Government and local law enforcement agencies have employed canines for decades to sniff explosives and such banned goods as fresh produce, exotic wildlife, undeclared currency and illicit drugs. Now researchers are developing electronic noses imitating the sense of smell of humans and animals especially dogs.
The smells are composed of molecules, which has a specific size and shape. Each of these molecules has a corresponding sized and shaped receptor in the human nose. When a specific receptor receives a molecule it sends a signal to the brain and brain identifies the smell associated with the particular molecule. The electronic noses work in a similar manner of human. The electronic nose uses sensors as the receptor. When a specific sensor receives the molecules, it transmits the signal to a program for processing, rather than to the brain.
It has been reported that the human nose has around 400 scent receptors and can detect at least one trillion odors (Bushdid et al., 2014). Although the human nose can rate a smell, individuals’ judgments may be bias, and human nose cannot be used to sense toxic gases. In addition, human nose has detection limits for difference gases. Those limitations prevent the human nose from being a universal tool for all smell-related discrimination and classification.
The electronic nose is an array of chemical sensors, connected to a pattern-recognition system that responds to odors passing over it. Different odors cause different responses in the sensors, and these responses provide a signal pattern characteristic of a particular aroma.Typically, the volatile molecules react with the sensing materials of the gas sensor and cause irreversible changes in electrical related properties, such as conductivity. The computer evaluates the signal pattern and can compare the aromas of different samples, using pattern recognition algorithms such as artificial neural network (ANN), to perform discrimination and classification.
Compared with traditional gas analytical equipment including, GC–MS, high-performance liquid chromatography (HPLC), and Fourier transform infrared (FT-IR) spectrometry, e-nose is a relatively inexpensive and less time-consuming approach. These electronic noses can assist security agencies to detect explosives in venues like stadiums, schools, airports, hospitals or any other crowded place, which offer an attractive target to terrorists.
The best-known electronic nose is the breathalyser. As drivers breathe into the device, an ethanol specific chemical sensor measures the amount of alcohol in their breath. This chemical reaction is then converted into an electronic signal, allowing the police officer to read off the result. Alcohol is easy to detect, because the chemical reaction is specific and the concentration of the measured gas is fairly high. But many other gases are complex mixtures of molecules in very low concentrations. Building electronic noses to detect them is thus quite a challenge. The electronic nose can detect gas molecules with more specificity and sensitivity than Breathalyzers, which can confuse acetone for ethanol in the breath. The distinction is important, for example, for patients with Type 1 diabetes who have high concentrations of acetone in their breath.
Common sensors used in electronic nose
Depending on the sensing materials, gas sensors can be classified into several types including, conducting polymers (CP), metal-oxide-semiconductor (MOS), quartz crystal microbalance (QCM), and surface acoustic wave (SAW) sensors. The sensing materials are coated onto a ceramic substrate, such as alumina.
Many research groups in academia and industry are focusing on the performance improvement of electronic nose (E-nose) systems mainly involving three optimizations, which are sensitive material selection and sensor array optimization, enhanced feature extraction methods and pattern recognition method selection. Researchers are also trying to make them affordable and widely available. The possible use of the electronic nose is almost limitless.
Electronic nose system
The physical part of a typical e-nose system that consists of gas sensor arrays, reaction chamber, valves, air pumps for sampling and cleaning, control devices, and data acquisition (DAQ) devices. Typically, the device also has a heating element. One example of a typical e-nose system is shown in Fig.
Pattern recognition algorithms and classification methods for e-nose data
Signals obtained by an e-nose system combined with pattern-recognition algorithms or pattern classifiers such as Principal Component Analysis (PCA), support vector machines (SVM), artificial neural networks (ANN), random forest (RF), and other machine learning classifiers enable the recognition of different sample types via aggregation of similar emissions into clusters representing compounds from related food volatiles.
