Periodic Reporting for period 4 - PLASMIC (Plasmonically-enhanced III-V nanowire lasers on silicon for integrated communications )
Okres sprawozdawczy: 2020-10-01 do 2021-09-30
Electronics provide means for complex computations with a versatile and nanoscale hardware foundation. Photonics on the other hand deliver extremely fast, high data density and low loss communication channels. Thus, scientists have long aspired towards enabling continued growth of electronic circuit complexity using photonic interconnects.
In order to combine electronics and photonics a game-changing technology is required, which possesses the same capacity and speed as photonics, but which is scalable towards the realm of electronics.
Another main limitation is found at a materials level. Si is an indirect bandgap material; hence, it is inefficient for light emission. Integration of direct-gap III-V materials used for light emission and detection in the near-IR entails substantial technological challenges; due to the large lattice- and thermal-mismatch high-quality III-Vs cannot simply be grown on Si. and thick buffer layers must be used to reduce the amount of defects.
In addition, hybrid photonic-plasmonic cavities provides a path for scaling dimensions beyond the diffraction limit. PLASMIC addresses the monolithic integration of active III-V emitters and detectors on silicon and seamlessly integrated with silicon passives, and the use of metals to improve scalability and wavelength stability.
The overall project objectives are:
• Integration of active III-V nanostructures locally and aligned to silicon features
• Development of nanoscale fabrication technology commensurate with VLSI processes to achieve: alignment, parallelization and control of layer thicknesses and roughness on the sub-nm scale
• Explore new low-loss VLSI compatible tunable plasmonic materials for the near-IR wavelength region
• Development of new active photonic-plasmonic device concepts to allow for sub-wavelength confinement, integrated on silicon.
The conclusions of the action are:
• We demonstrated monolithic III-V lasers (GaAs, InP and InGaAs-based) integrated on Si and emitting in the NIR up to 1550 nm
• We demonstrated a novel concept for hybrid III-V/Si photonic crystal emitters based on a self-aligned local integration of III-V within a Si mirror.
• We established the use of metal-clad cavities for micro-lasers as a mean to achieve improved dimensional scaling and temperature stability.
• We showed that plasmonic Au nanoantennae could be used as an efficient means for improving wavelength and temperature stability of emission in InP microdisk lasers
• Lastly, we demonstrated the first in-plane integration of high-speed monolithic InGaAs detectors directly coupled to Si waveguides and operating above 50GHz
In order to combine electronics and photonics a game-changing technology is required, which possesses the same capacity and speed as photonics, but which is scalable towards the realm of electronics.
Another main limitation is found at a materials level. Si is an indirect bandgap material; hence, it is inefficient for light emission. Integration of direct-gap III-V materials used for light emission and detection in the near-IR entails substantial technological challenges; due to the large lattice- and thermal-mismatch high-quality III-Vs cannot simply be grown on Si. and thick buffer layers must be used to reduce the amount of defects.
In addition, hybrid photonic-plasmonic cavities provides a path for scaling dimensions beyond the diffraction limit. PLASMIC addresses the monolithic integration of active III-V emitters and detectors on silicon and seamlessly integrated with silicon passives, and the use of metals to improve scalability and wavelength stability.
The overall project objectives are:
• Integration of active III-V nanostructures locally and aligned to silicon features
• Development of nanoscale fabrication technology commensurate with VLSI processes to achieve: alignment, parallelization and control of layer thicknesses and roughness on the sub-nm scale
• Explore new low-loss VLSI compatible tunable plasmonic materials for the near-IR wavelength region
• Development of new active photonic-plasmonic device concepts to allow for sub-wavelength confinement, integrated on silicon.
The conclusions of the action are:
• We demonstrated monolithic III-V lasers (GaAs, InP and InGaAs-based) integrated on Si and emitting in the NIR up to 1550 nm
• We demonstrated a novel concept for hybrid III-V/Si photonic crystal emitters based on a self-aligned local integration of III-V within a Si mirror.
• We established the use of metal-clad cavities for micro-lasers as a mean to achieve improved dimensional scaling and temperature stability.
• We showed that plasmonic Au nanoantennae could be used as an efficient means for improving wavelength and temperature stability of emission in InP microdisk lasers
• Lastly, we demonstrated the first in-plane integration of high-speed monolithic InGaAs detectors directly coupled to Si waveguides and operating above 50GHz
We have established the basis for Template-Assisted-Selective-Epitaxy for photonic devices. TWe started with the binary material GaAs as this is the easiest from a growth perspective. Using this we achieved room temperature lasing in monolithically integrated micro-discs using two different integration approaches. Direct cavity growth (S. Wirths et al. ACS Nano, 12(3), 2018) and via a two-step virtual substrate approach (B. Mayer et al. IEEE Photonic Tech. Lett 2019). We moved on to InP and compared the performance of monolithic InP microdisk lasers to identical devices fabricated by state-of-the-art direct wafer bonding and etching, and found the performance to be comparable which is evidence of a high crystal quality (S. Mauthe et al. IEEE J. Select. Top. Quantum Electron. 25 (6) 2019).
We then moved to ternary compound, and demonstrated InGaAs micro-cavity lasers using the virtual substrate approach (S. Mauthe et al. , SPIE Europe, 2018). However, the two-step growth mode means that it is very hard to control the composition for ternary compounds. Hence those InGaAs devices only lase up to about 150K. In a more recent work we grew InGaAs micro-disk lasers directly in a one-step growth and achieved room temperature lasing (P. Tiwari et al. CLEO Europe 2021).
