. Cappa, S. Barreda, N. M. Bouvier, W. D. Ristenpart, Effect of voicing and articulation manner on aerosol particle emission during human speech. PLOS ONE 15, e0227699 (2020).
Information & Authors
Information
Published In
Science Advances
Volume 6 • Issue 36 • September 2020
History
Received: 12 June 2020
Accepted: 22 July 2020
7 August 2020
Published online: 2 September 2020
Copyright
Copyright © 2020 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution License 4.0 (CC BY).
This is an open-access article distributed under the terms of the Creative Commons Attribution license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Authors
Funding Information
http://dx.doi.org/10.13039/100014989Chan Zuckerberg Initiative: 2019-198099
RE:
In Fig. 3A of the published manuscript we show the relative droplet count for each mask, obtained by counting the number of emitted particles (corrected for the appearance of the same particle in successive video frames) relative to the no-mask reference. A quick visualization (shown in some news reports) can be obtained from a maximum intensity projection over the duration of the experiment, but the processing steps outlined in the manuscript are required for quantification.
Fig. 3 displays the number of particles, irrespective of their size. As detailed in our work, we cannot accurately determine the size of very small particles. In the supplementary materials we show a qualitative size distribution of the larger emitted particles, constrained by our image resolution. Even in this limited size range we see a relative shift of particle size to smaller sizes for some masks, particularly our neck gaiter.
There is an ongoing debate about the contribution of particles originating from the mask itself, generated during speaking. From our studies (including several follow-up experiments) we believe that for our masks the contribution of mask fragments were minor. However, for practical mask use this distinction of origin is not important since viruses can be contained in (or attached to) the particles in either case.
RE: Low-cost measurement of face mask efficacy
I found this research previously with pictures of most if not all of the face masks being tested showing the particles in the air. I've seen a few videos online that briefly show this but it's not complete showing each mask, and it's not images like it was originally. Is it possible to get the pictures showing the snapshots of the fleece gaiter where it showed more particles than the baseline?
I ask because I specifically remember seeing obvious signs of the extra particles for the fleece gaiter being the fibers of the material being released into the air, and not particles from the face or mouth being split into smaller particles. I'd read the paper and seen the pictures before it was picked up by mainstream media. I thought the error being reported in the news would be corrected.
RE: Collective response to comments on "Low-cost measurement of face mask efficacy for filtering expelled droplets during speech"
We appreciate the interest that many people have shown in our work, both here and in emails. Many comments have led us to put some clarifications in the final online version (for example, while “neck gaiter” and “neck fleece” are both terms in common use, the latter term is confusing to many people, so the final version uses “gaiter” only).
We would like to address the comments in the eLetters in broad sweep.
1. As the title indicates, our work focused on developing a simple technique for mask evaluation that can be replicated at other labs, rather than a comprehensive mask test. As we stated in the paper, our work was a preliminary study that included a set of masks worn by one person, and a subset of masks worn by four persons. We did not do a systematic study involving many masks, speakers, and wear conditions - that would have delayed publication by many months. We expect that there are variations of performance for different styles and different materials, as well as different users wearing identical masks. More detailed studies are needed to make specific use recommendations, hence we avoided rank ordering masks other than for clarity in the graphs.
2. We detect particles of size greater than about 0.5 µm, but cannot reliably determine the size of the smallest particles. In addition, our measurement makes no attempt at discriminating particles transmitted through the mask from particles escaping through gaps around the mask. Nor do we make any prediction about how far the particles travel, what their viral load is, or how infectious they are.
3. In this work we demonstrated one neck gaiter only (in the first version of the paper we called it fleece mask, which caused some confusion - we changed that in the final version). The neck gaiter was a 92% polyester / 8% spandex mask (see the photo in the paper - the neck gaiter is number 11 in Fig. 2). The gaiter we tested was single-layer and fairly thin to provide breathability, which is likely the reason for the high particle transmission, possibly as smaller particles originating from larger droplets. We are not implying that droplets are 'sliced up', but rather point to larger droplets as a possible source of mass. We have not tested thicker gaiter material and/or multiple layers, but expect the results to be different, likely better. We believe that for the neck gaiter the material, not the style, is the major determining factor for the measured particle count.
4. This study was a technique development effort, not a systematic mask study. As a demonstration, we utilized masks that we had at hand, some of which home-made, others commercial. Because of the small scale of the demonstrations, we deliberately kept the brand names and model numbers anonymous. For the cotton masks, we do not have access to detailed specifications (such as thread density). In the final version of the paper, we included the material and density of the bandana and neck gaiter (cotton 0.014g/cm2, one-layer polyester/spandex 0.022g/cm2, respectively). While we tested the bandana with four speakers, they were similar in style and material. The N95 with an exhalation valve had no filter inserted in the valve (no information is available on make, model, or approval rating).
