From valves to microprocessors and beyond
Electronics have come a long way since the development of the vacuum tube. Around 70 years ago, these bulky, fragile objects, known colloquially as valves, gave way to semiconductor transistors. Reduced to the nanoscale, and assembled in their millions and billions, transistors form the basis of today’s computer processors.
There is a developmental rule of thumb known as Moore’s Law, which states that the number of transistors in an integrated circuit doubles every two years. The law has so far proven accurate, with the maximum transistor count in a microprocessor having reached over four billion. When it comes to component size, Moore’s Law is running up against the laws of physics, but there are issues involved in addition to physical scale, the principal ones being cost and power consumption.
Searching for new materials
The electronics industry is increasingly focused on mobile devices which could benefit from the replacement of silicon-based transistors with components that draw significantly less power. It is hoped that low-power transistors made from graphene and related two-dimensional materials will re-energise the electronics industry, leading to disruptive technologies and products that to date remain the province of innovative minds and fertile imaginations.
Looking at the bigger picture
The potential for graphene and related materials in electronics is the subject of a review
recently published in the journal Nature Nanotechnology
. Five of the eight authors are associated with Europe’s Graphene Flagship
The Graphene Flagship is an international consortium of academic and industrial partners, funded in part by the European Commission, which focuses on the need for Europe to address the big scientific and technological challenges through long-term, multidisciplinary research and development efforts.
CMOS and two-dimensional materials
Led by device modelling expert Gianluca Fiori
, based at the Information Engineering department at the University of Pisa, and Francesco Bonaccorso at Graphene Labs of the Italian Institute of Technology
, the authors of the review article look at the current status of electronic devices based on two-dimensional materials. They also discuss prospects for the development of next-generation devices, and do so with a mid- to long-term perspective.
Commenting on the situation today, Gianluca Fiori says: “We currently face technical issues which can only be overcome through the definition of new device architectures and exploitation of new materials. Two-dimensional materials provide an option for ultra-thin transistors in low power electronics.”
The Nature Nanotechnology
review focuses on established CMOS (Complementary Metal Oxide Semiconductor) technology. It does not cover proof-of-concept devices which exploit spintronics and similar concepts.
Production and processing
Production processes are key to the exploitation of graphene and related materials, and this is especially true when it comes to the tuning of structural and electronic properties at the atomic level.
Two-dimensional materials may be produced using a number of methods, the most common of which involves depositing monolayers of materials of interest onto metal substrates. The process known as chemical vapour deposition (CVD) can produce high-quality two-dimensional materials, but it requires high temperatures, catalyst residues may contaminate the materials, and metal substrates can have a significant effect on their electronic properties. Lower temperature processes such as molecular beam epitaxy and atomic layer deposition are possible alternatives to CVD.
One promising approach to the industrial-scale production of two-dimensional nanomaterials is liquid-phase exfoliation, and the technique is well suited to printed and flexible electronics applications based on low-cost, roll-to-roll manufacturing processes. The structural and electronic qualities of two-dimensional films produced by liquid-phase exfoliation are lower than those obtained by mechanical exfoliation (e.g., the sticky-tape technique) or CVD, but the cost and scaling advantages of printable inks are obvious, and for many applications the quality of the material produced is sufficient.
Graphene has many uses, but, when it comes to field-effect transistors (FETs), the material cannot comply with the requirements of the International Technology Roadmap (ITRS) owing to its lack of an electronic band gap. In other words, the high electrical conductivity of graphene in its pure form means that it does not function well as an insulator or semiconductor. However, graphene is just one of a large number of two-dimensional materials, some of which have semiconductor properties that make them suitable for application in digital electronics.
The Nature Nanotechnology
review looks in particular at the class of materials known as transition metal dichalcogenides (TMDs). One well-known example of a TMD is molybdenum disulphide (MoS2), which has been used to construct FETs with small leakage currents and other qualities which approach those desired of FETs for CMOS circuits.
Transistors based on MoS2 can comply with the ITRS down to a minimum channel length of 8 nm, which is a significant improvement over silicon. The relatively low electron mobility in MoS2 is not an issue, given that the transport of electrons at such small scales is supposed to be effectively ballistic. This is another way of saying that electric charge carriers in the material encounter negligible scattering.
