Separate-confinement heterostructure (SCH, GRINSCH)
A laser layer stack in which thin quantum wells confine the carriers and a thicker waveguide layer of intermediate bandgap, typically 0.1–0.4 µm, confines the light. A single 8 nm well in an optimized SCH captures about 1.6% of the optical mode.
A separate-confinement heterostructure (SCH) is the vertical layer design used in nearly every quantum-well diode laser. It assigns the two confinement jobs to different layers: one or more quantum wells, each 5–10 nm thick, hold the injected electrons and holes, while a surrounding waveguide core of intermediate bandgap, typically 0.1–0.4 µm thick, holds the optical mode between higher-gap, lower-index cladding layers about 1–2 µm thick. In a GRINSCH (graded-index SCH) the composition of the waveguide core varies continuously from the cladding toward the well, so that both the bandgap and the refractive index are graded.
Why the two confinements are separated
In a bulk double heterostructure, described under carrier confinement, a single active layer about 0.1–0.2 µm thick does both jobs, because the lower-gap material is also the higher-index material. A quantum well cannot: a layer 8 nm thick with an index step of 0.15 confines only about 0.05% of a mode on its own, as the confinement factor entry shows. The SCH solves this by building a conventional slab waveguide whose core is thick enough to guide, and placing the wells at its center where the field is strongest. The wells contribute gain in proportion to the slice of the mode passing through them, while the transparency current scales with their small volume.
Confinement factor of a single well
For a symmetric slab, the confinement factor of a thin well at the center is approximately the well thickness times the normalized intensity at the peak of the mode. Taking indices of 3.17 for InP cladding and 3.40 for a quaternary core at 1550 nm, an exact TE slab solution gives, for an 8 nm well at the center:
| Core thickness | Core Γ | Well Γ |
|---|---|---|
| 0.10 µm | 0.11 | 0.009 |
| 0.20 µm | 0.33 | 0.014 |
| 0.35 µm | 0.61 | 0.016 |
| 1.0 µm | 0.94 | 0.011 |
The well confinement has a broad maximum, about 0.016 here, near a core thickness of 0.3–0.4 µm. A thinner core lets the mode spread into the cladding, and a thicker one spreads the mode over the core; both lower the peak intensity at the well. Five wells near the optimum capture roughly 0.08 in total, consistent with the range quoted for multi-quantum-well active regions. The numbers depend on the assumed indices, but AlGaAs/GaAs structures follow the same pattern.
Graded-index SCH
In a GRINSCH, the aluminum or phosphorus fraction of the core is graded, often linearly or parabolically, from the cladding composition down toward the well barriers. Two effects follow. The index profile narrows the mode slightly, raising the intensity at the well. The graded band edges produce a built-in field that drives injected carriers toward the well, shortening their transport through the undoped core and improving capture. W. T. Tsang demonstrated the structure in AlGaAs by molecular beam epitaxy in 1981–1982 and obtained very low threshold current densities, and graded or stepped cores remain common in GaAs-based lasers.
Where it matters in practice
Carrier transport across the SCH layers adds a delay to the laser's modulation response. The diffusion time across a layer of thickness is of order
so a 0.1 µm layer with an ambipolar diffusion coefficient of 5 cm²/s gives about 10 ps, the example worked under transit time. Carriers captured by the wells can also escape back into the core, and the population left in the SCH contributes no gain; this capture and escape balance enters high-speed laser models as an effective reduction of the differential gain.
At high temperature, carriers spill over the core barriers into the cladding, a leakage path that contributes to thermal rollover. The core composition is therefore a trade-off: a higher core bandgap gives a deeper well and less escape, but a smaller index step and a weaker waveguide.
The SCH thickness also sets the near field in the growth direction and thus the fast-axis far-field divergence. High-power lasers often use a deliberately thick, low-contrast core, called a large optical cavity, which lowers the peak intensity at the facet and narrows the fast-axis beam at the cost of a smaller well confinement factor.
Common questions
What is the difference between SCH and a double heterostructure?
A double heterostructure uses one active layer to confine both carriers and light. An SCH uses quantum wells for the carriers and a separate, thicker core layer for the light, so each can be optimized independently.
Why does the confinement factor of a quantum-well laser look so small?
Because the well occupies only a few nanometers of a mode several hundred nanometers wide. A well Γ of 0.01–0.03 is normal; it is compensated by the high material gain of the well and, where needed, by stacking several wells.
Is the SCH layer doped?
The core and wells are usually left undoped, forming the intrinsic region of the p-i-n diode, while the cladding layers carry the p and n doping. This keeps free-carrier absorption low where the intensity is highest.
References: W. T. Tsang, "Extremely low threshold (AlGa)As graded-index waveguide separate-confinement heterostructure lasers grown by molecular beam epitaxy," Applied Physics Letters 40, 217 (1982); L. A. Coldren, S. W. Corzine and M. L. Mašanović, Diode Lasers and Photonic Integrated Circuits, 2nd ed. (Wiley, 2012); S. L. Chuang, Physics of Photonic Devices, 2nd ed. (Wiley, 2009).