A new materials group to implement dense wavelength division multiplexing (DWDM) in Si photonics is proposed. A large thermo-optic (TO) coefficient of Si malfunctions multiplexer/demultiplexer (MUX/DEMUX) on a chip under thermal fluctuation, and thus DWDM implementation, has been one of the most challenging targets in Si photonics. The present study specifies an optical materials group for DWDM by a systematic survey of their TO coefficients and refractive indices. The group is classified as mid-index contrast optics (MiDex) materials, and non-stoichiometric silicon nitride (SiNx) is chosen to demonstrate its significant thermal stability. The TO coefficient of non-stoichiometric SiNx is precisely measured in the temperature range 24–76 °C using the SiNx rings prepared by two methods: chemical vapor deposition (CVD) and physical vapor deposition (PVD). The CVD-SiNx ring reveals nearly the same TO coefficient reported for stoichiometric CVD-Si3N4, while the value for the PVD-SiNx ring is slightly higher. Both SiNx rings lock their resonance frequencies within 100 GHz in this temperature range. Since CVD-SiNx needs a high temperature annealing to reduce N–H bond absorption, it is concluded that PVD-SiNx is suited as a MiDex material introduced in the CMOS back-end-of-line. Further stabilization is required, considering the crosstalk between two channels; a ‘silicone’ polymer is employed to compensate for the temperature fluctuation using its negative TO coefficient, called athermalization. This demonstrates that the resonance of these SiNx rings is locked within 50 GHz at the same temperature range in the wavelength range 1460–1620 nm (the so-called S, C, and L bands in optical fiber communication networks). A further survey on the MiDex materials strongly suggests that Al2O3, Ga2O3 Ta2O5, HfO2 and their alloys should provide even more stable platforms for DWDM implementation in MiDex photonics. It is discussed that the MiDex photonics will find various applications such as medical and environmental sensing and in-vehicle data-communication.

No abstract

Hyperspectral (HS) imaging provides rich spatial and spectral information and extends image inspection beyond human perception. Existing approaches, however, suffer from several drawbacks such as low sensitivity, resolution and/or frame rate, which confines HS cameras to scientific laboratories. Here we develop a video-rate HS camera capable of collecting spectral information on real-world scenes with sensitivities and spatial resolutions comparable with those of a typical RGB camera. Our camera uses compressive sensing, whereby spatial–spectral encoding is achieved with an array of 64 complementary metal–oxide–semiconductor (CMOS)-compatible Fabry–Pérot filters placed onto a monochromatic image sensor. The array affords high optical transmission while minimizing the reconstruction error in subsequent iterative image reconstruction. The experimentally measured sensitivity of 45% for visible light, the spatial resolution of 3 px for 3 dB contrast, and the frame rate of 32.3 fps at VGA resolution meet the requirements for practical use. For further acceleration, we show that AI-based image reconstruction affords operation at 34.4 fps and full high-definition resolution. By enabling practical sensitivity, resolution and frame rate together with compact size and data compression, our HS camera holds great promise for the adoption of HS technology in real-world scenarios, including consumer applications such as smartphones and drones.

A method for reduction of threading dislocation density (TDD) in lattice-mismatched heteroepitaxy is proposed, and the reduction is experimentally verified for Ge on Si. Flat-top epitaxial layers are formed through coalescences of non-planar selectively grown epitaxial layers, and enable the TDD reduction in terms of image force. Numerical calculations and experiments for Ge on Si verify the TDD reduction by this method. The method should be applicable to not only Ge on Si but also other lattice-mismatched heteroepitaxy such as III-V on Si.

Ge laser is expected as a solution for a key issue in electronic-photonic integrations, CMOS compatible on-chip light source. Two important issues remain in Ge laser technologies; the large threshold current (Ith) and the multiple growth of Ge epi-structures to co-integrate Ge devices such as photodetectors, whose pn junction profiles are much different from the Ge lasers. The reported Ith values are larger than those for III-V lasers by two orders of magnitude, and an in-situ heavily n-type doping in Ge prevents to form various pn structures unless the multiple growth with different doping profiles are performed. To solve these issues, the present paper proposes a "reverse-rib" Ge epi-structure in combination with an ex-situ P diffusion for heavily n-type doping, instead of in-situ doping. P-diffused n+-Ge reverse-rib structures first verify a clear threshold behavior of light emission at a low power of optical pumping.

The kinetics of Ge lateral overgrowth on SiO2 with line-shaped Si seeds is examined. The growth process is described by the difference between the growth rates of Ge on (100) planes (GR100) and <311> facets (GR311). The theoretical calculations well reproduce the growth kinetics. It is shown that narrowing the line-seeds helps Ge coalescence and flat film formation.

No abstract

No abstract