Thermal Path Design for Pulsed LiDAR Emitters: Junction to Heat Sink

The thermal stack in a pulsed emitter module

For pulsed lidar emitters, the submount is the critical spreading layer: single-crystal SiC at 350–400 W/m·K is the default because heat must cross the ceramic faster than the pulse width; polycrystalline ALN at 170–210 W/m·K is usually insufficient above ~100 W/cm² peak power density. Heat generated at the laser junction must cross a series of interfaces before reaching the module heat sink or cold plate. In a typical TO56 or cavity-package pulsed lidar emitter:

1. Laser junction — heat source during the current pulse.
2. Die attach layer — AuSn, solder, or epoxy; bond line thickness and void fraction dominate contact resistance.
3. Submount (ceramic spreader) — spreads heat laterally before it reaches the header; material conductivity and thickness set spreading resistance.
4. Submount-to-header attach — solder or epoxy to Kovar post or package floor.
5. Header base / flange — metal thermal mass and path to exterior case.
6. TIM (thermal interface material) — grease, pad, or solder to module heat sink.
7. Heat sink — aluminum or copper spreader, often finned for automotive modules.

For pulsed operation, the critical metric is not only steady-state R_th (case-to-ambient) but how fast the junction temperature rises during the pulse. The submount must spread heat faster than the pulse duration to avoid overshoot that drives facet damage and wavelength chirp.

Instantaneous power during a lidar pulse

A gain-switched 905 nm emitter at 30 A peak and 2 V forward bias dissipates roughly 60 W in the junction for a 5 ns pulse — even when average power is only hundreds of milliwatts at sub-1% duty cycle. The thermal mass of the active region is tiny; temperature spikes locally before the header or heat sink respond.

Why single-crystal SiC is the default submount

Polycrystalline ALN at 170–210 W/m·K is sufficient for many CW telecom lasers. Pulsed lidar pushes transient thermal spreading: phonons must cross the submount volume before the next pulse in a burst train. Single-crystal SiC at 350–400 W/m·K — without grain-boundary phonon scattering — provides roughly twice the spreading capability of ALN for the same geometry and thickness.

SiC is also semi-insulating, supporting both p-side-down and n-side-down die attach in edge emitters. For automotive lidar qualification, pairing SiC submounts with hermetic TO or cavity packages is standard practice; see FMCW vs. pulsed lidar packaging for how pulsed and coherent systems diverge.

Product specifications and standard part numbers: FerraLink SiC submounts. If SiC spreading is still insufficient after geometry optimization, see CVD diamond submounts (pre-release).

Thermal resistance budget (conceptual)

Engineers often target case-to-die thermal resistance below 5°C/W for uncooled pulsed modules. A simplified budget:

• R_junction-spreader (die attach): 1–3°C/W — minimize voids, optimize AuSn thickness.
• R_spreader (submount conduction): scales as t / (k·A) — lower thickness and higher k reduce this term; SiC wins on k.
• R_spreader-header: 0.5–1.5°C/W — solder voiding is a common failure mode.
• R_header-case + TIM + sink: remainder of stack to ambient.

When simulation shows junction overshoot above safe T_j during burst firing, the first packaging lever is usually submount material (ALN → SiC) or submount thickness reduction — before redesigning the heat sink.

Assembly choices that affect the thermal path

Related articles

The part that depends on your die

The rules above hold for most edge-emitter modules. What changes from program to program is geometry, duty cycle, and how hard you are pushing junction temperature — those inputs decide material, thickness, and whether catalog samples are enough.

• ΔT and junction temperature for your die size, power, and duty cycle.
• Whether your thermal path needs SiC spreading or AlN is sufficient.
• Header and TIM choices once submount material is fixed.