Limitations and future trends of e-noses
It is well-known that sample preparation and sampling are error-prone steps for e-nose measurements. Gas sensors are very sensitive to temperature, humidity, pressure, gas velocity, and vapor concentration. Sample preparation of e-nose sensing is also very challenging since the amount of volatiles released from foods depends on many factors such as temperature, pressure, and humidity. High repeatability and precision of e-nose measurement require strict control of sample preparation and sampling environment (Kiselev et al., 2018). Therefore, it is difficult to use e-noses in an open field or mobile sensing.
A great number of sample size (typically >10) for each type of sample is often required for training and validation. In some cases, the sample size needs to be even greater. Some types of gas sensors, such as MOS sensors, required heating to reach an equilibrium state before measuring, which is both time and energy consuming. Also, high energy demands also limit the portability of the e-noses. Unlike the human nose where hundreds of scent receptors reside, one e-nose typically employs a few numbers of sensors (less than 20); thus, their capability to discriminate different aroma is hampered.
One direction for future e-nose development is to minimize sample handling procedures and reduce the influences of the sampling environment. This requires the development of new sensing materials that are insensitive to environment variation while having high specification to some volatiles. E-nose will also employ nano gas sensors so that the number of gas sensors employed by an e-nose will be close to the number of scene receptors in the human nose. This will expand the capability of e-nose to differentiate different aroma. Another trend for e-nose is the employment of big data and artificial intelligence. For example, the development of a shared online library where data obtained from users all over the world using standardized e-noses can be used by other users to train their e-nose.
Nanotechnology based electronic nose
The recent nanotechnology advancements allow the fabrication of the E-noses in the size domain where miniaturization of the “classical” thin film micro analytical systems encounters principal limitations. The evolution of the sensors and electronic noses include: single nanowire sensors, analytical systems-based distributed nanowire sensors and integrated sensors based on nanowire networks and, finally, the simplest and yet fully functioning E-nose made of an individual single-crystal metal oxide quasi-1D nanostructure.
Carbon nanotubes (CNTs) based Compact electronic nose to identify human lung diseases and Covid
Researchers from Russia and Italy have proposed a compact sensor system that can implement the functionality of an electronic nose device and have developed a reproducible technology for its manufacture. The device is a flexible electronics platform that can analyze exhaled air as well as identify pathologies of the respiratory tract and organs. During the experiments, the device demonstrated high accuracy in determining patients with chronic obstructive pulmonary disease (COPD), an inflammatory disease of the respiratory tract, which increases the risk of complications when during COVID-19 infection.
The system is based on modified carbon nanotubes (CNTs), which allows the electronic nose to combine multiple properties. For example, flexible conductive films can be made from carbon nanotubes. Such films are needed in order to provide the system with an electronic structure layer responsible for the operation of the device. “CNTs were synthesized by aerosol chemical vapor deposition and deposited in the form of thin transparent and conductive films. This technology is highly reproducible, easily scalable and allows applying films of nanotubes to any surface,” said Albert Nasibulin, professor at the Skolkovo Institute of Science and Technology and the Russian Academy of Sciences.
Chinese Scientists develop graphene based electronic nose
In a study published in ACS Nano, a research group in China has developed a gas sensor that mimics the dog’s nose in terms of structure and sensitivity. It is well known that dogs have a better sense of smell than humans. The inside of a dog’s nose is lined with millions of tiny capillaries, which creates a super-sensitive sense of smell. Since the capillaries cover such a large surface area, they can detect odors at extremely low concentrations.
For years, researchers have been trying to develop an artificial detector that is just as good as a canine’s nose. Previous studies have had some success in using graphene-based nanoscrolls, which are nanosheets of graphene rolled up in continuous and uniform manner. These nanoscrolls have a large surface area, are stable at high temperatures and are strong and durable. But they are also difficult to manufacture, consume a lot of energy and difficult to scale up.
