No abstract

Monolithic lasers on Si are ideal for high-volume and large-scale electronic-photonic integration. Ge is an interesting candidate owing to its pseudodirect gap properties and compatibility with Si complementary metal oxide semiconductor technology. Recently we have demonstrated room-temperature photoluminescence, electroluminescence, and optical gain from the direct gap transition of band-engineered Ge-on-Si using tensile strain and n-type doping. Here we report what we believe to be the first experimental observation of lasing from the direct gap transition of Ge-on-Si at room temperature using an edge-emitting waveguide device. The emission exhibited a gain spectrum of 1590-1610 nm, line narrowing and polarization evolution from a mixed TE/TM to predominantly TE with increasing gain, and a clear threshold behavior.

Non-reciprocal photonic devices, including optical isolators and circulators, are indispensible components in optical communication systems. However, the integration of such devices on semiconductor platforms has been challenging because of material incompatibilities between semiconductors and magneto-optical materials that necessitate wafer bonding, and because of the large footprint of isolator designs. Here, we report the first monolithically integrated magneto-optical isolator on silicon. Using a non-reciprocal optical resonator on an silicon-on-insulator substrate, we demonstrate unidirectional optical transmission with an isolation ratio up to 19.5 dB near the 1,550 nm telecommunication wavelength in a homogeneous external magnetic field. Our device has a small footprint that is 290 mm in length, significantly smaller than a conventional integrated optical isolator on a single crystal garnet substrate. This monolithically integrated non-reciprocal optical resonator may serve as a fundamental building block in a variety of ultracompact silicon photonic devices including optical isolators and circulators, enabling future low-cost, large-scale integration.Non-reciprocal photonic devices that break the time-reversal symmetry of light propagation provide critical functionalities such as optical isolation and circulation in photonic systems. Although widely used in optical communications, such devices are still lacking in semiconductor integrated photonic systems 1,2 because of challenges in both materials integration and device design. On the materials side, magneto-optical garnets used in discrete nonreciprocal photonic devices show large lattice and thermal mismatch with semiconductor substrates, making it difficult to achieve monolithic integration of garnets with phase purity, high Faraday rotation and low transmission loss 3,4 , and requiring wafer bonding to incorporate them on a semiconductor platform. On the device side, non-reciprocal mode conversion (NRMC) and non-reciprocal phase shift (NRPS) integrated optical isolators have large footprints with length scales from millimetres to centimetres 5,6 , which severely limits the feasibility of large-scale and low-cost integration. Efforts have been pursued both in the monolithic integration of iron garnet and the exploration of other magneto-optical materials with better semiconductor compatibility. Polycrystalline Y 3 Fe 5 O 12 (YIG) films 3 , epitaxial Sr(Fe-doped InP films 9 have been demonstrated to have promising magneto-optical performance at a wavelength of 1,550 nm. In relation to device design, several monolithic non-reciprocal photonic devices capitalizing on optical resonance effects (for example, magneto-optical photonic crystals 10 , garnet thin-film based optical resonators 11 , silicon ring resonators with magneto-optical polymer cladding 12 and modulated ring resonators using non-reciprocal photonic transitions 1 ) have been theoretically analysed with a view to reducing the device footprint. However, the experimental realization of monolit...

We analyze the optical gain of tensile-strained, n-type Ge material for Si-compatible laser applications. The band structure of unstrained Ge exhibits indirect conduction band valleys (L) lower than the direct valley (Gamma) by 136 meV. Adequate strain and n-type doping engineering can effectively provide population inversion in the direct bandgap of Ge. The tensile strain decreases the difference between the L valleys and the Gamma valley, while the extrinsic electrons from n-type doping fill the L valleys to the level of the Gamma valley to compensate for the remaining energy difference. Our modeling shows that with a combination of 0.25% tensile strain and an extrinsic electron density of 7.6x10(19)/cm(3) by n-type doping, a net material gain of ~400 cm(-1) can be obtained from the direct gap transition of Ge despite of the free carrier absorption loss. The threshold current density for lasing is estimated to be ~6kA cm(-2) for a typical edgeemitting double heterojunction structure. These results indicate that tensile strained n-type Ge is a good candidate for Si integrated lasers.

No abstract