IEEE Future Directions
Related Articles
Molecular computers are using molecules to perform some computations. You cannot expect them to tell you how much 2+2 would be, unless you are happy with a result such as: “well there is a good probability that it makes 4 but of course it can also be 5 or 3 and sometimes ever 7…”. Now, it doesn’t go exactly like that but you get the gist.
They use billions of molecules that statistically provide sensible result. They work pretty well in areas like genome sequencing where the input is based on molecules and some sort of pattern matching is required.
Molecular computers are not based on silicon, although they may use silicon to detect the result of computation converting laser (diffraction) beams into bits and data. They use microfluidics, tiny pipes where molecules meet and part.
Now a team of scientists at Nanchang University, in China, have been able to use a few molecules as information processors. They have exploited the property of some molecules that can present a left or right form and have demonstrated that it is possible first to dispose the molecules in a given form and then have this form changed as result of the introduction of a molecules having a specific form. In a way it is like creating a device made up with cogs, each molecule being a cog that can rotate left or right and then adding a specific molecule whose orientation is affected by the existing orientation of two molecules. By assigning a value to a specific orientation (left handed is 1, right handed is 0) the scientists have been able to create a “micro”, actually a “nano” logic gate.
The property they exploited is the in-plane orientation, in other words the phenomenon that lead an organic molecule landing on a solid surface to (randomly) take a left or right handed geometry. By using two other molecules with a specific orientation it is possible to steer the new molecule into a specific orientation. Basically this means that by using two input signals (the existing molecules) you get a desired output signal: that’s a logic gate.
It is the first demonstration of the possibility of transferring a signal from one molecule to another.
Clearly this is a proto-proto-proto computer but the team is positive that this achievement demonstrated the possibility of using molecules in computation in a way that is similar to silicon logic gates.
Work will go on to try setting up different molecules configurations (in practice creating logic gates circuits) that would allow the processing of a signal through a pre-designed algorithm.
In the future this might have application in sensors.
