FEATURE: PHOTONIC INTEGRATION
THE QUESTION WAS: CAN WE BUILD AN ELECTRO- OPTIC DEVICE LIKE A PHOTODETECTOR OR MODULATOR?
semiconductor industry. ‘In order for graphene to be applied in most electronic or optoelectronic applications, you need to first grow it inside a CVD reactor as a wafer – traditionally six-inch but we’re moving to eight- inch,’ she explained. ‘Ten you need to transfer it onto a silicon-based wafer. We do all that.’
InP or LiNbO3 devices, although Templ acknowledges that there is fast progress.
From lab to fab Another important challenge is geting graphene component production from research labs into large-scale production. ‘Seting up mass production of components with reproducible high quality and good yield observing all the commercial requirements and constraints is a huge task on its own,’ Templ said. ‘Te pilot line, initiated by the European Graphene Flagship project, is an important first step in that direction. But it will be necessary that large component manufacturers adopt the results and take over as soon as a certain level of maturity has been reached.’ Graphenea, in San Sebastián, Spain, is another
Graphene Flagship member, supplying the graphene material in many cases, explains Amaia Zurutuza, the company’s scientific director. She highlights that using graphene in modulators and detectors for optical communications could reduce costs, as well as improving speed. Zurutuza recognises that graphene is more expensive than silicon, reflecting its small market. ‘As any material, it goes with economies of scale,’ Zuruzuta said. ‘In the history of Graphenea, since 2010 our prices have gone down dramatically. We don’t think this is going to be a limiting factor.’ Importantly, Graphenea’s processes are carefully engineered to fit with the existing
www.fibre-systems.com @fibresystemsmag
Mass production Reflecting the need for mass production capabilities, Graphenea is now moving to fabricating devices. ‘Tere is no foundry for graphene,’ said Zurutuza. Yet the details of that transition will need to be resolved, she acknowledged. Examples include finding ways to make graphene uniform on large wafers, and to deposit metal contacts reproducibly. Interfacing graphene with silicon well is essential for integration with conventional electronics. ‘All these steps haven’t been established because it’s a new material,’ Zurutuza noted. ‘So, you need to improve and optimise each step first with a small wafer, then you increase the size and number of wafers and then you need to do that reproducibly.’ Antonio D’Errico, senior researcher at Ericsson
in Pisa, believes that his company has a clear idea of how the value chain that combines graphene with silicon should be made up.
He notes that Ericsson’s desire is to have the supply chain in Europe. He highlights that there is a three-year window until funding for the Graphene Flagship runs out to fully establish the risks and benefits of further investment. Prospects based on experimental results so far are ‘really encouraging’, he said.
Cleanly ranging further Benefits come because graphene’s high electron mobility influences the critical role electrons and holes as charge carriers play in photodetectors and modulators. Charge carriers form when photodetectors absorb light. In modulators, charge carriers help control the modulation process, for example by switching graphene between a state where it absorbs light, and a state where it lets light propagate. As such, carrier mobility and lifetime ultimately determine the performance of optical communication systems. Current materials limit networks to 50GBd/s capacity, according to D’Errico, but graphene can overcome that. ‘We can efficiently rise up to 100GBd/s, if we have the electronics to drive it,’ he said. And because one atomic layer of graphene
enhances electro-optical modulation efficiency, it can enable modulators around 100µm thick long, D’Errico says. ‘Today, you need to drive a Mach Zehnder modulator hundreds of millimetres long based on LiNBO3 with 2-5V,’ he added. ‘We aim to demonstrate on a graphene- on-silicon Mach Zehnder modulator the driving voltage can be lower than 1V and compatible with CMOS technology.’ In switching between absorbing light and
leting it propagate, graphene releases chirp onto optical signals, D’Errico said, varying their phase. Tis is helpful because chromatic dispersion as signals travel in optical fibre, also generate chirp that limits signal range. Graphene’s chirp compensates for chromatic dispersion, under control of applied voltage. Te Ericsson team has studied this compensation in 10Gb/s signal propagation up
Graphenea’s processes are carefully engineered to fit with the existing semiconductor industry Issue 26 n Winter 2020 n FiBRE SYSTEMS 11
Graphenea
Shutterstock
Google
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40
