MIT engineers are creating vertical micro-LEDs for next-generation displays and virtual reality glasses.
Perspective: An international team of scientists led by MIT engineers has developed a way to make defect-free micro-LED wafers using a vertical approach that could pave the way for a new generation of virtual reality displays.
Vertically stacked micro-LEDs can provide higher pixel densities while being much easier and cheaper to manufacture, making them a boon for VR devices where low pixel densities and the annoying “screen door effect” are long-term issues.
Traditional displays use light emitting diodes side by side with red, blue and green subpixels packed as tightly as possible. As with transistors in CPUs, this side-by-side arrangement quickly reaches its theoretical density limit, forcing researchers to find innovative solutions to increase pixel density.
MicroLEDs are another potential solution. They are made from inorganic semiconductor materials that are 100 times smaller than conventional LEDs and can provide better performance, consume less power and last longer than OLEDs. Conversely, they require a much more precise manufacturing process to perfectly align the sub-pixels in a traditional RGB configuration.
Jihwan Kim and Jiho Shin, two MIT researchers leading the study published in Nature, have developed a new method for manufacturing micro-LEDs that does not require such high precision. They were able to grow and stack ultra-thin membranes of red, green and blue LEDs in a vertical configuration, with each vertical pixel only four microns wide.
Shin said that when using a vertical pixel configuration, the pixel area could theoretically be reduced “by a third”. Vertical micro LEDs can achieve a density of over 5000 pixels per inch, which is highest reported to date. “Vertical pixelization is the way to higher resolution displays in a smaller footprint,” the researchers say, paving the way for virtual reality indistinguishable from reality.
The scientists were able to get their vertical pixels thanks to a previously developed method for growing and separating the ideal 2D material from silicon wafers and other surfaces, a process they called “2D material-based layer transfer” or 2DLT.
They also demonstrated how to stimulate a custom vertical structure to produce the full commercial spectrum of colors, but now they need to develop an active matrix system capable of driving “25 million LEDs individually”.