4.7 Article

Chip-Level Multiple Quantum Well Modulator-Based Optical Interconnects

Journal

JOURNAL OF LIGHTWAVE TECHNOLOGY
Volume 31, Issue 24, Pages 4166-4174

Publisher

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JLT.2013.2290974

Keywords

Multiple quantum well modulators; optical interconnects; optical waveguides; photonic integrated circuits

Funding

  1. DARPA under ARL [W911NF-09-1-0080]

Ask authors/readers for more resources

High performance computing systems are becoming increasingly limited by the capacity of interconnects due to the continued scaling down of CMOS critical dimensions, resulting in the implementation of optical interconnects at ever decreasing distances. At the chip level, the communication bottleneck and energy consumption per bit are now major limitations to the continued performance scaling of microprocessors. In this paper, a novel integrated photonic approach is presented that uses polymer waveguides and surface-normal GaAs/AlAs multiple quantum well devices integrated directly onto a silicon chip. The concept provides sub-pJ/b performance and seamless interfacing between the on-and off-chip domains. This is the first demonstrated waveguide-coupled surface-normal MQW-based approach to be fully integrated within a photonic layer and to a large extent mitigates packaging issues for future photonics systems integrated with Si chips. Key aspects of the architecture are efficient and minimum-footprint optical fabrics and low-power-consuming optical transceivers. Gray-scale lithography is used to fabricate the 3-D coupling structures directly in the waveguide polymer layer. Analyses and experimental results show that the optical fabric concept provides the necessary bandwidth density and low power consumption for future chip-scale interconnections.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available