A team of researchers at the Massachusetts Institute of Technology (MIT)’s Auto-ID Lab are developing a simple RFID-based sensor that could provide a cost-effective method of detecting the exposure of vaccines to freezing temperatures. The accidental freezing of vaccine shipments is a large problem, with the majority of shipments being exposed to freezing temperatures at some point, according to one study conducted by scientists at the University of Kansas and another study performed by researchers at PATH, a nonprofit organization focused on health-care innovation. Such exposure can result in potency loss for freeze-sensitive vaccines, the scientists found, yet this issue has not been well addressed.
The MIT team, consisting of Professor Sanjay Sarma, postdoctoral associates Rahul Bhattacharyya and Isaac Ehrenberg, and graduate students Dylan Erb and Partha Bhattacharjee, has been working to find a solution to this problem. In a poster presented at the 2015 IEEE International Conference on RFID, the researchers wrote that their sensor, as a simple and inexpensive method of detecting freezing, has the potential to address a significant need “across all segments of the global supply chain.”
The idea, according to the researchers, is to integrate a passive ultrahigh-frequency (UHF) RFID tag with a glass vial (similar to the vials used to hold actual vaccines) that would be filled with distilled water, so that as the liquid freezes, the vial would deform and eventually crack. If a tag were mounted on this vial, the deformation could change the antenna’s shape and eventually damage the antenna enough that the tag would be rendered unreadable. One or more of the sensor vials would be inserted into a shipment of vaccines. In that way, the state of the vaccines could easily be monitored at all stages of the supply chain. If any vial tags were found to be unreadable, or if visual inspection revealed broken sensors, health-care workers could then assume that the vaccines contained within had also been frozen and might be damaged or unusable.
The researchers initially experimented with a copper-tape antenna attached to a Higgs3 RFID chip harvested from an Alien Technology passive Squiggle tag, but found that the tape was too strong to be affected by the cracking of the vial. They overcame this problem by employing a technique called argon sputtering to deposit a thin adhesion layer of titanium atoms on the vial’s surface, followed by a film of copper atoms in the pattern of the antenna. The copper film proved conductive enough to allow the tags to be interrogated from a distance of up to 30 centimeters (11.8 inches), yet flexible enough to easily deform along with the vial’s surface. When vials froze, the team found, their antennas were sufficiently mangled that the tags could not be read at any distance.
The team performed a couple of tests by placing sensors in a household freezer fitted with the antenna cabled to an Impinj Speedway Revolution R420 RFID reader. For one test, the group kept the freezer’s door closed so that the temperature within would remain fairly constant. For the other, the freezer door was opened and closed at intervals to mimic the vagaries of conditions in the field. In the latter case, the amount of time that the vials took to break was longer and more variable, yet in both scenarios, the reader could consistently distinguish frozen from unharmed sensors.
The researchers plan to continue their experiments to refine the technology further, Bhattacharyya says, adding that he and his colleagues will work to improve the consistency and robustness of the sensor, as well as experiment with various vial designs. “What we want to do now is refine it so that we can say accurately what temperatures it was exposed to,” he states. If researchers changed the composition of the fluids within the vials, he notes, it could allow them to customize the temperature range, since liquids with different chemical compositions would freeze at different temperatures.
The group is also considering the use of vials made of advanced materials that could improve the sensors’ reliability and precision—for example, by deforming more quickly below the fluid’s freezing point. Another important goal of their research, Bhattacharjee says, is to increase the range at which the sensors are readable, from the present 30 centimeters up to perhaps 1.5 meters (4.9 feet). This would make it more practical to monitor the state of vaccines at all stages of their distribution.
The goal is to improve the technology to the point at which it can be tested under real-world conditions, with a view toward producing a marketable product in the future. The researchers are aiming for a sub-$1 price per sensor, though the exact price will become clearer after field trials have been carried out. Bhattacharyya estimates that the sensors would be ready to begin trials sometime in early to mid-2016.
The MIT Auto-ID lab, Bhattacharjee reports, is having “ongoing discussions with potential partners that undertake technology implementation for vaccine supply chains,” and that have expressed “significant interest” in the sensor. He says he cannot divulge the names of the interested companies, but that he hopes to make a formal announcement within the next six months. The lab attempts to commercialize any promising technologies or inventions, Bhattacharjee adds, and he is optimistic that these sensors will have a large potential market.