Drawing inspiration from the capillary structure in the dog nose, researchers led by Professor Zhou Goufu of South China Normal University, China, created graphene-based nanoscrolls by subjecting graphene to a freeze-drying method in the presence of a polymer, poly(sodium-p-stryrenesulfonate). Upon examination, the nanoscrolls had wide, tubular shapes, and almost all of the graphene was rolled up. The researchers also used molecular dynamics simulations to model how these uniform, unaggregated structures were formed. The researchers then incorporated the nanoscrolls into a gas sensor, which was highly selective and sensitive for the gas, nitric oxide. The team noted that their method has the potential for large-scale production.
Metal-organic frameworks, or MOFs based electronic nose for monitoring air quality, diagnosing disease
Research at Oregon State University has pushed science closer to developing an electronic nose for monitoring air quality, detecting safety threats and diagnosing diseases by measuring gases in a patient’s breath. Recently published research led by Cory Simon, assistant professor of chemical engineering in the OSU College of Engineering, in collaboration with chemical engineering professor Chih-Hung Chang focused on materials known as metal-organic frameworks, or MOFs.
The research took aim at a critical yet understudied hurdle in using MOFs as gas sensors: Out of the billions of possible MOFs, how do you determine the right ones for building the optimal electronic nose? MOFs have nanosized pores and selectively adsorb gases, similar to a sponge. They are ideal for use in sensor arrays because of their tunability, enabling engineers to use a diverse set of materials that allows an array of MOF-based sensors to deliver detailed information.Depending on which components make up a gas, different amounts of the gas will adsorb in each MOF. That means the composition of a gas can be inferred by measuring the adsorbed gas in the array of MOFs using micro-scale balances.
The challenge is that all MOFs adsorb all gases – not to the same extent, but nevertheless the absence of perfect selectivity prevents an engineer from simply saying, “let’s just dedicate this MOF to carbon dioxide, that one to sulfur dioxide, and another one to nitrogen dioxide.””Curating MOFs for gas sensor arrays is not that simple because each MOF in the array will appreciably adsorb all three of those gases,” Simon said. Human noses navigate this same problem by relying on about 400 different types of olfactory receptors. Much like the MOFs, each olfactory receptor is activated by many different odors, and each odor activates many different receptors; the brain parses the response pattern, allowing people to distinguish a multitude of different odors.
“In our research, we created a mathematical framework that allows us, based on the adsorption properties of MOFs, to decide which combination of MOFs is optimal for a gas sensor array,” Simon said. “There will inevitably be some small errors in the measurements of the mass of adsorbed gas, and those errors will corrupt the prediction of the gas composition based on the sensor array response. Our model assesses how well a given combination of MOFs will prevent those small errors from corrupting the estimate of the gas composition.”
Though the research was primarily mathematical modeling, the scientists used experimental adsorption data in real MOFs as input, Simon said, adding that Chang is an experimentalist “who we are working with to make a real-life electronic nose to detect air pollutants.” “We are currently seeking external funding together to bring this novel concept into physical realization,” Simon said. “Because of this paper, we now have a rational method to computationally design the sensory array, which encompasses simulating gas adsorption in the MOFs with molecular models and simulations to predict their adsorption properties, then using our mathematical method to screen the various combinations of MOFs for the most accurate sensor array.”
Meaning that instead of an experimental trial-and-error approach to decide which MOFs to use in a sensor array, engineers can use computational power to curate the best collection of MOFs for an electronic nose. Another exciting application of such a nose could be diagnosing disease. The volatile organic compounds humans emit, such as through our breath, are filled with biomarkers for multiple diseases, and studies have shown that dogs — which have twice the number of different olfactory receptors as humans — can detect diseases with their nose. Marvelous though they are, however, dogs’ noses aren’t as practical for widespread diagnostic use as a carefully crafted and manufactured sensor array would be
MOF based Electronic nose smells pesticides, nerve gas
Researchers from KU Leuven have now built a very sensitive electronic nose with metal-organic frameworks (MOFs). “MOFs are like microscopic sponges,” postdoctoral researcher Ivo Stassen explains. “They can absorb quite a lot of gas into their minuscule pores.””We created a MOF that absorbs the phosphonates found in pesticides and nerve gases. This means you can use it to find traces of chemical weapons such as sarin or to identify the residue of pesticides on food. This MOF is the most sensitive gas sensor to date for these dangerous substances. Our measurements were conducted in cooperation with imec, the Leuven-based nanotechnology research centre. The concentrations we’re dealing with are extremely low: parts per billion — a drop of water in an Olympic swimming pool — and parts per trillion.”
