For non-scientists this project webpage may raise different questions about the aim of PHELICITI and the exploitation of its findings. In a try to answer them without scientific and technical terminology, the following sections address the most important questions that may have been raised by the interested reader. In case your interest did not get satisfied, feel free to contact the technical coordinator for further information.
What is it all about?
In a nutshell PHELICITI is a technology project funded by Austrian research programmes and develops a next generation of integrated chips, similar as they are nowadays found inside personal computers or mobile phones. The big difference, however, is the presence of a second technology layer in these chips that will significantly boost their performance.
Conventional electronics chips as produced in masses and included in every electronic device have enabled the modern information age, reaching from telephony and television to the rollout of the modern internet and massive computational powers. The technology behind has been successively improved during the last decades. It is now possible to host 100s of millions transistors on a microprocessor with a size smaller than a fingernail – produced at a cost that guarantees technology that is accessible for everybody. Such silicon integrated chips use so-called CMOS processes which allow access to geometrical on-chip structures in the sub-micrometer range.
Nevertheless, when it comes to massive workloads as required by the staggering information capacities required for modern services, electronic chip solutions are reaching their limits. For this reason PHELICITI introduces a second “key enabling technology”, namely photonics.
Why do we need photonics in addition to electronics?
…simply because technology based on light provides a better performance for simple tasks than technology based on electrons. An important fact is that the particles of light, the so-called photons, are much more “robust” for purposes of communications than their electrical counterparts, the electrons. For example, it is possible to transmit ~10.000-times more data over a single optical fibre compared to an electrical cable and on top of this, the overall communication link is more energy efficient, leading to a greener solution and a global reduction of wasted electrical power.
When talking about “links”, one can imagine different use cases for photonic technologies. On the one hand, these can be communication links between buildings, cities or continents: At the moment, it is possible to transmit more then 100 Tb/s over a fibre as thin as a human hair. Moreover, photonic links allow us to operate networks over much longer distances, e.g. more than 10.000 kilometres.
On the other hand, these links can be also between processor cores or between a processor and memory chips hosted on the same motherboard. Also here it is required to reach capacities in the order of more than a Tb/s when it comes to data centres. Note that a Terabit/second is ~20.000-times the capacity a wireless LAN link can achieve.
Still, although photonics provides access to super-fast communication links, there are yet no intelligent photonic building blocks that can perform mathematical operations as a microprocessor is easily performing. For this reason, the interplay between both worlds, photonics and electronics, is as important as the inclusion of a photonic layer in the next-generation of “optoelectronic” chips. This kind of chip can be thought like two separate chip layers, one for photonics and one for electronics, being brought together like the two breads of a sandwich and acting as a single chip device thanks to an interposer technology that enables the seamless interplay between photonic and electronic chip components.
What are the specific goals of PHELICITI and how does its outcome look like?
PHELICITI aims at developing the technological building blocks for integrating all functions required for applications in telecommunications. This includes the physical design of photonic modulators, photodetectors, passive photonic functions for signal manipulation, electronic amplifiers and other electronic circuitry optimised for the co-integration with novel photonic components. At the same time the “connections” between photonic and electronic components needs to be developed and optimised in order to unlock the synergy of the two technology layers that leads to the good performance aimed at. One key challenge to this end is to retain conventional CMOS processes for processing wafers in silicon chip manufacturing.
The output of the project will be a chip that can be flexibly used in many fields of telecom applications, for example as an end-user internet modem for a high data rate of 10-80 Gb/s for each household. The difference to conventional electronic chips will not really be visible – the only difference in its appearance will be a thin optical fibre that originates at the surface of the chip package used to interact with other optoelectronic chips through a purely photonic interconnect with massive data throughput.
