What can we learn from the evolution of semiconductor device structures?

As we discussed previously, the flat panel display (FPD) industry does not follow Moore’s Law per se, but leaders in the industry have formulated their own versions.

Examining what drove Moore’s Law in the semiconductor industry, and looking for parallels in the FPD industry, is an insightful exercise nevertheless; stimulating thoughts on how to drive display performance and create value.

Really we don’t have to look any further than the evolution of semiconductor and FPD device structures to see there are some literal and figurative parallels between the industries.

As pointed out by Tsu-Jae King Liu, from the Dept. of EECS at UC Berkeley in 2012, in their slide deck entitled “FinFET History, Fundamentals and Future”, it was the evolution of the FinFET structure that drove Moore’s Law.

Like MOSFET, there are a number of performance gains can be realized by using a double gate TFT structure. Arguably the most important to meeting Wang’s Law for panel performance is the 2X increase in mobility that can be realized.

An increase in mobility not only enables higher resolutions to be achieved, but can also impact other important parameters like refresh rate, dynamic range, and power consumption.

Following the dual-gate era, MOSFET structures evolved to the FinFET. FinFET use a split active area along with two gates.

We believe that the use of the split structure for TFT is envitable and we are already seeing the emergence of it.

The recent paper by S. Lee et al., presented at DisplayWeek2019 and entitled “Highly robust oxide TFT with bulk accumulation and source/drain/active layer splitting”, demonstrates that a 4X mobility increase can be realized by using a split TFT structure.

In addition to the mobility boost, the authors demonstrated that mechanical performance was enhanced.

We believe that the evolution of TFT towards dual gate and then split structures is key for meeting the performance levels the FPD industry needs to keep driving demand.

Key to the Amorphyx approach with AMeTFT development is starting with a great foundation, quite literally.

We begin by laying down a super smooth amorphous metal gate which then enables us to use a thin high-K dielectric for the gate insulator and form better interfaces throughout the device stack.

Just by transitioning to this materials set we are able to achieve a 3X increase in mobility for AMeTFT. Moving towards dual gate (additional 2X increase) and split (additional 4X increase) structure offer the possibility of increasing another 8X increase (24X total) for AMeTFT.


FinFET Evolution

J. K. Liu, “FinFET History, Fundamentals and Future,” p. 55.

Examples of Double-Gate TFT

K.-H. Moon, Y.-S. Cho, H. Choi, C.-K. Ha, C.-G. Lee, and S.-Y. Choi, “Characteristics of Amorphous Silicon Dual-Gate Thin Film Transistor Using Back Gate of Pixel Electrode for Liquid Crystal Display Driver,” Jpn. J. Appl. Phys., vol. 48, no. 3S2, p. 03B017, Mar. 2009.

K.-S. Son et al., “Characteristics of Double-Gate Ga–In–Zn–O Thin-Film Transistor,” Electron Device Letters, IEEE, vol. 31, pp. 219–221, Apr. 2010.

N. Münzenrieder et al., “Flexible Self-Aligned Double-Gate IGZO TFT,” IEEE Electron Device Letters, vol. 35, no. 1, pp. 69–71, Jan. 2014.

M. Nag et al., “Dual-Gate Self-Aligned a-IGZO TFTs using 5-Mask Steps,” in International Display Workshops, 2015, pp. 239–241.

Example of a Split Structure TFT

S. Lee, Y. Chen, J. Kim, H. Kim, and J. Jang, “Highly robust oxide TFT with bulk accumulation and source/drain/active layer splitting,” Journal of the Society for Information Display, vol. 27, no. 8, pp. 507–513, 2019.

PerspectiveSean Muir