The chemical sensor can easily be integrated into existing electronic devices, Professor Rob Ameloot adds. “You can apply the MOF as a thin film over the surface of, for instance, an electric circuit. Therefore, it’s fairly easy to equip a smartphone with a gas sensor for pesticides and nerve gas.” “Further research will allow us to examine other applications as well,” Professor Ameloot continues. “MOFs can measure very low concentrations, so we could use them to screen someone’s breath for diseases such as lung cancer and MS in an early stage. Or we could use the signature scent of a product to find out whether food has gone bad or to distinguish imitation wine from the original. This technology, in other words, offers a wide range of perspectives.”
Airbus Wants its Bomb-detecting ‘E-nose’ to Sense COVID-19
In 2017, aircraft manufacturing giant Airbus agreed to a partnership with a Silicon Valley startup named Koniku Inc., to develop an “electronic nose” that could detect explosives in airports and on aircraft. In light of the current COVID-19 pandemic, however, Airbus said they are pivoting to incorporate “biological hazard detection” capabilities into the e-nose. They expect to conduct in-situ testing by later-2020.
The biotechnology solution is molecular-based. Oshiorenoya Agabi, founder of Koniku told that Koniku’s scientists genetically modify either HEK cells (kidney stem cells) or astrocytes (brain cells) to have olfactory receptors that can identify compounds by smell. Essentially, the e-nose continually “smells” the air looking for molecules associated with known explosives or biohazards. When a potentially harmful molecule binds to the living cells inside the e-nose, an identification is made and the proper authorities are alerted. This all occurs within 10 seconds or less. In fact, according to Koniku, its sensors can uniquely identify bioweapons with high specificity (either directly or through metabolites) in milliseconds. A single chip includes redundant neurons that can detect explosives simultaneously, and react to new threats in just hours.
While the synthetic biotechnology tech is new, the idea behind sensing differences in a person’s breath is not. Medical doctors and scientists have hypothesized for hundreds of years that certain diseases cause detectable changes in a person’s breath. For example, breath that smells overly fruity could be a sign of diabetes, as acetone builds up to a dangerous level in a person’s body. Researchers in the forensic space are continually developing methods and tools to sense specific metabolites present in a person’s breath after they have indulged in drugs, especially cannabis. In relation to COVID-19, researchers think they can auditorily differentiate between types of coughs in infected versus non-infected people. Although it is not approved as a diagnostic tool, a team at Carnegie Mellon have trained an algorithm and created an app to do just that.
The Airbus/Koniku e-nose is currently in the prototype phrase. In photos (like the one above), the purple-tinted device appears to be able to “suction” to a variety of surfaces both on an aircraft and in an airport. While airport security systems are notoriously controversial (and somewhat ineffective), Airbus says this e-nose is specially “developed to meet the rigorous operational regulatory requirements of aircraft and airport security operations.”
Scientific Gains May Make Electronic Nose the Next Everyday Device, reported in 2016
Researchers at the Texas Analog Center of Excellence (TxACE) at UT Dallas are working to develop an affordable electronic nose that can be used in breath analysis for a wide range of health diagnosis. The new research was presented in a paper titled “200-280GHz CMOS Transmitter for Rotational Spectroscopy and Demonstration in Gas Spectroscopy and Breath Analysis” at the 2016 IEEE Symposia on VLSI Technology and Circuits in Honolulu, Hawaii.
