← SSC (spot size converter)

2% 4 µm × 4 µm SSC @ 1550 nm

2% Δ GeO₂-doped silica core (4×4 µm) ↔ SMF-28 at 1550 nm — a mode-expanding segmented SSC reaching 0.114 dB coupling loss (mode-overlap integral), below the 0.2 dB target.

Target & SMF parameters

SMF-28 MFD is ≈ 10.4 µm at 1550 nm. The bare 4 µm core guides only a ≈ 4.6 µm mode, so the SSC expands it up to ≈ 10–11 µm to match the fiber. The last field is the mode-field diameter the SSC facet presents to the SMF (the design target).

n_core (from Δ)
n_clad (silica, Sellmeier)
Bare WG mode field diameter (D4σ)
Coupling loss — bare WG ↔ SMF
Coupling loss — SSC tip ↔ SMF
Improvement
SSC quality

Analytic Gaussian-overlap estimate for a quick look; the authoritative numbers below are the mode-overlap integral between the true 2-D waveguide/facet mode (from a mode solver) and the SMF-28 field.

Mode field cross-section — blue: SMF Gaussian, red dashed: bare WG mode, green: expanded SSC-tip mode. Larger overlap with the SMF means lower loss.

Why a spot size converter

How the coupling loss is evaluated (mode-overlap integral)

Butt-coupling loss is the mode-overlap integral between the fiber mode and the waveguide (facet) mode — the robust, standard definition:

η = |∫ E_wg·E_smf* dA|² / (∫|E_wg|² dA · ∫|E_smf|² dA),    Loss(dB) = −10·log₁₀(η)

The 2-D waveguide mode E_wg (the bare chip mode, or the duty-averaged effective-medium facet mode) is found with an imaginary-distance mode solver on the real index cross-section; E_smf is the SMF-28 field (MFD 9.2 µm). The segmented taper's role is to expand the chip mode to the facet mode adiabatically, so the device coupling loss equals the facet-mode / SMF mismatch.

Note on method. The low-loss facet mode is weakly guided (near cutoff). A "launch a Gaussian, propagate it, overlap the output" scalar-BPM metric is unreliable in that regime: even propagating the exact expanded eigenmode returns a self-overlap far below 100 % (a paraxial mode-beating artefact), which spuriously inflates the loss. The mode-overlap integral above is free of that artefact, so it is used for the reported numbers.

Result @ 1550 nm — coupling loss < 0.2 dB

Chip waveguide 4 µm × 4 µm solid; the segmented taper ramps the duty 0.95 → 0.20 (cosine) and the segment width 4 → 7 µm over a gentle ≈ 530 µm length, expanding the mode from 4.6 µm to ≈ 11 µm at the SMF facet.

QuantityValue
Coupling loss — SSC facet ↔ SMF-280.114 dB (η = 97.40 %)
Coupling loss — bare 4 µm chip (no SSC)2.46 dB (η = 56.8 %)
Chip mode field diameter (D4σ)4.6 × 4.6 µm  (n_eff = 1.46149)
SSC facet mode field diameter (D4σ)11.2 × 10.0 µm  (n_eff = 1.44547)
Facet effective width / duty7.0 µm / 0.20  (cosine ramp from solid chip)
Facet n_eff margin above cladding+0.0015  (weakly guided, near cutoff)

Expanding the mode to D4σ ≈ 11.2 × 10.0 µm brings it onto the SMF-28 field (10.4 µm), lifting the overlap from 56.8 % (bare) to 97.4 % — a coupling loss of 0.114 dB, below the 0.2 dB target. A facet duty in ≈ 0.18–0.24 keeps the loss under 0.16 dB (and 0.16–0.24 under 0.2 dB), giving good lithography tolerance. Because the low-loss facet mode is weakly guided (n_eff only ≈ 0.0015 above the cladding), the design is more sensitive to duty/width fabrication error and to substrate proximity than a strongly-guided mode; a thick cladding and an adiabatic taper are assumed.

2% 4um 1550 nm SSC: duty/width ramp, chip and facet modes, mode cut, loss vs duty
Mode-expanding SSC @ 1550 nm (2% Δ, 4 µm core) — duty/width ramp, the chip mode (4.6 µm) and expanded facet mode (11.2 × 10.0 µm), a y = 0 mode-field cut against the SMF-28 field, and coupling loss vs facet duty (0.16–0.24 all under 0.2 dB, minimum ≈ 0.11 dB at duty 0.20).
Bare 4um chip mode vs mode-expanding SSC facet matched to SMF-28 at 1550 nm
Mode matching to SMF-28 (white dashed = 10.4 µm MFD): the bare 4 µm chip mode (4.6 µm, 2.46 dB) badly under-fills the fiber, while the expanded SSC facet mode (11.2 × 10.0 µm, 0.114 dB) matches it.

Device layout (GDS)

The SSC built from this result as a ready-to-place mask cell. A solid 4 µm chip bus feeds the mode-expanding segmented taper (120 segments, duty 0.95 → 0.20 cosine, width 4 → 7 µm cosine over 530 µm); the wide low-duty facet at the chip edge is where the SMF-28 butt-couples.

Top-view GDS layout of the 2% 4x4um SSC device: solid input bus, mode-expanding segmented taper, and SMF facet
Top view of ssc_device.gds (core layer 1/0) — solid 4 µm input bus → segmented mode-expanding taper → 7 µm-wide, 0.20-duty facet at the chip edge (red dashed = dicing / SMF butt-coupling line, layer 10/0). Zoom shows the low-duty facet segments.

Layer map — (1,0) waveguide core (2% Δ GeO₂-silica), (10,0) facet / dicing reference, (63,0) text label. The taper geometry matches ssc_optimized.gds below; this cell adds the routing bus, facet marker, and label for direct placement.

Downloads

Everything needed to reproduce the result above:

Run with python3 run_modal_ssc_2pct_4um_1550.py --out . (needs numpy, gdstk, matplotlib, and bpm3d.py). Coupling loss is the mode-overlap integral between the solver's 2-D waveguide mode and the SMF-28 field; pass --dx 0.07 for a finer convergence check.

Equations

SMF mode (Gaussian): E_SMF(r) = exp(−r²/w²),   w = MFD/2
Overlap: η = (∫ E_a·E_b dA)² / (∫E_a² dA · ∫E_b² dA),    Loss(dB) = −10·log₁₀(η)
Index from Δ: n_core = n_clad / √(1 − 2Δ),   n_clad from the Malitson silica Sellmeier model.

Reference implementations