Graphene-on-SiC Heads From Lab To Fab
To fulfil the tremendous potential of graphene, high-quality material must be shipped in significant volumes. One supplier looking to do just that is Graphensic, which has developed graphene-on SiC products for making structures for metrology, high-speed transistors and biosensors. Company founders Rositza Yakimova, Mikael Syväjärvi, and Tihomir Iakimov detail their progress.
Cubic SiC formed at grapheneSiC
It may smack of hype to refer to graphene as the wonder material, but it is an accolade that is richly deserved: This allotrope of carbon, which is based on atomic sheets, is destined to make massive improvements in numerous areas, including the construction of lighter, stronger airplanes, the manufacture of ultra-fast batteries and the production of new or better electronics.
The pioneers of this material are Andre Geim and Konstantin Novoselovfrom the University of Manchester, UK, who reported their ground-breaking discovery of this new form of carbon in 2004. Since then they have shot to fame, netting a Nobel Prize for their efforts in 2010. Following their discovery, research into of graphene has mushroomed, with commercial interest in this material taking off. However, graphene is never going to fulfil its potential as a wonder material that can serve mankind in many ways unless it can be manufactured by industrial processes in high volumes.
At Graphensic AB, a spin-off of Linköping University, Sweden, we are one of a handful of companies addressing this. We produce high-quality films of graphene on SiC substrates with a unique manufacturing method involving high-temperature processing.
For us, forming our company was a natural evolution. When we had been working at Linköping University, we received an increasing number of requests for both material and research collaborations. These requests eventually exceeded those we could manage within unfinanced research collaborations, and at that point in time it was an obvious step to form a company. We did this in late 2011. We both manufacture materials by direct sales and through development projects aiming to build up the company through customer sales.
Development and production of products utilizing graphene demands a steady, reliable supply of this material. Several companies are meeting this need, and one of the key differences between them is that they are producing different types of graphene on different foundations. The graphene that is manufactured can be a single layer of carbon atoms (monolayer graphene), or two or a few layers of this element (bilayer and multilayer graphene, respectively). It can be produced as flakes, or on a substrate, such as a metal or SiC. Graphensic is one of the few companies in the world specialized in producing high quality graphene on SiC.
Flakes can be made by various methods. There is the scotch tape method, which was used by the Nobel Prize winners to make their first samples, and there are also chemical methods, such as those employed by the UK firm Graphene Industries. These processes are attractive from a cost perspective, but the graphene that is formed is small, preventing it from being used for various electronic applications.
Larger sizes are possible by producing graphene on a metal or SiC. Graphenea in Spain provides the latter type of product, which is used to transfer the graphene film to the active region of a device, such as a flexible polymer or silicon. Using graphene in this manner, it is preferable to use a metal rather than SiC as the substrate, due to lower substrate costs.
So what is the benefit of SiC? It's not for transferring graphene to another material, because that transfer process is challenging from technical point of view, due to the close bonding to the substrate, and consequently costly, due expenses associated with chemicals and facilities. But graphene-on-SiC is best suited to applications where the SiC substrate forms either an active part of a device, or acts as a suitable template. This is possible, because SiC offers biocompatibility and chemical inertness.
The key difference between using a metal and SiC as the substrate is that a metal is always conductive, while SiC can be semi-insulating or doped. This equips graphene-on-SiC with an advantage for various electronic applications, and allows this to target markets that are impenetrable by either flakes of graphene and SiC-on-metals products. There are also markets where all three classes of product can compete. In those cases, processing and cost issues will determine which format is most successful.
Substrates and processes
We produce our graphene films on 6H and 4H-SiC substrates. Although these are commercially available in diameters of up to 150 mm, there are some issues regarding substrate defects and the large bandgap to consider. Due to the latter, we are also interested in the development of the cubic form of SiC (3C-SiC) as a foundation for graphene.
In our process for forming graphene, the SiC substrate has a dual role, acting both as a precursor and a substrate for epitaxy. When the substrate is heated to 1500-1600oC in a gas ambient, SiC vapour species start to leave the surface, which rearranges itself to form a buffer layer. Graphene then forms on top.
The nature of the graphene formed by this approach is strongly influenced by the processing temperature. When SiC species vaporize, the ratio of silicon-to-carbon atoms varies, depending on the conditions. Silicon has the higher vapour pressure, so the ratio of silicon to carbon is large at a lower temperature, and decreases with increasing temperature to approach an ideal ratio of 1. Given this, it is favourable to apply a high temperature, such as 2000oC, to liberate similar populations of silicon and carbon atoms.
