In 2020, while looking at data from a local air quality sensor, Assistant Professor David Burnett of Portland State University (PSU) saw something interesting: A sudden large spike in carbon monoxide levels.
What did it mean?
"A gas bubble was coming through," Burnett explains, likely related to the wildfire smoke that was overwhelming the region at the time. It would have been useful, however, to know more about that bubble.
"How big was it? Where was it headed? Where did it come from?" Burnett said.
Per Environmental Protection Agency (EPA) guidelines, every metropolitan area in the United States of a certain size needs at least one environmental air quality sensor. Portland's is managed by the Oregon Department of Environmental Quality (DEQ), and is what Burnett was using that day. It detects particulate matter, ozone, and air pollutants including carbon monoxide, nitrogen dioxide and sulfur dioxide. While the DEQ's data are useful for monitoring our region's air quality, the single sensor is of limited use in determining trends and investigating causes—something which is of great interest to climate scientists.
The More Data, The Better
The single DEQ sensor isn't the only game in town. PurpleAir, a technology company, sells air quality sensors to individuals who then opt into their crowdsourced data system. Once installed in private homes, the sensors measure particulate matter (PM) in the air and upload their data to the web. The community data of PurpleAir users can offer a more comprehensive picture of the region's air quality, but this still has limitations.
For one, the people buying PurpleAir sensors tend to be more affluent, which can perpetuate environmental injustice through a lack of data for lower-income neighborhoods. For another, the sensors require a power source. Even if every single household had a PurpleAir, we still wouldn't have data for our deserts, forests and parks—natural areas where people don't live.
"Environmental sensing ideally happens at a high spatial resolution. What would be more beneficial to scientists is to have a grid of sensors," Burnett said. This would enable researchers to track changes and movements over a large area of land, or even the ocean with pH and water quality sensors.
An assistant professor of Electrical & Computer Engineering in the Maseeh College, Burnett's research focuses on environmental sensing and monitoring through the use of electronic devices, and he hopes one day to deploy enough sensors to get granular, high-resolution data for the entire region.
Undergraduate Students Built Their Own Environmental Sensors
As an advisor to undergraduate engineering students, Burnett encourages them to be creative with off-the-shelf components. Shown here on the left is a student-constructed sensor made as part of a senior capstone project that measures indoor air quality. Mounted on the wall of the Wireless Environmental Sensor Technology (WEST) Lab, the shoebox-sized sensor has been in continuous operation since June 2023.
"Students can make a device like this for $100 using commercial parts. But an environmental sensing microchip could be mass produced for pennies, once the design is perfected," Burnett said.
Working with Intel, TSMC (Taiwan Semiconductor Manufacturing Company), and other industry partners, along with academic collaborators like UC Berkeley, UW's Washington Nanofabrication Facility, and Inria Paris, Burnett has designed several iterations of a semiconductor-based wireless sensor: A miniscule chip, able to detect different gases and particulate matter, that will be attached to an antenna and a small solar panel. The antenna and special network protocols allow the chips to relay signals to each other, so as long as they are within communication range of the nearest fellow chip, a computer located in a research center would eventually receive information from all of them.
Since these are sealed in inert materials that do not interact with the environment, it would be affordable and beneficial to place them everywhere. One delivery method that Burnett's lab has considered is attaching the chips to small, 3D printed replicas of a maple seed made using biodegradable materials.
When maple seeds detach from the tree, their unique shape causes them to spin rapidly as they descend (that's why they're sometimes called whirlybirds or helicopter seeds). The movement generates lift, slowing the seed's descent and allowing it to stay airborne longer. The extended time aloft increases the likelihood of wind carrying the seeds farther from the parent tree, facilitating effective seed dispersal and reducing competition for resources among seedlings.
This same principle could enable a drone to release a large amount of semiconductor-laden whirlybirds over, say, a national forest, and let them disperse evenly throughout the area.
Looking To The Future
Providing scientists with a wealth of data wouldn't be the only benefit of having a multitude of tiny sensors in place: They could also help protect nearby people who have smartphones or other Bluetooth devices.
"Having the chips mounted all over a metro area could serve as an early warning system, alerting nearby Bluetooth devices if the air is dangerous," Burnett said.
While this isn't a reality yet, collaborations within PSU and with community partners in the region are ongoing, making this and other applications a possibility for the future. By developing the cost-effective, wireless environmental sensors, Burnett and his team are taking a key first step toward creating a comprehensive air quality monitoring network—not just for the Portland region, but around the world.
To learn more about this research, visit Burnett's website or explore his publications.