Computing goes into solution

South Korean scientists have developed the first soluble molecular logic gates - one step along the way to designing molecular computers and biological lab-on-a-chip devices. 

A team of scientists led by Juyoung Yoon of Ewha Womans University has used solutions of fluorescent sensor molecules that respond to ’inputs’ of pH, metal ions and, for the first time in a molecular logic gate, proteins. 

The researchers made a microfluidic device consisting of a network of tiny channels and valves that lead to a central loop filled with a fluorescent sensor solution. The valves allow input solutions to flow into a central loop where they mix with the sensor and, in certain combinations, switch the fluorescence ’output’ on or off. 

Using this approach the team were able to construct logic gates, the basic components of computing. ’Once you have the gates, you can solve Boolean logic and then construct a Turing machine, which is the basis of our digital electronic computers’ said Yoon. 

One gate that Yoon has designed is an ’XOR’ gate using a derivative of fluorescein that emits green light only at neutral pH. 

The fluorescein solution was added to the loop in a basic environment so no fluorescence was seen. When either one of two acidic inputs was turned ’on’ by injection into the loop, they neutralised the solution - switching on the fluorescence. But when both inputs were injected the pH became too acidic and the solution stopped fluorescing. 

In another gate, they used Cu²⁺ solution as an input instead because it also turns off emission from the fluorescein derivative. But input of a protein called Tf+, which can complex the Cu²⁺, turned fluorescence on again. 

A Prasanna de Silva, who researches molecular logic at Queens University Belfast, praised the work’s originality. But he believes that for molecular logic to have the edge over silicon circuitry it has to be miniaturized even further, to the scale of single molecules. 

’Semiconductor engineers will question whether liquid-flow in micrometric channels can do any more than imitate electron flow in wires’ de Silva said. Logic systems at the molecular scale, however ’can enter small spaces such as cells where semiconductor computing devices cannot go’ he added. 

Yoon accepts there are many more challenges to overcome before molecular chips can replace silicon. But he believes that the demonstration of a logic gate with a protein input makes lab-on-a-chip sensors for biomedical analysis a practical possibility. 

With systems like these, Yoon said, a molecular input can be processed to a molecular output. ’We foresee in the future that samples from patients can be processed in such devices and even produce a molecule for treatment’ he told Chemistry World. 

Tom Westgate