WHY THIS MATTERS IN BRIEF
There are plenty of situations when you want to know what something’s made from, such as food or goods, and soon that power will be in your smartphone.
Imagine pointing your smartphone at a snack you found at the back of your pantry and immediately knowing if its ingredients had turned rancid, or were made from horse meat not beef, like the UK scandal recently, or using it to detect the ripeness of fruit, or the chemical composition of, well, anything, from cosmetics to drugs.
Devices called spectrometers do this all the time by virtue of the fact that they’re able to detect the chemicals within objects based on a unique “fingerprint” of absorbed and emitted light. But spectrometers, other than the one that was shown off a little while ago at CES, have long been both bulky and expensive, which has obviously limited their use outside the lab.
Until now that is after engineers at the University of Wisconsin-Madison announced they’ve developed a spectrometer that is so small and simple that it can integrate with the camera of a typical cell phone without sacrificing any accuracy.
“This is a compact, single-shot spectrometer that offers high resolution with low fabrication costs,” says Zhu Wang, who was among the team of electrical engineers that created the device.
A pin-sized revolution in chemical analysis
The researchers published a description of the devices in the journal Nature Communications.
The team’s devices also have an advanced capability called hyperspectral imaging, which collects information about each individual pixel in an image order to identify materials or detect specific objects amidst a complicated background. Hyperspectral sensing, for example, could be used to detect seams of valuable minerals within rock faces or to identify specific plants in a highly vegetated area, as well as, again, almost analyse the make up of almost anything and everything. So combine these two capabilities and you have one very powerful sensing system.
Every element’s spectral fingerprint includes unique emitted or absorbed wavelengths of light – and the spectrometer’s ability to sense that light is what has enabled researchers to do everything from analyse the composition of unknown compounds to reveal the makeup of distant stars.
Spectrometers usually rely on prisms or gratings to split light emitted from an object into discrete bands, each corresponding to a different wavelength. A camera’s photodetector can capture and analyse those bands, for example, the spectral fingerprint of the element sodium consists of two bands with wavelengths of 589 and 590 nanometers.
Human eyes see 590-nanometer wavelength light as a yellowish-orange shade. Shorter wavelengths correspond to blues and purples, whereas longer wavelengths appear red. Sunlight contains a complete rainbow mixed together, which we see as white.
To resolve the difference among a mixture of different colours, spectrometers usually must be relatively large with a long path length for light beams to travel and separate.
Yet the team created tiny spectrometers, measuring just 200 micrometers on each side, which is roughly one-20th the area of a ballpoint pen tip, and delicate enough to lie directly on a sensor from a typical digital camera.
That small size was possible because the researchers based their device on specially designed materials that forced incoming light to bounce back and forth several times before reaching the sensor. Those internal reflections elongated the path along which light travelled without adding bulk, boosting the devices’ resolution. And the devices performed hyperspectral imaging too, resolving two distinct images, in this case of the numbers five and nine, from a snapshot of an overlaid projection that combined the pair into something indistinguishable to the naked eye.
Now the team hopes to boost the device’s spectral resolution as well as the clarity and crispness of the images it captures, and those improvements could pave the way for even more enhanced sensors that help us open up a whole new world of use cases and capabilities for the supercomputers we all hold in our hands.