Final Report Summary - NEXCENTRIC (Next-generation on-chip supercontinuum light sources based on graphene-enabled extreme nonlinear optics)
Nonlinear optics is the scientific discipline where light and matter interact with each other in a nonlinear manner e.g. to create new wavelengths. To develop compact nonlinear-optical devices that could be widely used in telecom, biomedical imaging and other application domains, it is a natural strategy to fabricate these devices on photonic chips using nanoscale waveguides where the nonlinear interaction between light and matter can be made very intense. Yet, the state-of-the-art on-chip nonlinear-optical devices are not yet suitable for widespread deployment as most of them are pumped with impractical pump lasers and/or rely on non-standard waveguide designs which cannot be manufactured in large volumes.
In the ERC NEXCENTRIC project it has been our aim to develop on-chip nonlinear-optical devices made of standard waveguide designs and pumped with practical near-infrared lasers, while at the same time enabling unprecedented device performances so that a wide range of new wavelengths could be created and so-called supercontinuum generation could be achieved. To make this possible, we have introduced the nonlinear-optical two-dimensional material of graphene as cover layer on top of the standard waveguides. Below one can find our main findings resulting from the NEXCENTRIC project:
- Graphene has a very strong third-order nonlinear-optical response that is orders of magnitude higher than that of conventional materials, and at the same time it is a very challenging material to investigate. Particularly, its nonlinear-optical behavior is totally different from what was assumed previously in the literature. Indeed, whereas third-order nonlinear processes such as self-phase modulation (SPM) and four-wave mixing (FWM) in undoped or weakly doped graphene were previously interpreted as so-called Kerr processes in a perturbative framework, we have shown theoretically and experimentally that the actual physics at play relies on saturable photoexcited carrier refraction (SPCR) in a non-perturbative framework. This way we have solved the long-existing paradox of the discrepant graphene nonlinearity predictions and experimental observations reported over the past decade.
- As a result of this novel SPCR physics, near-infrared-pumped graphene exhibits a strong nonlinearity with a negative sign rather than a positive one. As such, when depositing graphene on a waveguide material like silicon featuring by itself a strong positive nonlinearity, there will be a significant counteraction between the two materials and the overall nonlinear response of the graphene-covered waveguide will be rather weak. If, however, graphene is transferred on waveguide materials like silica or silicon nitride with a much smaller positive nonlinearity than silicon, the total nonlinear response of the graphene-cladded waveguide can be very pronounced. As such, supercontinuum generation e.g. by cascading SPM-based spectral broadening and FWM becomes possible provided that also the light absorption in the graphene top layer is properly tailored and that the wavelength spacing considered for the FWM process falls within the range of high nonlinearity.
- Another consequence of the novel physics at play is that graphene-covered silica-core waveguides can exhibit a strong nonlinear response already at low pump powers. As such, the minimally required pump power levels for on-chip supercontinuum generation sources can be substantially reduced, further improving the practical applicability of these sources. The latter plays a key role in the rationale of the NEXCENTRIC project.
To conclude, the NEXCENTRIC project has had a disruptive impact on the in-depth understanding of graphene’s nonlinear-optical behavior, where non-perturbative nonlinear effects can dominate over the usually studied perturbative phenomena. Through our breakthroughs in modeling, fabrication and experimental device demonstration and through our publications in prestigious journals such as Nature Communications, we have impacted both fundamental graphene science and integrated nonlinear optics research while also opening up new routes for widely deployable on-chip supercontinuum light sources.
In the ERC NEXCENTRIC project it has been our aim to develop on-chip nonlinear-optical devices made of standard waveguide designs and pumped with practical near-infrared lasers, while at the same time enabling unprecedented device performances so that a wide range of new wavelengths could be created and so-called supercontinuum generation could be achieved. To make this possible, we have introduced the nonlinear-optical two-dimensional material of graphene as cover layer on top of the standard waveguides. Below one can find our main findings resulting from the NEXCENTRIC project:
- Graphene has a very strong third-order nonlinear-optical response that is orders of magnitude higher than that of conventional materials, and at the same time it is a very challenging material to investigate. Particularly, its nonlinear-optical behavior is totally different from what was assumed previously in the literature. Indeed, whereas third-order nonlinear processes such as self-phase modulation (SPM) and four-wave mixing (FWM) in undoped or weakly doped graphene were previously interpreted as so-called Kerr processes in a perturbative framework, we have shown theoretically and experimentally that the actual physics at play relies on saturable photoexcited carrier refraction (SPCR) in a non-perturbative framework. This way we have solved the long-existing paradox of the discrepant graphene nonlinearity predictions and experimental observations reported over the past decade.
- As a result of this novel SPCR physics, near-infrared-pumped graphene exhibits a strong nonlinearity with a negative sign rather than a positive one. As such, when depositing graphene on a waveguide material like silicon featuring by itself a strong positive nonlinearity, there will be a significant counteraction between the two materials and the overall nonlinear response of the graphene-covered waveguide will be rather weak. If, however, graphene is transferred on waveguide materials like silica or silicon nitride with a much smaller positive nonlinearity than silicon, the total nonlinear response of the graphene-cladded waveguide can be very pronounced. As such, supercontinuum generation e.g. by cascading SPM-based spectral broadening and FWM becomes possible provided that also the light absorption in the graphene top layer is properly tailored and that the wavelength spacing considered for the FWM process falls within the range of high nonlinearity.
- Another consequence of the novel physics at play is that graphene-covered silica-core waveguides can exhibit a strong nonlinear response already at low pump powers. As such, the minimally required pump power levels for on-chip supercontinuum generation sources can be substantially reduced, further improving the practical applicability of these sources. The latter plays a key role in the rationale of the NEXCENTRIC project.
To conclude, the NEXCENTRIC project has had a disruptive impact on the in-depth understanding of graphene’s nonlinear-optical behavior, where non-perturbative nonlinear effects can dominate over the usually studied perturbative phenomena. Through our breakthroughs in modeling, fabrication and experimental device demonstration and through our publications in prestigious journals such as Nature Communications, we have impacted both fundamental graphene science and integrated nonlinear optics research while also opening up new routes for widely deployable on-chip supercontinuum light sources.