Layer-by-layer growth of highly aligned MoS$_2$ nanoribbon arrays

In the relentless drive to make electronic devices smaller and faster, engineers have reached a point where traditional silicon transistors are becoming too thin to control effectively. To solve this, scientists are turning to a new class of materials known as transition metal dichalcogenides. Imagine these as sheets of atoms so thin that they are essentially two-dimensional, yet they possess the ability to conduct electricity in ways that silicon cannot. While flat sheets of these materials are useful, the most promising path for future ultra-small transistors involves shaping them into long, narrow strips called nanoribbons. These one-dimensional structures act like tiny, efficient channels for electric current, offering better control over the flow of electrons. However, a significant hurdle has remained: while researchers could easily grow these strips as a single layer of atoms, they struggled to add more layers on top in a controlled way. Adding layers is crucial because thicker strips often conduct electricity much better, but previous methods either failed to add layers or created messy, uneven structures that ruined the device's performance.

A team of researchers has now demonstrated a way to build these nanoribbons layer by layer, like stacking sheets of paper, with precise control over how many layers are added. They grew strips of molybdenum disulfide, a specific type of these atom-thin materials, on a surface of sapphire. The process began by growing a single, perfectly aligned layer of nanoribbons. Once this foundation was set, the researchers changed the conditions inside their growth furnace, specifically by increasing the supply of the molybdenum precursor. This adjustment triggered a shift in behavior: instead of the material spreading out sideways to cover more of the surface, it began to stack vertically, adding a second layer directly on top of the first. By carefully tuning the temperature and the amount of material supplied, the scientists could stop the process at two layers, or continue it to create strips with three or even four layers. Crucially, this vertical growth happened without widening the strips, keeping them narrow enough to be useful for tiny transistors.

The researchers confirmed that these new, thicker strips were not just random piles of atoms but were highly ordered, single-crystal structures. They found that the layers stacked in a specific, repeating pattern known as the 2H configuration, which is ideal for electronic devices. This structure is so uniform that the strips remained free of the internal boundaries and defects that often plague materials grown by other methods. To prove the quality of their work, the team built tiny electronic switches, or transistors, using these nanoribbons. They compared devices made from single-layer strips against those made from double-layer strips. The results were clear: the double-layer strips allowed electricity to flow much more easily. The devices made from the two-layer strips moved electrons at speeds roughly twice as fast as the single-layer versions and carried three times more current. This improvement suggests that controlling the thickness of these nanoribbons is a key to unlocking the full potential of next-generation electronics.

The study also showed that this method is versatile enough to create more complex structures. By swapping the materials used in the second step of the process, the team was able to grow a different material, tungsten disulfide, directly on top of the molybdenum disulfide strips. This created a vertical stack of two different materials, known as a heterostructure, which is a building block for advanced electronic components. The ability to switch materials mid-growth without breaking the alignment of the strips opens the door to creating a wide variety of custom-designed electronic layers. The researchers achieved this by leveraging the unique interaction between the nanoribbons and the sapphire surface. Unlike other surfaces where a new layer might only form after the entire surface is covered, the sapphire surface allowed the new layers to form exclusively on top of the existing strips. This selective growth was confirmed through detailed computer simulations, which showed that the atoms prefer to stick to the existing ribbon rather than the empty sapphire surface once the first layer is present.

The success of this approach relies on a delicate balance of time and temperature. The researchers found that simply growing the material for a longer time did not automatically add more layers; instead, the key was increasing the heat of the precursor material to drive more atoms into the reaction. At lower temperatures, the strips remained single-layered regardless of how long they were grown. As the temperature rose, the strips began to stack, first forming a complete second layer, and then, with even higher temperatures and longer times, adding a third and fourth layer. Throughout this process, the width of the strips remained remarkably consistent, staying below 20 nanometers in many cases. This precision is vital because the performance of these tiny transistors depends heavily on their width. The fact that the researchers could control the thickness without sacrificing the narrow width of the strips represents a significant step forward in manufacturing high-quality, atomically thin electronics.

By demonstrating that it is possible to grow these complex, multi-layered structures with such precision, the researchers have provided a general method for creating better transistors. The devices they built showed performance metrics that rival or exceed the best results reported so far for similar materials, even though the channels in their devices were extremely narrow. This suggests that the combination of a clean, bottom-up growth process and the ability to control the number of layers is a powerful strategy for future chip design. The work highlights that the thickness of the channel is not just a minor detail but a fundamental parameter that can be tuned to optimize how well a transistor works. As the industry moves toward even smaller and more efficient electronic components, the ability to stack these atomic layers with such control offers a clear and practical path forward, moving beyond the limitations of single-layer materials to a new era of three-dimensional, yet atomically thin, electronics.