5. While Table 1 provides an index connecting the mask names in Fig. 3 to the pictures in Fig. 2, we have clarified this connection in the updated version by providing the image numbers directly in the axis description in Fig. 3.
We greatly appreciate the constructive comments and ideas for future studies (e.g. saturation studies).
RE: Low-cost measurement of facemask efficacy for filtering expelled droplets during speech
No doubt the authors are aware of the importance of encouraging the use of face masks while at the same time publishing research on which masks might be more or less effective. One significant limitation in trying to interpret these results is the difficulty in determining the types of masks/fabrics that were tested. For example, four of the types of masks listed are "knitted, cotton, fleece, and bandana." These four terms refer to four distinct categories that are not mutually exclusive. Knitted is a type of fabric, cotton is a fabric content, fleece is a type of fabric construction and bandana refers to a shape. One could easily construct a fleece bandana out of knitted cotton!
Further, the authors focus on "neck fleece" as being particularly ineffective. Many if not most neck gaiters (which the authors term "neck fleece") are not made of fleece but of some type of polyester jersey microfiber. Was the "neck fleece" that was tested actually fleece, or some other fabric? Within the category of neck gaiters, the variability of available fabrics and potential effectiveness of each (partly depending on density of the weave) makes the information contained confusing.
While the authors say "… we do not attempt a comprehensive survey of all possible mask designs…. We merely demonstrated our method on a variety of commonly available masks and mask alternatives", in future testing more detail and accuracy describing fabric construction, including density of the weave, would be helpful.
Unfortunately, the importance of face coverings has become a contentious issue in this country. The scientific community needs to communicate its guidelines as clearly as possible which will hopefully lead to greater compliance.
RE: Efficiency of fleece gaiters
Do you have any data on fleece gaiters being folded over to create multiple layers?
Your results indicate that fleece gaiters increase the particle count and decrease particle size, but do they affect particle velocity in any way?
RE: Valved masks
I was surprised at how well the valved N95 masks performed, in the chart, it appears to reduce particles by about 85%. compared to no mask, and is in the middle of the pack, 7 masks are worse, some considerably worse, 7 better.
Nevertheless, the article states:
"... the performance of the valved N95 mask is likely affected by the exhalation valve, which opens for strong outwards airflow. While the valve does not compromise the protection of the wearer, it can decrease protection of persons surrounding the wearer. In comparison, the performance of the fitted, non-valved N95 mask was far superior."
Quoted in the Washington Post, one of the authors went further:
"Those relief valves are fantastic if what you want to do is protect yourself from the outside world because air doesn't come in through them," Warren said. "If what you're trying to do in this pandemic is protect the outside world from you, it completely defeats the purpose."
Seems to me that the results in the article don't support these statements. Yes, a valved mask is inferior to an unvalved, well fitting 95 mask. But it's a lot better than nothing, and better than many masks. Moreover, a comfortable mask is more likely to be worn -- how many times do you see people wearing a surgical mask with their nose hanging out?
I own several masks with valves that appear to be designed to direct the airflow down, presumably to avoid fogging glasses, but I would think that would have the beneficial effect of directing particles toward the flow and would reduce dispersion.
Sincerely,
Henry M. Cohn
RE: statements of exhaust valved masks
Hello. Very interesting and compelling research and findings! Several recent media reports have cited this publication, and some mask policy is being shaped partially based on these findings. I am particularly focused on presumptions made regarding "valved" masks, and also the this study's apparent singular focus on emissions only toward the front of the speaker, and not the sides of speakers face. In the Discussion section of this publication, the author's make several definitive statements about the performance of the valved N95 mask "Furthermore, the performance of the valved N95 mask is likely affected by the exhalation valve, which opens for strong outwards airflow. While the valve does not compromise the protection of the wearer, it can decrease protection of persons surrounding the wearer. In comparison, the performance of the fitted, non-valved N95 mask was far superior." Some valved masks incorporate a filter in the exhaust valves - such as the xMask by XSUIT.com. Also, a non optimal fit allowing gaps at the wearer's nose, cheeks or chin would seemingly allow extensive expelled droplets to exhaust - though not necessarily to the front of the speaker. The typical surgical mask specifically is notoriously loose fitting. It would sure seem this is a huge liabaility for moisture exhaust out the sides of the wearer - potentially an enormously greater risk in a side-by-side passenger situation in an airplane or public transportation scenario! Yet, several major airlines have already instituted a mask policy specifically banning ONLY valved masks! I would welcome further discussion on this.
