Daniel Vargas Ramos loved drinking coffee, energy drinks and tea, but he sometimes had trouble sleeping and wondered how much caffeine he was taking. That was when Haozheng Ma, a graduate student studying biomedical engineering, approached Vargas Ramos with an idea for a caffeine sensor. 

The caffeine sensor that resulted was called DECAF. It looked like a stir bar and could measure the caffeine concentration of any beverage it was dipped into. After stirring the drink with the sensor, the fluid was drawn up to an electrochemical sensor that measured its caffeine concentration. 

Ma and Vargas Ramos, a graduate student studying biomedical engineering, became co-lead authors of the paper published in the journal Small, which described the caffeine sensor they created.

Maral Mousavi, another coauthor on the paper, saw two types of users of this device: those who drink coffee and are conscious about their health and those who want to determine how much caffeine is in a drink for more technical reasons. 

According to the United States Food and Drug Administration, consuming 400mg of caffeine every day is a healthy amount. Mousavi, an assistant professor of biomedical engineering, said the device would help those drinking too much caffeine.

Mousavi also gave the example of testing how coffee varies across different brewing conditions as a use case for the second group.

“They’re these very professional baristas that are really looking into the science of coffee, that would love to know how brewing affects their oxidation of caffeine in the drink,” Mousavi said.

There are existing methods of quantifying caffeine content, but they are time-consuming or use bulky equipment that needs to be calibrated. The device is made up of more than just the final sensor. It also includes components meant to make it robust against varying factors, like acidity or temperature. 

The sensor was tested in various different conditions varying in acidity, temperature and other interfering compounds. The researchers tested how well the sensor worked in different liquid environments, like carbonated drinks or lattes, by adding a known amount of caffeine and using the device to measure the concentrations before and after the change. The sensor seemed to work better than other tests used to find the amount of a substance.

“We’ve actually tested against interfering compounds such as these fake sugars,” Mousavi said. “We tested milk components, which some of them were actually reported to interfere with the prior assays. So far, honestly, we haven’t noticed anything. It works great in beverages, but we would have to do a lot more testing.”

USC has filed a provisional patent for the device. Mousavi said she sees commercialization as a way to reach more people with this product, as the materials used to make the sensor are relatively inexpensive. Even though there are additional costs beyond materials in a commercial context, she said the product can be accessible.

“We want you to be able to just go to the website and order these and just make informed decisions about your health,” Mousavi said.

Vargas Ramos said DECAF’s fluidics component, which controls liquid flow to the electrochemical sensor, could be improved. He said the electrochemical sensor takes around a minute to determine caffeine concentration, but the fluidics component of the device takes five minutes. 

Another way Vargas Ramos said the device can be improved is to develop an app that directly gives someone the information they need about their drink, since the device currently measures caffeine concentration in millimolars.

“The idea would be that the app would have all those mathematical equations already in the algorithm,” Vargas Ramos said. “So, you just need to put the volume, and it’s already calculating the millimolars, so it will do that by itself, showing you the milligram number.”

Michael Chen, a graduate student studying computer science with a game development focus, said he would use a device like DECAF. Chen estimated that he drinks 16 to 30 ounces of coffee every day. He remembered one night he couldn’t sleep until 2 a.m. due to drinking coffee at night. 

“I’m not sure how much caffeine I take per day,” Chen said. “That’s why I want to use this kind of machine, since I don’t know the number.”