Researchers can now connect them to other chip elements more effectively using atomically thin materials.
Moore’s Law, a famous prediction that the number of transistors that can fit onto a microchip would double every two years, is now at the limit. These limits could only stop decades of progress if new approaches are found.
One direction being explored is using atomically thin materials as the basis of new transistors. However, connecting these “2D” materials with other electronic components has proven difficult.
Researchers at MIT, the University of California at Berkeley, and Taiwan Semiconductor Manufacturing Company have discovered a new method to make these electrical connections. This could allow for 2D materials to be unleashed and further miniaturization of components. It may even extend Moore’s Law.
These findings were published in the journal Nature in a paper by Pin-Chun Shen, Ph.D. ’20 and Cong Su, Ph.D. ’20, recent MIT graduate Pin-Chun Shen Ph.D. 20, postdoc Yuxuan L Ph.D. 19, MIT professors Jing Kong and Tomas Palacios, and 17 other MIT and UC Berkeley researchers.
Illustration of the monolayer semiconductor transistor. Credit: The researchers
Su, now at UC Berkeley, says that “we solved one of the most difficult problems in a miniaturizing semiconductor device, the contact resistance zwischen a metal electrode and monolayer semiconductor materials.” It was easy to find a solution: using a semimetal (the element bismuth) to replace ordinary metals to connect to the monolayer material.
These ultrathin monolayer materials (in this case, molybdenum disulfide) are a strong contender to overcome the limitations of silicon-based transistor technology’s miniaturization. Su says that the challenge of creating a highly conductive interface between these materials and metal conductors to connect them and other devices and power sources could have improved progress toward such solutions.
Metal-induced gap states are formed when semiconductor materials and metals interact. This causes a Schottky barrier to form. This prevents charge carriers from flowing through the Schottky border. This problem was solved by using a semimetal whose electronic properties are similar to those of semiconductors.
Lin explains how the rapid pace at which transistors are being miniaturized has been stalled in the past, around 2000. However, a new technology allowing for a three-dimensional structure of semiconductor devices on a single chip broke the logjam and allowed rapid progress to resume. He says, “we believe we are at the edge of another bottleneck.”
This technology has demonstrated miniaturized transistors of extraordinary performance, which meets the requirements for the technology roadmap to future microchips. Credit: Credit to the researchers
Two-dimensional materials are thin sheets of one to a few atoms in thickness. They meet all requirements for miniaturization. This could reduce the channel length by several times. It is currently around 5-10 nanometers in cutting-edge chips. But, it can be scaled down to a sub-nanometer scale. Many of these materials, including the whole family of compounds called transition metal dichalcogenides, are being explored. This family includes the molybdenum disulfide used in the new experiments.
Basic research into the physics of novel monolayer materials is also needed by the difficulty of making low-resistance contact with metals. Existing connection methods are so resistant that the signals required to monitor electron behavior in the material’s nanostructures cannot be transmitted. “Many examples from the physics world require a low contact resistance between the metals and semiconductors. Su states that it is also a significant problem in the physics community.
It could take time to figure out how to scale these systems up and integrate them commercially. This will require more engineering. However, researchers believe that the new findings could immediately impact physics applications. Su states that physics can directly benefit from technology in many cases.
The researchers are still exploring further and reducing the device’s size. They also continue to look for combinations of materials that could enable better electrical contact with the other charge carriers (known as holes). The problem with the N-type transistor was solved. However, finding a combination channel and electrical contact material to enable a monolayer P-type transistor would open up new opportunities for next-generation chips.