The process yields high quality monolayer graphene over a large area of a wafer. This success partly results from our strong background in SiC research at Linköping University. At this institution, SiC growth has been developed for almost 20 years. Methods that have been pioneered range from liquid phase epitaxy −which was initiated by a program to make the world's first SiC growth in microgravity using the sounding rocket MASER7 at Esrange in northern Sweden −to various sublimation growth methods.
Experience in SiC growth system opens the potential of large area growth of grapheme
As stated previously, commercially available, hexagonal forms of SiC are mainly used for the formation of graphene. This provides interesting features of the crystal and materials behaviour. That's because these classes of SiC are polar, with opposing sides terminated by either silicon or carbon atoms. These two surfaces have a substantial difference in the surface free energy, and this accounts for the far greater challenge in preparing graphene on the carbon face than the silicon face.
Graphene formation on a SiC wafer.
Another challenge stems from the not perfect orientation to a low index surface. The slight off-axis orientations of the SiC substrates create atomic steps, and when this wafer is heated, the surface rearranges and undergoes a step-bunching process: Initial small steps turn into larger ones, with larger terraces and a step edge. Often monolayer graphene is formed on the terrace, and bilayer graphene on the step edges.
Surface rearrangement is a natural process in SiC, with the first atomic layers rearranging to abuffer layer. In graphene, this is believed to induce doping, which is not wanted in some applications. To prevent this, some researchers are trying to expose graphene on SiC to certain elements that can penetrate beneath the graphene and change the buffer layer into a graphene layer. Success in this endeavour creates bilayer graphene without a buffer layer.
Low-energy electron microscopy images of 3C (left) and 6H (right) showing large domains (>50 µm) and dominating monolayer graphene with dark areas of bilayer graphene. Images: Alexei Zakharov, MaxLab.
Many researchers around the world are investigating the properties of monolayer, bilayer, and multilayer graphene on silicon and carbon faces. One highlight of these efforts is associated with the development of monolayer graphene with a buffer layer, which has shown outstanding performance in metrology. We have found that our graphene on SiC provides a resistance standard in quantum Hall measurements that is several orders of magnitude better than the current one based on GaAs (quantum Hall measurements relates Planck’s constant, h, to the electron charge, e). This could aid the International System of Units, providing quantum units of mass and current based on these fundamental constants of nature.
Another exciting opportunity for graphene-on-SiC is its use in the creation of a monolithic transistor for combining an on/off ratio or more than 104 with the absence of damping at megahertz frequencies. Fabrication, in its most simple form, requires just a single lithography step to build transistors, diodes, resistors and eventually integrated circuits, without the need of metallic interconnects.
One hurdle to the realization of such circuits is the lack of a bandgap for graphene. However, this can be addressed by turning to ribbons of graphene, which have a bandgap of 0.5 eV and can be produced by making forced topographical changes on SiC. Note that it is not possible to modify a metal substrate so that it yields graphene ribbons.
Yet another area where graphene-on-SiC could make a commercial impact is in the field of biosensors. Electrical properties of graphene channels in transistors are influenced by minor perturbations, such as molecules on the surface. This high degree of sensitivity stems from the combination of a high surface-to-volume ratio and tuneable electron transport properties, which result from quantum confinement effects. The upshot of this is that these devices have the potential to detect single molecules.
Biosensor operation is based on the use of a target disease biomarker. This provides a change in the surface charge density, which is detected as an electrical signal. SiC and graphene are very promising materials in this field, because they exhibit excellent biocompatibility with in vivoand in vitrostudies, and they show no cytotoxicity responses. Commercial opportunities include the development of miniaturised systems for the detection of disease biomarkers for use in the early diagnosis and monitoring of diseases.
Development of graphene on SiC is on going, and we are keen to pursue growth of the cubic form of this wide bandgap material. It is rare in bulk-like form, but it appears that this platform could create structures that are free from the buffer layer. We have developed a sublimation method for growth of 3C(111) on hexagonal SiC substrates at a high growth rate of up to 1 mm/hr. Traditionally this material is grown on silicon, but this causes stress by a high thermal and lattice mismatch. Our aim is to be able to provide substrates of 3C using this approach.
Although all forms of SiC are not cheap platforms for graphene, we are convinced that this class of material system has great commercial opportunity. We are shipping material to customers from all around the world, and as we watch the number of publications reporting advances in graphene-on-SiC rise, we are confident that this will spur the establishment of markets where this form of graphene dominates. Commercial exploitation of this material is yet to begin, but given its wonderful set of attributes, what it can realise is only limited by our imaginations.
Company founders Rositza Yakimova, Mikael Syväjärvi, and Tihomir Iakimov