High-frequency analogue electronics based on silicon semiconductor transistors is a mature technology, with circuits operating at speeds of up to a few hundred gigahertz. But we are fast approaching a bottleneck imposed by material properties, and this when there is huge interest in expanding deep into the terahertz regime, for example in communications systems and sensor applications.
Given its high electrical conductivity, graphene is a suitable material for radio-frequency electronics applications, but the best reported maximum oscillation frequency for graphene is a few hundred gigahertz, whereas with indium phosphide devices it is 1.2 THz. Improving the conductivity of graphene will raise the maximum frequency, but it remains uncertain as to whether this will be sufficient to bring it above 1 THz.
Alternative two-dimensional materials have their pros and cons when it comes to high-frequency applications. One possibility is non-standard transistor geometries based on combinations of different materials. These could make terahertz frequencies accessible without struggling with the absence of a band gap in pure graphene.
From components to integrated circuits
Moving from single components based on graphene and other two-dimensional materials to integrated circuits involves a complex fabrication process. So far we have seen voltage amplifiers, mixers, frequency doublers, oscillators and combinations of such components, but in all cases the circuits have transistors with a gate length of at least a few hundred nanometres. This is considerably longer than that of state-of-the-art silicon devices.
Graphene is not yet in a position to surpass the performance of mature and ever-improving silicon semiconductor technologies. That said, the unique properties of two-dimensional materials, together with the rapid improvements seen in recent years, could soon give them the advantage.
The are many applications for flexible electronics, and considerable media attention has been devoted to the potential of graphene in this area. Earlier this year, Plastic Logic
and the Cambridge Graphene Centre
, a partner of the Graphene Flagship, demonstrated
the world’s first flexible display with graphene incorporated into the pixel backplane. This opens the way to the manufacture of flexible devices with roll-to-roll manufacturing techniques.
Two-dimensional materials are in principle a perfect solution for flexible electronics. They are extremely thin and transparent, and their chemical structure makes them very strong. They can also be prepared over relatively large areas, and transferred onto a number of flexible substrates, among them common industrial polyesters such as polyethylene naphthalate.
Material characteristics are critical in electronics applications, and there are practical trade-offs to be made between quality and ease and cost of fabrication. Graphene can be grown over large areas by CVD, but this is an expensive process. Producing and manipulating two-dimensional materials in solution is cheaper and more scalable, but the electronic performance of the materials will differ from that achieved with CVD.
We already have demonstration devices incorporating chemical and biological sensors, and the Nature Nanotechnology
review looks at a couple of these. In the case of body motion sensors with graphene infused into rubber, the process is cheap and highly scalable. Graphene and other two-dimensional nanomaterials are guaranteed to feature prominently in the flexible electronics field.
The way ahead for graphene in electronics
The potential for graphene and other two-dimensional materials in electronics applications is clear, and the question is whether we have the right materials and devices to enable the fast, low-power systems envisaged. Opportunities and challenges are laid out clearly in the Nature Nanotechnology
review, with the principal challenge being the need to create flat two-dimensional films over large areas, with sufficient yield such that devices constructed with these novel materials are viable.“Two-dimensional materials have huge potential for beyond-CMOS technology,” says Francesco Bonaccorso. “But material optimisation and industrial process engineering are required to translate these materials from laboratories to industry. The development of functional inks based on two-dimensional materials is the gateway for the advance of a new generation of printed, flexible and stretchable electronics.”
Once the technical issues discussed here are addressed, two-dimensional materials will lead to a whole new generation of electronic devices, some of them flexible. They will create entirely new markets for the electronics industry, with practical applications that have hitherto been impossible with standard semiconductor materials.
Two-dimensional technology will certainly be disruptive, in the positive sense of that word. But contrary to the impression given in much of the mainstream media, graphene and related materials will not necessarily displace established electronic materials. This is important to note when using the language of technological revolutions and paradigm shifts.
For media enquiries relating to the research and development activities of the Graphene Flagship, please contact Dr Francis Sedgemore.