The December 2022 issue of IEEE Spectrum is here!

Close bar

Test-Tube Hard Drives Compute with Chemicals

Storing data as small molecules and getting them to compute by reacting could speed some computations

2 min read
Test tubes and other objects forming a pattern of 0s and 1s.
Photo: Barry Rosenthal/Getty Images

A group of scientists and engineers at Brown University is planning to use chemicals in a droplet of fluid to store huge amounts of data and, eventually, get them to do complex calculations instantly. They’ve just received US $4.1 million from the Defense Advanced Research Projects Agency to get started, and plan to borrow robots and automation from the pharmaceutical industry to speed their progress.

“We’re hoping that at the end of this we’ll have a hard drive in a test tube,” says Jacob Rosenstein, assistant professor of electrical engineering, who is co-leading the project with theoretical chemist Brenda Rubenstein.

There’s been a big push recently to store data as molecules of DNA, but the Brown chemical computing project will do things differently, potentially ending up with greater data density and quicker readouts.

DNA data-storage techniques encode the data as long chains of chemical “letters.” But in the Brown team’s scheme, each data point could be represented by its own chemical. A simple version of that scheme using, say, the presence or absence of any of four chemicals, could then encode 16 states. Using a similar scheme, they encoded a simple 81-bit image using 25 chemicals, just to prove it could be done. In that scenario, each pixel position was represented by a different substance.

If it seems like they’ll need a lot of different chemicals to make this scheme do anything important, you’re right. But that may not be a big problem at all. The Brown team will rely on a class of chemistry called Ugi reactions. These are combinations of four chemicals reacted together all at once to produce a small organic molecule. The pharmaceutical industry uses automated systems that react different combinations to produce millions of chemicals for testing as drugs.

So how do they tell which of those millions is actually in a drop of liquid? Rubenstein, Rosenstein, and their colleagues will use a technology called mass spectrometry. It’s essentially the same technology used to find evidence of doping drugs in athletes’ urine. Basically, it charges the molecules in a sample and sprays them toward a detector. Their mass and charge determine exactly when they arrive at the detector, helping to identify to each molecule. [See the illustration in "New Tech to Find Cheaters at the 2016 Olympics," IEEE Spectrum, August 2016.]

Reading and writing data are just the beginning, though. The team wants to do computation by reacting data sets together. “That part is still being worked out,” says Rosenstein. It will require mapping the chemical data set carefully so that a reaction converts to meaningful data. “We have some ideas we’ll be exploring.”

The Conversation (0)

Why Functional Programming Should Be the Future of Software Development

It’s hard to learn, but your code will produce fewer nasty surprises

11 min read
A plate of spaghetti made from code
Shira Inbar

You’d expectthe longest and most costly phase in the lifecycle of a software product to be the initial development of the system, when all those great features are first imagined and then created. In fact, the hardest part comes later, during the maintenance phase. That’s when programmers pay the price for the shortcuts they took during development.

So why did they take shortcuts? Maybe they didn’t realize that they were cutting any corners. Only when their code was deployed and exercised by a lot of users did its hidden flaws come to light. And maybe the developers were rushed. Time-to-market pressures would almost guarantee that their software will contain more bugs than it would otherwise.

Keep Reading ↓Show less