While devices that can conduct breath analysis using compound semiconductors exist, they are bulky and too costly for commercial use, said Dr. Kenneth O, one of the principal investigators of the effort and director of TxACE. The researchers determined that using CMOS integrated circuits technology will make the electronic nose more affordable. CMOS is the integrated circuits technology used to manufacture the bulk of electronics that have made smartphones, tablets and other devices possible.
“Smell is one of the senses of humans and animals, and there have been many efforts to build an electronic nose,” said Dr. Navneet Sharma, the lead author of paper, who recently defended his doctoral thesis at UT Dallas. “We have demonstrated that you can build an affordable electronic nose that can sense many different kinds of smells. When you’re smelling something, you are detecting chemical molecules in the air. Similarly, an electronic nose detects chemical compounds using rotational spectroscopy.”
The rotational spectrometer generates and transmits electromagnetic waves over a wide range of frequencies, and analyzes how the waves are attenuated to determine what chemicals are present as well as their concentrations in a sample. The system can detect low levels of chemicals present in human breath.Breaths contain gases from the stomach and that come out of blood when it comes into contact with air in the lungs. The breath test is a blood test without taking blood samples. Breath contains information about practically every part of a human body.
The researchers envision the CMOS-based device will first be used in industrial settings and then in doctors’ offices and hospitals. As the technology matures, they could become household devices. Dr. O said the need for blood work and gastrointestinal tests could be reduced, and diseases could be detected earlier, lowering the costs of health care. The researchers are working toward construction of a prototype programmable electronic nose that can be made available for beta testing sometime in early 2018.
Skoltech team developed on-chip printed ‘electronic nose’ reported in Jan 2021
koltech researchers and their colleagues from Russia and Germany have designed an on-chip printed ‘electronic nose’ that serves as a proof of concept for low-cost and sensitive devices to be used in portable electronics and healthcare. The paper was published in the journal ACS Applied Materials Interfaces.
The rapidly growing fields of the Internet of Things (IoT) and advanced medical diagnostics require small, cost-effective, low-powered yet reasonably sensitive, and selective gas-analytical systems like so-called ‘electronic noses.’ These systems can be used for noninvasive diagnostics of human breath, such as diagnosing chronic obstructive pulmonary disease (COPD) with a compact sensor system also designed at Skoltech. Some of these sensors work a lot like actual noses — say, yours — by using various sensors to detect the complex signal of a gaseous compound.
One approach to creating these sensors is by additive manufacturing technologies, which have achieved enough power and precision to produce the most intricate devices. Skoltech senior research scientist Fedor Fedorov, Professor Albert Nasibulin, research scientist Dmitry Rupasov, and their collaborators created a multisensor ‘electronic nose’ by printing nanocrystalline films of eight different metal oxides onto 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 true solution inks. There are a few things that make it valuable. First, the printing resolution is close to the distance between electrodes on the chip, which is optimized for more convenient measurements. We show these technologies are compatible. Second, we managed to use several different oxides, enabling more orthogonal signals from the chip resulting in improved selectivity. We can also speculate that this technology is reproducible and easy to be implemented in industry to obtain chips with similar characteristics, and that is really important for the ‘e-nose’ industry,” Fedorov explained.
In subsequent experiments, this ‘nose’ was able to sniff out the difference between different alcohol vapors (methanol, ethanol, isopropanol, and n-butanol), which are chemically very similar and hard to tell apart, at low concentrations in the air. Since methanol is extremely toxic, detecting it in beverages and differentiating between methanol and ethanol can save lives. To process the data, the team used linear discriminant analysis (LDA), a pattern recognition algorithm, but other machine learning algorithms could also be used for this task.
So far, the device operates at rather high temperatures of 200-400 degrees Celsius. Still, the researchers believe that new quasi-2D materials such as MXenes, graphene, and so on could be used to increase the sensitivity of the array and ultimately allow it to operate at room temperature. The team will continue working in this direction, optimizing the materials used to lower power consumption.