Graphene May Challenge Dominance Of Silicon In Transistors
It will emerge as material of choice in transistors gradually as technological barriers slowly restrict the use of Silicon
Silicon, the second most abundant element in Earth’s crust is at the heart of the IT revolution for the last 50 years. As it is the basic building material of transistors, the electronic switches which power the microprocessors, memory chips, digital control logic etc.
The growth of IT industry implied that the transistors reduced in size while their number on a chip and frequency of operation continued to increase. Nowadays, microprocessors have tens of billions transistors, running at some GHz frequency and a few nanometers in size. However further reduction is size or increase in frequency is hitting roadblocks due to quantum effects and heat generation. This has led to search of alternative materials e.g. Germanium which was used to create first transistor in 1947;III-V materials, such as Gallium Arsenide, Indium Arsenide; materials which are just 1atom layer thick e.g. Tin monoxide, Phosphorene, Graphene, Carbon nanotube etc. But the material which is getting maximum research attention is Graphene.
Graphene an allotrope of Carbon in the form of a two-dimensional hexagonal lattice with each atom at the vertex. At 15000 cm2⋅V−1⋅s−1 its electron mobility is at least 10 times that of Silicon. And its hole mobility is even more favorable as compared to that of Silicon.The same holds for its thermal conductivity which is at least 10 times that of Silicon and hence could offer a solution to heat dissipation. Scientists at University of California at Los Angeles, at IBM etc. have demonstrated Graphene based transistors with speed about 100 GHz. Researchers are aiming to take the speed to THz. Silicon based commercial microprocessors are yet to touch 10 GHz. Graphene is 200 times stronger than steel, so transistors made up of Graphene would be rugged.
However, there are various technical and commercial challenges before Silicon is dethroned. First, Graphene does not has a band gap and in theory is not a semiconductor and more importantly, it cannot work as a transistor which is a switching device. One way to solve this issue is to use 2 layers of Graphene called bilayered. Another approach is to introduce imperfections in structure or impurities in the form ofdoping. The thinness of Graphene implies that it is difficult to produce and is vulnerable to damage when used to create transistors. Its thinness also introduces health hazards during its manufacture. But the biggest challenge is commercial. Silicon is abundant and cheap to produce. Currently, the cost for electronic-grade silicon is only $50 per kilogram, while graphene costs $40,000 per kilogram. 50 years of research has matured the processes to create Silicon based transistors besides investments in billions of dollars have been done. Cost of creating a new foundry is at least $ 1 billion. In fact, Graphene or any other substance would find easier adoption if they canpartially utilize the Silicon based infrastructure and processes. Here again IBM has taken lead and demonstrated creation of Graphene based IC using a process that is compatible with standard Silicon CMOS processes.
Graphene will emerge as material of choice in transistors gradually as technological barriers slowly restrict the use of Silicon. It will initially be used in niche applications e.g. wireless or radio frequency communications, photovoltaics, high frequency devices and then move to heart of semiconductors, i.e. general purpose transistors.
Disclaimer: The views expressed in the article above are those of the authors' and do not necessarily represent or reflect the views of this publishing house