FP7Reintegration grant2013–2016

NAPOLI · Nano Photonic Optical Link

FP7 — People (Marie Curie Actions)

Duration
2013-09-01 → 2016-07-31
EU contribution
€72,917
Participants
1
Scheme
MC-CIG

Lines connect the coordinator with its partners.

Results in brief

Nano Photonic Optical Link

Today’s information society relies on huge computer systems to handle a massive amount of data that grows every day. Computers the size of warehouses are used, where the interconnection of individual processing cores is limiting the total system performance. The interconnects already take up about 50% of the systems power consumption and this value will drastically increase with system complexity and data capacity. This “interconnect bottleneck” can only be overcome by employing optical interconnects for rack-to-rack communication. To sustain the growth of computer performance, we need to develop optical technologies that can be applied at ever shorter distances down to connecting processing cores on the same chip. This implies lasers and photonic components, which are more energy efficient, cheaper and, above all, much smaller. The CIG Grant Napoli allows us to contribute to a better fundamental understanding of light-matter interaction in nanoscale cavities and develop novel technologies for integration of nanoscale devices in a versatile photonic platform. Special care needs to be taken when designing laser cavities at the wavelength scale. We studied a number of promising novel cavity concepts to find out which are most suitable for small and efficient lasers. Sub-wavelength confinement of light at metal interfaces becomes possible with plasmonic structures [1]. Previously we have worked on plasmonic lasers in which the light is concentrated in active regions with dimensions down to 20 nm. Cavity losses in these lasers are very high, however, and it is difficult to get them lasing. In metallo-dielectric structures, which are on the scale of a wavelength, we use the high reflectivity of metal rather than the plasmonic properties, enhancing efficiency [2,3]. Based on this concept, we investigated the realization of the pillar lasers with an active region smaller than 0.2 µm². They are coated with Silicon Oxide and a Silver film to form a metallo-dielectric cavity. On route towards a lasing device we demonstrated highly efficient light emission from a nanoscale metal clad LED. Finally, we considered the use of photonic crystals for micrometer scale lasers. The precise arrangement of periodic holes in a semiconductor beam allows the control of guiding and reflection properties at the wavelength scale. We designed and simulated devices with promising efficiencies exceeding 20%. With fabrication tests and the development of selective area regrowth we showed the feasibility of making such complex devices in our clean room. Though scientifically successful and promising, the goals of this Career Integration Grant could not be realized fully because the main researcher has accepted a job in an industrial R&D laboratory. This lead to the premature end of the grant after 2 years. As a consequence the ambitious goals of the original grant proposal, i.e. the creation of a nano-scale laser and densely integrated optical links, were not reached. Nevertheless, the technologies established during this project positively aid the ongoing development of our micron-scale photonic platform IMOS. The NAPOLI project most noticeable contributed to the IMOS development with: - low optical loss ohmic contacts for compact active devices. - highly efficient metal grating couplers. - short adiabatic tapers for compact grating couplers. - design of a twin-guide layer stack for membrane semiconductor optical amplifiers. - compact pin-diode detectors. - selective area regrowth for active-passive integration in photonic membranes. The IMOS platform is pioneering the attachment of thin, high-index membranes with photonic ICs bonded directly on top of Silicon and CMOS wafers. It promises higher integration densities and bandwidth, better sensitivities and lower energy consumption through the use of membrane photonics and the tight co-integration of driver and control electronics. This should enable new applications in optical communication and sensing. [1] M. Hill, “Status and prospects for metallic and plasmonic nano-lasers“, J. Opt. Soc. Am. B, Vol. 27, No. 11, 2010. [2] Dolores Calzadilla, V.M., Geluk, C.T.T., Vries, T. de, Smalbrugge, E., Veldhoven, P.J. van, Ambrosius, H.P.M.M., Heiss, D., Fiore, A. & Smit, M.K. (2013). Fabrication technology of metal-cavity nanolasers in III-V membranes on silicon. Conference Paper : Proceedings of the 18th Annual Symposium of the IEEE Photonics Benelux Chapter, 25-26 November 2013, Eindhoven, The Netherlands, (pp. 243-246). Eindhoven: Technische Universiteit Eindhoven. [3] Heiss, D., Dolores Calzadilla, V.M., Fiore, A. & Smit, M.K. (2013). Design of a waveguide-coupled nanolaser for photonic integration. In H. Chang, V. Tolstikhin, T Krauss & M Watts (Eds.), Conference Paper : Integrated Photonics Research, Silicon and Nanophotonics (IPRSN), 14-17 July 2013, Rio Grande, (pp. IM2A.3). Rio Grande: Optical Society of America.

Data: CORDIS, © European Union

Project objective

In today’s ware house sized computers the interconnection of individual processing cores is limiting the total system performance. Interconnects already take up about 50% of the systems power consumption and this value increases with system complexity. Optical interconnects are nowadays employed for rack-to-rack communication to overcome this “interconnect bottleneck” by reducing space requirements and power consumption. In the future, such optical technologies must penetrate deeper in the system design and be applied for chip-to-chip, or even on-chip interconnection to sustain the exponential growth of computer performance. New approaches are needed to meet the extreme requirements in integration density, power consumption and cost for optical interconnects in future high performance computers. By exploring the limits of miniaturization and energy efficiency in integrated active optical components we want to demonstrate a compact optical link and assess the potential of nanophotonic technology for integrated optical chip-to-chip or even on-chip interconnects. We will employ metallo-dielectric cavities to shrink the footprint of devices, which in turn will increase operation speed, reduce power consumption and allow efficient cooling of the highly integrated devices through the metal surfaces. We will develop a first waveguide coupled nanolaser, demonstrate optical detectors with record small footprint and demonstrate for the first time an optical interconnect that satisfies the requirements of future computing systems with respect to transmission density, power consumption and device size. The acquired techniques will directly contribute to the development of a photonic technology platform. Such innovations in photonic technology are essential to overcome the interconnect bottleneck and enable next generation computing technology.

Original text from CORDIS.

Participants

  • TECHNISCHE UNIVERSITEIT EINDHOVEN · EindhovenCoordinatorNetherlands

Links

Data: CORDIS, © European Union