We were just walking outside, passing by some flowers that had a nice smell. I breathed the scent in deeply. It reminded me of something, but what?
I imagined how great it would be to have something like a portable analysis laboratory that would give me a list of the most abundant compounds in the flower smell. A mobile electro-chemical nose, if you will. (The idea isn’t new. The original Star Trek crew already had tricorders, which ostensibly included a chemical analysis capability. I wouldn’t be surprised if the idea has been floating around Sci-Fi circles for a hundred years, but it must have been around for certain since the 1950s.)
The technology exists (I think), but not at this size (yet). The chemical analysis is performed by a gas chromatography/mass spectrometry (GC/MS) combination, which has been standard equipment in any chemical analysis lab for decades. But all the devices I have seen are at least desktop-size and far from portable; except maybe for the compact chemical analysis devices carried by – and custom-made for – certain space probes, such as Huygens.
What is GC/MS? The gas chromatograph (GC) takes in volatile molecules and basically sorts them into packs of the same compound – it separates the input. The separated substances are then passed on to the mass spectrometer (MS), in which the molecules are essentially smashed into little fragments, which are then detected, each individual substance having a certain characteristic profile which allows it to be identified.
The device I’m imagining contains a full, nano-scale GC/MS system as well as some kind of interface (OLED or eInk display with touchscreen, probably), a small embedded computer, and a power source (LiPoly battery or something similar). The interface, computer and power supply bits are straightforward, as is the software that would be required. You could probably build a variant of such a device without its own interface but a connection to your smartphone (wirelessly, via NFC or bluetooth, for instance).
As a user, I would want the technology to have been simplified to such a degree that you’d only have to point the inlet towards the source of the scent you want analysed, press a button, and after a few seconds, you’d get a list of all the compounds identified in the sample, ordered by their relative concentrations (most abundant component first). This aroma profile would be characteristical for certain plants or foods or other things that produce smell (or even vapour that humans can’t smell), so the device would actually be able to tell you what it is that it’s »smelling«.
For example, if I pointed the device at the flower I smelled earlier and pushed the button, I’d get a list of compounds, and then the device would connect to the internet to search for matching aroma profiles (or use its on-device cached data), and I’d get a list of the best matches, hopefully including the name of the correct flower. (As with any such technology, there will be false matches.)
If this technology can be built to a high degree of precision and reliability, I’d imagine that identification of flowers would be trivial. You could analyse practically anything that gives off a sufficient number of molecules for the device to detect and identify. Of course, the less volatile molecules that can be detected, the less reliable the result. If you’d just point the device into the air, you’d get »random smell noise«, where no particular set of compounds is featured prominently enough to afford any identification.
Maybe it will even become possible to build the tiny GC/MS system using the same (or similar) process that’s used to produce integrated circuits today, lowering the production cost and making the device affordable to many. Since the market for such devices is probably not extremely large (yet), I’d imagine the cost at first to be around 2000 EUR, but then becoming gradually cheaper as production is simplified.
Possibly, there will be low-cost variants of such devices aimed at the consumer market. They would have much less of the precision of professional, scientific models, but they would still be good enough for personal use. Low-end consumer devices would possibly skip the chemical analysis part altogether and just identify the source of the sample, as most people don’t have the appropriate knowledge of chemistry, and they probably don’t care.
If devices like this can be mass-produced to such an extent that the price falls really low, it would be feasible for everybody to have such a device permanently installed at home, constantly monitoring the quality of the air, and warning of the release of hazardous smells.
For chefs, you could have specialised devices that could tell you when the food is ready to eat, based on the scent profiles being monitored. This would need a lot more intelligence on the software side, but I think it’s perfectly conceivable.
I can imagine a device like this, but I can’t build it. My knowledge is too limited. So I’m going to sit back and wait for about 10 to 20 years, when this vision will be reality. This is seriously a toy that I want to have!