Epitaxial Process 1961 Confirm Breakdown Voltage Survives

# Confirm Breakdown Voltage Survives Despite the Thin Lightly-Doped Layer: The Other Half of Step 1's Promise

## 1. Why the Margin Step 4 Protected Now Has to Survive an Entirely Different Threat

This step applies reverse bias to the collector-base junction and measures the voltage at which avalanche breakdown occurs, confirming that the thin epitaxial layer — thinner than a uniformly-doped 1959-style collector would ever need to be, because it no longer has to carry current efficiently the way the substrate now does — still supports a depletion region wide enough to sustain the specified breakdown voltage before the heavily-doped substrate beneath it interferes. Step 6 already proved this series delivers on the resistance half of Step 1's original conflict; this step is where the breakdown-voltage half gets its own proof, under the specific new risk this series' own structure introduces. Step 4 worried about a diffusion physically reaching the substrate during fabrication; this step worries about something that happens later and under bias, every time the device operates — the depletion region itself reaching that same substrate interface, which the margin alone, undisturbed, was never actually being tested against.

$$W_{\text{dep}}(V_R) < t_{\text{epi}} - x_j$$

where $W_{\text{dep}}(V_R)$ is the depletion region's width at the specified maximum reverse bias $V_R$, and the right-hand side is the identical margin Step 4 defined between the junction depth and the substrate interface — Step 4 guarded that margin against the diffusion itself closing it during fabrication; this step guards the same margin against the depletion region closing it during operation, and if it closes either way, the heavily-doped substrate's own much lower breakdown voltage takes over and this device's breakdown voltage collapses to whatever the substrate alone could support.

A Margin Closing Under Bias, Not Under Diffusion the depletion region widens with reverse bias until it fills Step 4's own margin diffused base / emitter, depth xj undepleted margin, tepi − xj Wdep(VR), widens as reverse bias increases → substrate interface — reach-through if depletion reaches here first heavily-doped substrate Wdep(VR) < tepi − xj — the same margin Step 4 defined, now tested by bias instead of diffusion if the depletion region wins the race to the interface, the substrate's low breakdown voltage takes over

## 2. Real Diagram: Two Breakdown Curves, One Sharp, One Premature

The measured curves below plot collector current against reverse voltage for two devices built to the same intended specification: a correctly structured device showing a sharp avalanche knee exactly at the specified breakdown voltage, and a reach-through device whose depletion region reached the substrate first, breaking down early at a voltage closer to what the substrate alone could support.

Sharp Avalanche Versus Premature Reach-Through collector current against reverse bias, both devices built to the same intended specification reverse bias, VR → collector current correctly structured — sharp knee at specified VR reach-through device — premature knee, substrate-limited specified VR premature VR the gap between the two knees is the margin reach-through actually cost this device

## 3. A Failure Mode 1959's Own Structure Never Needed a Name For

The 1959 planar series' single-substrate collector had no internal interface anywhere beneath its diffused junction — a depletion region expanding under reverse bias simply kept expanding into more of the same uniformly-doped material, and breakdown occurred entirely on that material's own terms, with no second region of different doping waiting to interfere. This series deliberately introduced that interface in Step 2, for the reason Step 6 just quantified: the heavily-doped substrate beneath a thin, lightly-doped epitaxial layer is what makes this series' resistance improvement possible at all. But the same interface responsible for that improvement is also, for the first time in this project's history, a surface a depletion region can reach — and when it does, the device's breakdown voltage stops being set by the carefully controlled lightly-doped layer and starts being set by the substrate instead, a specific, nameable condition this project calls reach-through, because no earlier series in this project's history ever needed a name for it.

Step 7 does not merely re-confirm the breakdown voltage Step 1 demanded; it confirms that the very structure delivering Step 6's resistance improvement has not quietly taken that breakdown voltage back.

Take epitaxial process 1961 confirm breakdown voltage survives further

Ask the copilot about this term, or have our engineers assess it against your process.

Ask CFSGPT about this → Request a technical assessment → Open the CFS app