Researchers design an on-chip printed ‘electronic nose’

Skoltech researchers and their Russian and German colleagues have designed a printed ‘electronic nose’ that will prove the concept for the use of low-cost and sensitive devices in portable electronics and health care. The paper was published in the journal ACS activated product interface.

The fast-growing domains of the Internet of Things (IoT) and advanced medical diagnostics require small, cost-effective, low-power but reasonably sensitive gas analysis systems, and are optional as ‘nozzles’. electronic ‘. These systems can be used for noninvasive diagnosis of human breath, such as the detection of chronic obstructive pulmonary disease (COPD) with a tight sensory system also designed at Skoltech. Some of these sensors work much like real noses – say, me – by using different sensors to detect a complex gaseous compound signal.

One way to create these sensors is with additive manufacturing technologies, which have achieved enough power and precision to produce the most complex devices. Skoltech senior research scientist Fedor Fedorov, Professor Albert Nasibulin, research scientist Dmitry Rupasov, and their colleagues created a multisensor ‘electronic nose’ by printing nanocrystalline films of eight metal oxides between differentiated on a multielectrode chip (they were manganese, cerium, zirconium, zinc, chromium, cobalt, tin, and titanium). The Skoltech team came up with the idea for this project.

For this work, we used microplotter printing and real solution inks. There are a few things that make it valuable. First, the print resolution is close to the distance between electrons on the chip, which is magnified for a more convenient measurement. We show that these technologies are compatible. Second, we were able to use several different oxygens, enabling more orthogonal signals from the chip leading to better selectivity. We can also imagine that this technology is reproducible and easy to implement in an industry to get chips with similar features, which is very important for the ‘e-nose’ industry. “

Fedor Fedorov, Chief Research Scientist Skoltech

In subsequent experiments, this ‘nose’ was able to detect the difference between different alcohol valves (methanol, ethanol, isopropanol, and n-butanol), which are very similar chemically and difficult to tell apart. , at low concentrations in the air. Because methanol is highly toxic, if found in beverages and distinguishes between methanol and ethanol, it can save lives. To process the data, the team used sequence analysis (LDA), a pattern recognition algorithm, but other machine learning algorithms could use this task.

To date, the machine operates at a relatively high temperature of 200-400 degrees Celsius. However, the researchers believe that new semi-2D materials such as MXenes, graphene, and so on could be used to increase the sensitivity of the series and ultimately allow it to operate at room temperature. -room. The team will continue to work in this way, making the best use of the materials used to reduce energy consumption.