IMAGE: Artistic drawing of the electro-absorb model based on graphene. view more
Credit: © ICFO / D. Alcaraz
Over the past few years, global data traffic has been successful, with more than 12.5 billion connected devices worldwide. The current worldwide use of 5G telecommunications standard drives the need for smaller devices with improved performance, such as higher speed, lower power consumption and lower cost as well as easier manufacturing .
In finding the right technology, photonic devices have emerged as the leading technology for the evolution of information and communication technologies, surpassing the capabilities of conventional micro-technologies and CMOS technologies.
Optical communication systems rely on three basic components: modulators, waveguides and photodetectors. Light switching is essential for integrated photonic circuits as it allows multiple signals to be transmitted simultaneously over a single channel. In particular, electro-absorbent (EA) modulators alter the amount of light that passes through the optical wavelength.
To date, silicon and graphene are winning the race by proving to be the most scalable, cost-effective and CMOS-compatible materials for optical modeling and detection. Graphene-based modulators have already demonstrated broadband optical bandwidth and temperature stability, but at times have not been able to demonstrate both high-speed efficiency and high switching at the same time. , due to the limited quality of graphene and the mixture between the graphene and the dielectric material.
Now, in a study published in Nature Communication, ICFO researchers Hitesh Agarwal, Bernat Terrés, Lorenzo OrsinI, led by ICFO ICREA Professor Frank Koppens, in collaboration with researchers from Universita di Pisa, CNIT, Ghent-IMEC University, and NIMS has reported a novel EA modulator capable of a 3-fold Increase in static and dynamic conversion efficiency while increasing high speed, a value that exceeds those for graphene EA modulators. previously reported.
To achieve this, the team of researchers developed a graphene-based electro-absorbent modulator by combining high-quality graphene and high-k dielectric, also used in microelectronics. The high quality of graphene was achieved by fusing it with the 2D-hexatic dielectric boron nitride (hBN). Interestingly enough, the team was then able to inject the high-k HfO2 dielectric material between two layers of boron nitride, which allowed it to work at much lower voltages, and, at the same time, achieves symmetry and hysteresis due to the high quality of the graphene. By doing this, the dielectric combination was able to increase the capacity of the EA modulator without compromising the strength of the machine against high voltage, maintaining high cost shift (increasing switching efficiency) while and increased low levels of doping.
As Hitesh Agarwal, a researcher at ICFO and first author of the study, says, “Because one of the key bottles for the integration of graphene into CMOS fab lines is as incompatible with high-k oxygen, encourage here we are to build the hBN structure. -HfO2-hBN. Not only have we been able to achieve high switching efficiency (due to high-K dielectric), but also higher speed (due to increased mobility). “
“We’ve been waiting for some time to see the unique basic capabilities of graphene in applications,” said Bernat Terrés, ICFO’s Postdoc researcher and author of the work’s reference. He also confirms that “Optoelectronics is one of the first where this 2D material overcomes the latest conventional technologies, providing an optimistic outlook for other commercial applications”.
In summary, the device was capable of outperforming previous modulators, operating at high speeds and, achieving very high switching efficiency, low power consumption, achieving bandwidth 39GHz which was breaking down, with operations up to 40Gbps speeds, thus exceeding the basic limits achieved so far with dual-layered graphene systems.
The compatibility of this device with silicon and microelectronics technology could help with the scaling developments that lie ahead today for the photonic industry as well as implement this type of technology for a much larger range of applications in electronic and optoelectronic applications. Such results could certainly be beneficial to applications for high-speed and low-latency optical networks such as autonomous vehicles, remote surgery, IoT, to name a few.
Research colleague Marco Romagnoli, a researcher at CNIT and director of a work package at Graphene Flagship, says, “This new scientific product of ultrafast electro modulation paves the way for the continuous race for speed features the highest electro-optical bandwidth achieved with graphene and 2D materials. Moreover, this work is also the first example of full 3D integration achieved by combining different types of 2D materials demonstrating the potential of these new paths in microfabrication of integrated circuits. ”
In addition, Wolfgang Templ, from Nokia Bell Labs, emphasizes “This work demonstrates that the discussed 2D-3D dielectric integration of high-quality double-graphene structures can double the opening the way to the production of new high-performance and miniature electro modulators (EAMs) that can be combined with Si-based electronics into the highly advanced photonic circuits ”.
Finally, Frank Koppens, ICREA Professor at ICFO and work package director at Graphene Flag, points out that “data mining is growing rapidly and will bring great benefits to society, for example by enabling autonomous vehicles Power consumption of high levels of data traffic, however, is a key challenge that needs to be addressed I am pleased to see that graphene-based modulators with much lower power consumption, as can be seen in this work, can address to two social challenges at the same time. “
###
Link to paper: https: /
Link to ICFO-led ICREA Prof research group Frank Koppens: https: /
Link to graphene at ICFO: http: // graphene.
Link to major graphene locations: https: /
Disclaimer: AAAS and EurekAlert! they are not responsible for the accuracy of press releases posted to EurekAlert! by attacking sites or for using any information through the EurekAlert system.