Amorphous Silicon Based Crosstalk Resilient Photonic Phase Shifters

In photonic integrated circuits (PICs), photonic phase shifters (PPS) [1] have numerous applications in sensing, switching, and advanced communication in simple components like ring resonators or interferometers and also in complex systems such as phase arrays. For a tunable phase shift, the PPS can use a heater. However, in densely packed PICs, the thermal crosstalk between different parts of the chip may lead to severe performance degradation. Previously, we have shown that in thermally tuned crystalline silicon (c-Si) rib waveguides, N++ doped silicon heaters provide a better thermal crosstalk performance than integrated metal heaters [2]. But, CMOS-compatible PICs can be fabricated on a hydrogenated amorphous silicon platform [3] by chemical vapor deposition as well. Amorphous silicon (a-Si) has a thermal conductivity of 1.8 W/(m.K) at room temperature [4], which is much lower than that for c-Si with 90 W/(m.K). Therefore, amorphous silicon might be a better platform for densely packed PICs. Here, we present a comparative study of the effect of thermal crosstalk on amorphous silicon rib waveguides and its mitigation by a deep trench [2] using the simulation software Lumerical [5]. A cross-sectional view of the simulation design is shown in Fig. 1(a). The rib waveguides (wg-1 and wg-2) and the slab are made of amorphous silicon, and an N++ doped Si heater with a doping concentration of 1 x 10
²⁰
cm
⁻³
is placed along these waveguides. The buried oxide (BOX) and the cladding layer are composed of silicon dioxide (SiO
2
). The proposed design is CMOS-compatible on a standard silicon-on-insulator platform, and the dimensions of the individual components and the air-filled deep trench are the same as in Ref. [2]. The respective temperature profile is shown in Fig. 1(b). Figure 1(c) shows the same design but with an additional deep trench [2] to mitigate the crosstalk. The thermal insulation and heat confinement by the deep trench can be seen in Fig. 1(d). The phase crosstalk, defined as the ratio of the phase shift in the secondary waveguide (wg-2) to that in the primary waveguide (wg-1) for different spacing between the waveguides
$d$
and for different slab widths w, can be seen in Figs. 1(e) and (f), respectively. As shown, the a-Si has a noticeably better phase crosstalk performance than c-Si, and for the deep trench, the phase crosstalk is minimal for both a-Si and c-Si, as presented in Fig. 1(g).