I have participating in research and development of the valved xMask product, which incorporates a removable filter for the exhaust valve.RE: Does test setup measure only in front of face?
Thank you for doing this research. I'd like to see a photograph of the test setup. If I understand how it operates, it only measures the droplets expelled through the front of the mask and passing before the camera. But since most masks do not adhere to the face or fit tightly, most of the air exhaled escapes around the sides of the mask (as anyone who wears glasses can tell you). That means that most of the viral particles spread by an infected person would float into the air even if the mask were 100 percent efficient at trapping particles that pass through it. Can you address this concern? Best, --Dan Littman
Interesting Methodological Article, Uncertain Relevance to COVID-19
While methodologically interesting, it appears that the technique employed by the authors likely reflected air concentrations of relatively large sized droplets (e.g. ca. 100 um) that penetrated the various fabrics. This appears to be the point made by Figure S5, and would explain the very superior performance of the N95 (valve-free). It would be worth determining whether the configuration of the chamber used in the experimental set-up results in passage into the chamber of droplets that escape around the mask seal. The results for the surgical mask suggest that might not be the case. Growing evidence suggests the small respiratory droplet fractions (i.e. << 5 um) expelled by breathing, vocalizing or coughing play an important role in COVID-19 transmission. If the methodology used in this demonstration trial was insensitive to this small droplet fraction, then no conclusions can be drawn from the results with respect to the efficacy of the various common fabric masks in reduction of COVID-19 transmission risk. That requires clarification as there would seem to be considerable potential for misinterpretation of the reported results.
RE: Neck Fleece
What material is the "neck fleece", and how many layers were worn over the speaker's mouth?
RE: Low-cost measurement of facemask efficacy for filtering expelled droplets during speech
I think is is a good article and timely and I think the authors do a good job laying out their methodology and the scope of their findings. But I'm shocked at how it has been spread in the national press and how broad generalizations are being made that the paper itself (and the authors as well I assume) would not support.
I think the take away from this study is just how desperate people are for information and opinions on masks.
Reading the actual study, the authors are very clear that this is a low-budget experimental set up with very few observations. I do not think they really intended it to have a national audience or to have the results generalized to cover public mask wearing. Clearly their finding that a neck gator produces more smaller particles than no mask at all is interesting. But there are a couple of barriers to generalizing the results that the press miss:
what is being measured are particles 1.4 inches out from the mask/speaker. It does not tell you anything about what happens a foot from the mask or 3 feet from the mask, which is where it probably matters for mask use in public.
It only measures the number of particles, not how fast they are going or how far they can go.
They only had 4 different people as test subjects
The statistics for the bandanna and neckgator versus no mask show huge variations in the standard deviation of the measurements. So given very few observations it is unlikely there is much statistical significance.
I am afraid this test will get popularized in people take away the message that no mask is as good or the same as a bandanna or a neck gator, and I'm pretty sure that is not the intent of the authors either. Of course, it shouldn't be the role of a academic journal to highlight disclaimers. But it would sure be nice if journalists and the general public had a better understanding of statistics and and how studies and their results should and should not be used in making public policy.
Its a good time time to write thank yous to your college of high school stats teacher.
RE: Low-cost measurement of facemask efficacy for filtering expelled droplets during speech
Your article is very confusing in its description of the material of the neck gaiter used in the test. In the photo, it looks like a polyester-spandex material, but in the article you call it a "fleece" neck gaiter. There is a huge difference between fleece material and polyester-spandex. "Fleece" material is typically thicker and fuzzier. Neck gaiters for winter are often made of "fleece" material, but neck gaiters for spring, summer and fall are made of lighter weight polyester-spandex material. You really should get a materials expert to provide you with the correct description of the material you used in the tests and then update your article.
There's too much confusing information about this material, especially when articles in newspapers, such as the Washington Post, refer to it as "stretchy polyester-spandex", not fleece.
RE: Low Cost measurment of facemasks, etc
This was a good study, but to be more useful, It would have been better to show each of the 14 masks on the graph so that we could match them up with the picture of masks. I have no idea what mask 1 or mask 2 refers too. A chart on ranking masks along with pictures would be helpful to lay people desperate to protect themselves. All I can tell from here is that N95 and surgical masks are good and fleece is worse than nothing.
