Open Source PPE: Inside the Project Decentralizing Face Mask Production

Before the COVID-19 pandemic, the United States federal government gave passing thought to its stockpile of respiratory masks.

Respiratory masks, long downplayed by health and government officials as an effective precautionary measure, are now a critically important tool in stopping the metastasization of COVID-19. Yet, from hospitals to pharmacy shelves, supplies are dwindling. 

“If it were to be a severe event, we would need 3.5 billion N95 respirator masks” to get through the coronavirus crisis, Dr. Robert Kadlec, assistant secretary for preparedness and response, testified before the U.S. Senate Health Committee in late March. At that time, the U.S. had approximately 42 million masks on hand, a fraction of the needed supply, and the situation has hardly improved since. Depending on how you count, the United States is still short between 70 to 100 million face masks per month.

Spurred by federal incentives, domestic manufacturers including Kimberly-Clark and Honeywell have stepped up their disaster preparedness response, with 3M alone pledging to nearly double output to two billion masks per year. But last Tuesday, after President Donald Trump pledged to buy 600 million masks from five domestic producers, at least one said it wouldn’t be able to fill its part of the order until September. Even well-funded conglomerations lack the capacity to ramp up short-run “burst capacity” production on any meaningful scale. 

That’s where the Open PPE Project comes in. A group of about 15 technologists and engineers based in the Midwest is forging a way to spin up mask-making facilities across the country. The idea is to give manufacturing capability to individual hospitals, towns and states, thereby alleviating the constraints of centralized production. 

There are supply chain and regulatory complexities that are obstructing other projects.

Beginning with their own factory churning out N95-like masks in the midwest, the group hopes its open-source model can be adopted by others, when the need strikes. Their plan includes tracking down raw materials with resilient domestic supply chains, designing a functional mask fit for mass production and reuse, and going through legal review so distribution is never cut off by the state. 

“There are supply chain and regulatory complexities that are obstructing other projects,” Matt Parlmer, brainchild of the Open PPE Project and CEO of applied computer science company Ohlogen, said. “We’re doing our best to essentially be the wedge in that crack and drive it open so as many people as possible can start manufacturing masks.”

Open sourcing

Parlmer has taken to giving daily updates on the project, so other organizations can know what hurdles stand in their way. 

“We realized from a structural perspective masks are not that complicated of a product. The really complicated part is handled by upstream manufacturers who make the filter material,” he said. This material, sometimes referred to as N95 Meltblown, is a type of polypropylene thermoplastic made from byproducts of the petrochemical industry. 

It normally costs pennies on the kilogram, but state and corporate buying has driven up the cost on the open market. That has forced Parlmer into MacGyver ways to limit wasted materials. Teaming up with the thousands of engineers and coders volunteering their time and expertise at Helpful Engineering, Parlmer and his team have designed an N95-equivalent mask that requires less filter material. Something as simple as making filter squares rather than circles significantly reduces the amount of plastic left on the cutting room floor, he said. 

By crowdsourcing designs, the team had the flexibility to go back to first principles. The Open PPE mask will be reusable, allowing people to swap out filters as needed, rather than tossing out or having to sterilize the entire product. Further, while most N95s are machined using “complex thermoforming or ultrasonic tools,” these masks are simple enough to be churned out by any factory. “You might be able to find that production capability in the United States or in Germany over the course of a year, but you’re not gonna find that in Uganda,” Parlmer said. 

While the Open PPE Project is beginning with local, autarkic production, the ultimate aim is to be able to ship this design around the world. Parlmer thinks it’s particularly insidious how a few superstates are able to buy out entire cargo ships’ worth of masks, leaving poorer nations completely exposed to the virus. “I’m a dyed-in-the-wool globalist,” he said, “but I think that the first world has a responsibility to not drive up the price for disposable PPE that could otherwise be going to the third world.” 

“Every mask that we bring into the United States is a mask that isn’t going on the face of somebody [less fortunate],” he said. 

Further, importing masks or raw materials is a dangerous game to play for Americans. At this point, the majority of melt-bent plastics is imported from China, just as American pharmaceuticals are said to be overly reliant on Chinese medicinal precursors. “We’re one bad tweet from losing PRC-based mass production,” Parlmer said.

Domestic hurdles 

Domestic production comes with its own hurdles. N95 masks are not necessarily brand-name goods, but a Centers for Disease Control designation for respiratory equipment that can filter out 95 percent of particulate matter. In trying to go through legal review, the Open PPE Project has come up against the “licensing raj” that is the United States medical review. 

Although the team is planning to make a similar product, using the same machines as 3M or Kimberly-Clark, the National Institute for Occupational Safety and Health (NIOSH), a branch of the CDC, told Parlmer it would take 45 to 90 days to approve the team’s factory. That is, as one oblique commenter noted, as “ER docs are making their own masks out of bandanas.”

“Our goal it to start producing as many masks as possible as quickly as possible to get them on people’s faces,” Parlmer said. A number of law firms, including Varnum LLP, are providing pro bono assistance to the team as they try to get emergency exceptions from onerous regulatory oversight. 

“This is not an easy task, but it’s not like making microchips or cars or airplanes,” Parlmer said. “Essentially, we’re trying to delegate the actual first principles of the safety and testing review to laboratories and qualified universities.” His hope is that once their design is standardized and approved, other “mask fabs” following their model will be able to immediately start producing. 

While the team has hit hiccups along the way, over the course of a month they’ve made significant in-roads, with Parlmer saying they are now in conversation with CDC and NIOSH. 

For his part, Parlmer was driven to action after hearing his anesthesiologist father was given one mask by his hospital’s administration and told to “use until it fell apart,” he said. “For context, pre-crisis protocols dictate that you swap out masks in between patients.” This dangerous shift in hospital protocols could eventually leave the country without enough medical experts to care for the ill. 

“The need is not only acute for medical professionals, but for anyone riding the subway. It would be a profound injustice if respiratory PPE became some sort of precious scarce thing.”