We developed a new type of hybrid III-V/Si 1D beam Photonic crystal resonator where individual Si rods are selectively replaced by III-V gain material to position the III-V material only at the center of the resonant cavity (S. Mauthe et al. Nano Letters 20 (12), 2020). A similar approach was used to demonstrate scaled monolithic InGaAs detectors on Si (S. Mauthe et al. Nature Communications 11 (1), 2020), and in a recent work this was extended to larger structures containing and InGaAs/InP heterostructure for improved carrier confinement and waveguide coupling to Si waveguides (P. Wen et al. accepted for publication - available as preprint: arXiv:2106.00620). In this work we also demonstrated LED operation even if we did not yet achieve electrically actuated lasing.
We also worked on the use of metals for the hybrid photonic-plasmonic cavities. We evaluated several in-house deposition methods for both Ag, Au as well as metal nitrides. We investigated the scaling potential of InP microdisk lasers of different geometries with and without a Au metal cladding and found that whereas the use of a metal cladding increases the threshold the devices can be scaled to smaller dimensions (P. Tiwari, Optics Express 29 (3), 2021). We carried out a thorough analysis on the thermal properties and found this to be dramatically impacted by the presence of the metal cavities (P. Wen et al. manuscript in preparation). We further investigated the use of Au nanoantennae coupled to InP micro-disk lasers and found them to have a strong impact on the emission properties in terms of side-mode suppression and temperature stability (pre-print available on: arXiv:2110.11204 article under review).
We then moved to ternary compound, and demonstrated InGaAs micro-cavity lasers using the virtual substrate approach (S. Mauthe et al. , SPIE Europe, 2018). However, the two-step growth mode means that it is very hard to control the composition for ternary compounds. Hence those InGaAs devices only lase up to about 150K. In a more recent work we grew InGaAs micro-disk lasers directly in a one-step growth and achieved room temperature lasing (P. Tiwari et al. CLEO Europe 2021).
We developed a new type of hybrid III-V/Si 1D beam Photonic crystal resonator where individual Si rods are selectively replaced by III-V gain material to position the III-V material only at the center of the resonant cavity (S. Mauthe et al. Nano Letters 20 (12), 2020). A similar approach was used to demonstrate scaled monolithic InGaAs detectors on Si (S. Mauthe et al. Nature Communications 11 (1), 2020), and in a recent work this was extended to larger structures containing and InGaAs/InP heterostructure for improved carrier confinement and waveguide coupling to Si waveguides (P. Wen et al. accepted for publication - available as preprint: arXiv:2106.00620). In this work we also demonstrated LED operation even if we did not yet achieve electrically actuated lasing.
We also worked on the use of metals for the hybrid photonic-plasmonic cavities. We evaluated several in-house deposition methods for both Ag, Au as well as metal nitrides. We investigated the scaling potential of InP microdisk lasers of different geometries with and without a Au metal cladding and found that whereas the use of a metal cladding increases the threshold the devices can be scaled to smaller dimensions (P. Tiwari, Optics Express 29 (3), 2021). We carried out a thorough analysis on the thermal properties and found this to be dramatically impacted by the presence of the metal cavities (P. Wen et al. manuscript in preparation). We further investigated the use of Au nanoantennae coupled to InP micro-disk lasers and found them to have a strong impact on the emission properties in terms of side-mode suppression and temperature stability (pre-print available on: arXiv:2110.11204 article under review).
The optically pumped photonic micro-cavity lasers are now well established, although we continue working on optimizing the cavities, improving growth control and extending to other materials. This represents a novel monolithic integration method for III-V active nanophotonics on silicon. The versatility of this approach in terms of materials and geometry allows for some unique device opportunities. We demonstrated the capability to grow radial quantum wells in the micro-disk lasers and will continue to explore this opportunity beyond the duration of PLASMIC in novel device designs.
The waveguide-coupled monolithic InGaAs detectors demonstrated represents clear progress beyond the sate-of-the-art, we will continue to develop this technique and the long-term goal is to demonstrate a fully monolithic III-V based optical link on silicon.
For the hybrid III-V/Si photonic crystal structures this represents a clear novelty, the precise self-alignment combined with the truly local integration is not achievable with other technique. We are continuing building on this foundation exploring topological designs, and also believe this to be the most promising platform for electrical actuation.
Whereas metal-cladding of cavities have been established for hybrid photonic-plasmonic lasers in the past, the novelty in this project lies in establishing this fro a wide range of geometries and based on conventional processes, so that our findings are portable to other platforms, also the coupling to extensive investigation of the thermal properties is novel. The Antenna-coupled micro-disk also explore a large range of devices compared to earlier studies and also demonstrate coupling to an active micro-disk rather than Si or SiN based resonators
The waveguide-coupled monolithic InGaAs detectors demonstrated represents clear progress beyond the sate-of-the-art, we will continue to develop this technique and the long-term goal is to demonstrate a fully monolithic III-V based optical link on silicon.
For the hybrid III-V/Si photonic crystal structures this represents a clear novelty, the precise self-alignment combined with the truly local integration is not achievable with other technique. We are continuing building on this foundation exploring topological designs, and also believe this to be the most promising platform for electrical actuation.
Whereas metal-cladding of cavities have been established for hybrid photonic-plasmonic lasers in the past, the novelty in this project lies in establishing this fro a wide range of geometries and based on conventional processes, so that our findings are portable to other platforms, also the coupling to extensive investigation of the thermal properties is novel. The Antenna-coupled micro-disk also explore a large range of devices compared to earlier studies and also demonstrate coupling to an active micro-disk rather than Si or SiN based resonators