Accelink is an optical device company in the Wuyou system. It released the research progress of EML yesterday, which was jointly researched by Taiyuan Engineering and Accelink.
The physical photo of the EML is given, and the length of the DFB region is 200 μm, but the length of the EA electroabsorption modulator is not provided, and this parameter has a very large correlation with the modulation bandwidth.
Actual photos, the proportion is certain. If deduced according to the DFB length of 200 μm, the length of the electroabsorption region is between 75-100 μm.
The EA junction capacitance is 0.09pF, the series resistance is 2.45Ω, and the parasitic capacitance is relatively large, 0.2pF
The simulated bandwidth is 19.8GHz, and the measured -3dB bandwidth is 43GHz, which is quite good.
Butt growth is commonly used in the EML industry, and a small part is selective growth or dual active layers. Wuyou uses the technology of combining selective growth and dual active layers.
Cross-section quantum well, DFB area is a double active layer, EA area is selective growth, and the wavelength detuning between the electroabsorption area and DFB area is 50nm. Compared with the data of other manufacturers seen before, the detuning range is larger, and the detuning is too The conference resulted in insufficient extinction ratio. After Accelink and their follow-up data on this EML are released, you can compare and see ER.
In the case of docking growth, the detuning range will not affect each other, but the double active layer, below the DFB region is the shared absorption region quantum well, but the layer thickness is different, the internal stress is different, and the wavelength detuning range will be slightly different. Make it wider.
If you choose the common detuning range of 10nm for docking growth, absorption will occur below the DFB. Taiyuan Engineering and Opticsun widen the detuning range to 110nm. The upper active layer in the DFB area is 1310nm, and the lower active layer is 1200nm (also That is, the 1310nm band does not absorb and belongs to the transparent region), the purpose is to reduce the intracavity loss of this structure.
Secondly, the optimization of the PNPN semiconductor BH structure is chosen, and the P-type-InP confinement layer is changed to a semi-insulating region doped with Fe ions. The n-Inp barrier design is retained to avoid mutual diffusion of metal ions. This design is different from Huawei, Lumentum and other factories, and is similar to a structure of Mitsubishi.
The optical field of the DFB, is confined to the active layer above
If you look at the picture for calibration, the width of the active layer is 2 μm
If you look at the picture for calibration, the width of the active layer is 2 μm
Inversely deduced from the two figures, the vertical divergence angle of the modulation area decreases, and the coupling efficiency with the fiber will increase.
Compared with the ridge waveguide, the BH structure has a larger confinement factor. In the traditional PNPN semiconductor-type BH heterogeneous buried structure, optimization to a part of the semiconductor and a part of the semi-insulating Fe-InP burial can further increase the confinement factor.
EML devices are usually packaged in TO-Can (metal case) or Butterfly packages. The TO-Can package is a metal case that houses the EML laser and the drive circuit. TO-Can packaging is usually used for low-rate applications (less than 10Gbps), such as 1.25Gbps and 2.5Gbps optical communication systems and LAN applications. Butterfly package is a more advanced package form, which adopts double-layer package structure. EML devices in Butterfly packages are usually composed of EML chips, pads, bonding wires, leads and sapphire window. This package is suitable for high-speed applications (greater than 10Gbps), such as 40Gbps and 100Gbps optical communication systems. In the Butterfly package, EML devices are usually built into modules together with other optical components (such as fiber optic connectors, waveguides, etc.) for higher-level integration and system applications.
Comentarios