However, it is generally agreed that today's transistors, with their planar architecture, are nearing the limits of miniaturization: there is a minimum size under which the gate control over the channel becomes less and less effective. Over the past 50 years, transistors have been steadily reduced in size, enabling the development of increasingly powerful microelectronic devices. Acting like a switch, the gate determines whether the transistor is on or off. The flow of current between these terminals is controlled by a third terminal, called gate. The 'building blocks' of microelectronics, transistors consist of a semiconductor component, called channel, linking two terminals. The use of 3D transistors could significantly increase the power of microelectronic devices. DOI: 10.1126/science.aah4698 ( About DOIs).Published in Nanoscale, these findings open the way toward alternatives to the planar structures used in microprocessors and memory units. We have not yet packed these transistors onto a chip, and we haven't done this billions of times over.' The concept, of course, is that materials other than silicon might help get us past the limits of today's transistors. 'This work demonstrated the shortest transistor ever,' said Ali Javey, the paper's senior author. While it worked, even the researchers involved caution people not to get too excited just yet. And the amount of current that transited across the MoS 2 in the on state was 10 6 times larger than the current in the off state. But the resulting circuit was fully functional the presence of current in the carbon nanotube allowed it to be switched on and off in a controlled manner. The nickel source and drain wiring was then linked to the MoS 2, completing the circuitry.Īll the wiring was quite a bit larger than the tiny dimensions of the two main ingredients, so this is not a complete 1nm feature. The authors tested a variety of flake sizes, but most of the work was done with one that was two atomically thin layers stacked on top of each other. Rather than move them around, the team simply identified where they were and then used standard lithography techniques to connect the nanotubes to wiring.Īll this was then embedded in a layer of zirconium oxide, after which flakes of MoS 2 were placed above the carbon nanotubes.

The carbon nanotubes were made first, through chemical vapor deposition onto a silicon substrate. So the production procedure for the tiny transistor was a bit involved. Unfortunately, two of these materials-MoS 2 and carbon nanotubes-are extremely difficult to make where you want or to move into place when you're done. The wires for the current source and drain were made of nickel. To get something of the appropriate size, the team making the device turned to carbon nanotubes. To get a functioning transistor, however, you need more than just a semiconductor like MoS 2: you need a gate to control whether it conducts or not.

Plus, MoS 2 naturally forms sheets that are just a single atom thick, making it relatively easy to make incredibly small devices.

This slows them down, which limits device performance, but it also makes it much harder for them to randomly leak across a transistor even as the transistor size gets ever smaller. In this material, electrons move as if they were heavier than they are in silicon.

MoS 2 offers a potential solution to this. Once silicon features get small enough (that 5nm limit mentioned above), leakage becomes large enough that it's impossible to tell whether a transistor is on or off. This unwanted movement causes an increase in current leaking across transistors when they're supposed to be off. On the positive side, the electrons move with less resistance when we want them to, but they also move more readily when we don't want them to. That property is the mobility of electrons within silicon. The idea behind the work is that a property of silicon we normally view as beneficial becomes an issue once things get small enough. But the work validates that MoS 2's properties can allow us to shrink electronics down below silicon's limits. Unfortunately, other parts of the hardware are quite a bit larger than that, and we have no way of producing these in bulk yet. Now, a large collaboration has taken a different material-molybdenum disulfide, or MoS 2-and used its distinctive properties to craft a transistor that has a gate size of just one nanometer. That's led to research into alternative materials such as carbon nanotubes. Below about five nanometers, quantum effects make their behavior unpredictable. Sujay Desai/Berkeley Lab reader comments 35 withĬonventional silicon-based electronics are rapidly approaching a fundamental barrier.

Size of transistors today