RE: Low-cost measurement of facemask efficacy for filtering expelled droplets during speech
Dear Madams and Sirs,
I read your report with great interest, particularly since my "mask" of choice is a collection of neck gaiters. Needless to say, the results pertaining to neck gaiters were very surprising. However, upon further reflection, I am not surprised, yet concerned that the results will be misconstrued.
The results of the study have already been picked up by several media outlets who state that neck gaiters are actually worse than not wearing any masks at all. I very much beg to differ and hope that Duke will issue some clarifying language.
Your study essentially tested the droplet count and size at about 15 cm (roughly 7 inches) from the subject's mouth. However, nowhere in the study is there any mention about the viability or distance of travel of these particles beyond the light sheet in the experiment.
I do not dispute the higher efficacy of certain types of masks over neck gaiters. However, I contend that the distance of travel of droplets when wearing a neck gaiter will be significantly less than when not wearing any face covering at all. So, will there be a higher droplet count using a neck gaiter than without one? Apparently. Will these droplets travel farther when wearing a neck gaiter than without one? I propose that they do not, and I believe that this should be mentioned as a caveat in the discussion area of the study report.
Very sincerely,
Nico Lacchini
RE: Low Cost Filtering of Facemask Efficacy
Thank you for your very interesting study of a very practical and urgent issue to help contain the pandemic . Your study nicely addresses the efficacy of a number of materials as masks. However, whether the position of the wearer's face allows leakage around the edges of the masks to be included in the measurements is unclear to me. My concern about so many masks, from surgical to homemade, including several styles I have made, is the fit and air/droplet/aerosol escape around the mask edges as they are usually worn. I hope that you will be able to address measurement of leakage in further studies. Also useful would be measurements of transmission when the wearer covers only the mouth but not the nose, as is often seen in public places.
RE: Valved N95 Mask
It is not clear from the article if mask #2 ("Valved N95") had its filter media installed correctly or not. This type of mask, fresh out of its package, will likely not have its filter media installed. Without it, it will perform similar to or worse than other cloth masks. So it would be good to know if any mask with replaceable filters had them installed correctly or not, and what filter media was used.
Note that it is easy to accidentally fold the filter media during installation which will greatly reduce the performance of the mask.
RE: Fleece mask
Can you describe the fleece mask? Is it like polar fleece that is warm and designed to allow warm, wet breath to penetrate? Or is it like the sun gators used for fishing and which many people use this summer.
RE: "fleece" material of neck gaiter
Fischer et al. have provided a detailed and thorough analysis of mask efficacy, but a simple piece of information is missing from Table 1. Although the material is identified for some masks, such as #5 ("Poly/Cotton"), it is not identified in Table 1, or elsewhere, for several face coverings, including #11 ("Fleece"). In the table, the description of "Gaiter type neck fleece" is worded such that it indicates the shape of the face covering (a "neck fleece") but not the material. In Figure 1, the garment shown as #11 does not appear to be made of the material commonly referred to as "fleece" (a synthetic, cold-weather fabric, a kind of man-made wool, such as the brand "Polar Fleece").
Rather, #11 appears to be a stretchy, synthetic fabric. "Buff" brand neck gaiters are made of 95% Repreve® Polyester, 5% Elastane ( https://buffusa.com/buff-products/multifunctional-headwear/original/blac... ).
The difference between "fleece" and a stretchy polyester synthetic blend seems critical to the apparent increase in droplet count, even relative to no mask, observed in this experiment.
We suggest that Table 1 be amended to include the type or composition of the material tested for all face coverings. Anecdotally, people engaging in outdoor activities such as running and bicycling have been observed using "Buff" style synthetic material neck gaiters as a way to cover the face when passing by others, then removing the facecovering when once again alone on a trail. Demonstrating that this practice may in fact be counterproductive is an important finding, but without further description of the fabric material, it could be challenging to disseminate the information to the outdoor exercise community.
Note to editors / authors from the author
Admittedly, my degree (Ph.D.) is in Anthropology, not a medical field. I am not seeking publication of any kind for this note but do hope to improve the scientific quality of the article. The question regarding the material of the "fleece" gaiter was brought to my attention by a member of my outdoor trail running community in La Crosse, WI.
RE: gaiter material
Castelli cycling made a gaiter that uses Soft Flex ( their proprietary material) which feels like a synthetic jersey material. No fleece. I would like to know if this fails your test, because I suspect that you'll need to further classify gaiters instead of a general statement.
Thank you.
RE: Mask types
Hi,
As a consumer how are we to determine if a particular mask we see on the internet matches the mask type you are using? You label one mask as 'gaiter type neck fleece', but I'm looking at neck masks made from polyester. Are they in the swath of polypropylene category since they aren't made of fleece?