Selectively Demarcating and Modifying Defects in Two-Dimensional Semiconductors in a 300 mm Fab

The world of future electronics is looking for materials that are thinner, faster, and more energy-efficient than the silicon chips that power our current devices. Scientists have identified a promising class of materials called two-dimensional semiconductors, which are essentially sheets of atoms so thin that they are only one layer thick. Imagine a sheet of paper that is a million times thinner than a human hair; these materials are even more delicate. Among them, a material called molybdenum disulfide has shown great potential for building the next generation of transistors, the tiny switches that control the flow of electricity in computers. However, making these materials for real-world use is difficult because the process of growing them often leaves behind imperfections. These flaws can take the form of unwanted extra layers of material, tiny holes in the atomic structure, or boundaries where different crystal grains meet. Just as a scratch on a lens can blur a photograph, these defects scatter electrons and ruin the performance of the electronic devices built from them. For these materials to move from a laboratory experiment to a factory floor, engineers need a way to find these flaws and fix them without damaging the delicate atomic sheet.

Researchers at IMEC and Pennsylvania State University have developed a method to do exactly that, using a gas commonly found in semiconductor factories to clean up these atomic sheets. The team focused on molybdenum disulfide films grown on large, industry-standard wafers, a crucial step toward mass production. They discovered that by introducing a specific amount of chlorine gas into the manufacturing chamber, they could selectively target and remove the unwanted defects while leaving the perfect parts of the material untouched. The process works because the chlorine gas reacts differently depending on what it encounters. It attacks the edges of extra layers of material, known as bilayer islands, and the boundaries between different crystal grains, effectively eating them away. At the same time, it ignores the smooth, single-layer areas of the film. This selectivity is vital because it allows engineers to smooth out the surface of the material, removing the bumps and ridges that would otherwise disrupt the flow of electricity.

The researchers found that the temperature of the treatment acts as a precise control knob for this process. When the material is heated to a specific range between 500 and 600 degrees Celsius, the chlorine gas removes the extra layers of material but stops before it begins to eat into the single layer underneath. This creates a perfectly flat, single-atom-thick surface. If the temperature is raised higher or the treatment is applied differently, the gas can also be used to highlight the boundaries between crystal grains, making them visible to standard factory cameras. This ability to see the invisible structure of the material is a significant breakthrough, as it allows manufacturers to check the quality of the film quickly and accurately during the production process, rather than relying on slow and expensive laboratory techniques.

Beyond simply removing the extra layers, the treatment also heals the tiny holes and missing atoms within the material. The study showed that the chlorine gas does not just act as an eraser; it also acts as a healer. When the gas reacts with the material, it can fill in the missing spots in the atomic lattice, effectively repairing the damage. To make this healing process even more effective, the researchers added a second gas, hydrogen sulfide, to the mix. This combination shifted the chemical balance, preventing the gas from eating away too much of the material while encouraging it to fill in the defects. The result was a material that was not only smoother but also electrically superior, with fewer traps for electrons and a more consistent performance across the entire wafer.

The impact of these changes was measured by building tiny transistors from the treated material. The results were striking. The devices made from the treated films conducted electricity much better than those made from the untreated material. The current flowing through the transistors increased by up to 80 percent, and the speed at which the electrons moved through the material improved significantly. Perhaps just as important was the consistency of the results. In the untreated samples, the performance of the transistors varied widely from one to the next, a problem that makes mass production difficult. After the treatment, the variation dropped dramatically, with the devices behaving almost identically to one another. This uniformity is essential for building reliable computer chips, where billions of transistors must work in perfect harmony.

The study also ruled out the idea that simple heating alone could fix these problems. When the researchers heated the material without the chlorine gas, the defects remained, and the extra layers of material stayed in place. This confirmed that the chemical reaction with the chlorine was the key to the improvement, not just the heat. Furthermore, the researchers demonstrated that this process is compatible with the large-scale equipment used in modern semiconductor factories. They successfully treated films on wafers that are 300 millimeters in diameter, the standard size for high-volume chip manufacturing. This proves that the technique is not just a laboratory curiosity but a viable path for industrial adoption.

By combining the removal of unwanted layers with the healing of atomic defects, this new approach offers a complete solution for improving the quality of two-dimensional semiconductors. It addresses the three main hurdles that have prevented these materials from being used in commercial electronics: the presence of extra layers, the existence of invisible grain boundaries, and the scattering of electrons by atomic defects. The ability to selectively target these issues with a gas that is already familiar to the semiconductor industry suggests a smooth transition from research to production. As the demand for faster and more efficient computing grows, the ability to manufacture these ultra-thin materials with high precision and low defect rates will be critical. This work provides a clear, manufacturable path to achieving that goal, turning a fragile atomic sheet into a robust foundation for the electronics of the future.