As a PhD student at Harvard University, engineer Sindy K.Y. Tang studied under famed chemist George M. Whitesides—a pioneer in nanoscience, a field that now informs everything from electronics to medical diagnostics. While Tang was on his team, Whitesides was involved in a DARPA project to find ways of encoding messages in bacteria. In the system he and his colleagues developed, messages could be encoded as dots of bacteria on a plate and decoded by adding a particular chemical agent that, when it met the bacteria, would cause a flourescent glow. The pattern could then be translated to reveal a secret message.
Four years later, Tang is applying that same idea in her lab at Stanford, where she is an assistant professor of mechanical engineering. But instead of sending messages back and forth, she’s using chemistry to spot contaminants in water. When dropped into a stream or well, her device, a prototype that was recently described in the journal Lab on a Chip, produces a barcode that indicates both the concentration and the whereabouts of pollutants, such as lead, in water—no electricity necessary.
The device, which is currently about the size of a pinky finger, facilitates a controlled chemical reaction as it moves through water. The clear silicone housing contains two thin tubes, each filled with a gel compound. One end of each tube connects to a reservoir containing a reactant chemical; the other end is open to the environment, so that water can seep into the device.
The chemical in the reservoir moves through the tubes of gel at a predictable rate. As the device moves down a stream, water flows into the gel from the other side. If the chemical being screened for is present—in this initial case, lead—a reaction takes place, creating an insoluble, visible mark in the tube. Those markings create a barcode that scientists can read to determine the amount and location of lead in a particular water supply.
Tang's team has successfully run tests with two different water samples, both in beakers in her lab. The researchers slowly added lead to the water samples, one from the lab and the other from a water hazard on the Stanford golf course, and then were able to see their additions encoded on the sensor afterward. Before they can test the capsules in the field, however, they will need to set up a way to collect them after deployment. One possible solution would be to add small magnetic particles into the silicone housing and use a magnet to fish them out at the other side.
Right now, the sensor still isn’t very precise. “Our detection limit is very high, so we won’t be able to detect [lead] until it’s already very concentrated,” Tang explains. And its chemistry is only able to detect lead at this point. But, going forward, the capsule could be modified to check for other common contaminants. The silicone shell could contain multiple tubes tuned for different contaminants, such as mercury and aluminum, allowing users to conduct a broad-spectrum screening in one test. Tang stresses that the device is still only a proof of concept and is far from implementation. “We wanted to show how the idea would work—that you can use it and apply other chemistry,” she says.
If successful, Tang’s system would solve a big water-testing puzzle. The current prototype represents the first time anyone has been able to detect more than a “yes or no” answer about heavy-metal contamination in water sources. Current methods, such as the handheld remote called ANDalyze, must remove samples from a water source for testing. In that case, she explains, users can identify the presence of metals, but have no means to isolate their source in the water supply. Even if the sensors could travel into cracks and fissures to reach groundwater, the delicacy of the electronic components also means that they might not survive well underground, where heat and pressure rise significantly.
At its current size, Tang’s sensor could be used to find pollutants and their sources in streams, but getting the system down to a nanoscale—about one millimeter—is her ultimate goal. “The real original motivation was in the need for sensing underground, where you would have a hole or well where you can’t possibly disperse sensors and collect [them] at the other end [using current technology],” she explains. As Tang told Stanford News, “The capsules would have to be small enough to fit through the cracks in rock layers, and robust enough to survive the heat, pressure and harsh chemical environment below ground.” Another large piece of the puzzle: Tang isn't yet sure how to collect the sensors after dispersion.
There’s plenty of water to screen. According to the Environmental Protection Agency, about 95 percent of all the fresh water resources in the U.S. are underground. Those sources are susceptible to a wide variety of pollutants that leech into the supply from plumbing, industry and general waste. There can also be a fair amount of prescription drugs in there as well.
Ultimately, the miniaturization process, which Tang says is still years away, might also breed a change in design. Instead of linear tubes that run in parallel, the millimeter-sized sensors would be round dots, she posits. In that case, the barcode would present itself as circles instead of stripes, “like rings on a tree,” she says.