Epitaxial Process 1961 Measure Collector Resistance Baseline

# Measure Collector Series Resistance Against the 1959 Planar Baseline: The First Number This Series Exists to Improve

## 1. Why This Series Finally Has to Put a Number on Its Own Promise

This step measures the finished transistor's collector series resistance directly — the voltage drop across the collector at a fixed current, divided by that current — and sets it beside the equivalent measurement on a 1959-style single-substrate planar transistor built to the identical breakdown-voltage specification, because collector resistance was never a headline metric the 1959 series tracked. That series' whole attention was fixed on the exposed junction; it never needed to defend a resistance number, since its single-doping-level substrate offered no alternative to compare against. This series exists specifically because a single substrate could not deliver low collector resistance and high breakdown voltage at once, and every step since Step 1 has been building toward a structure that can — but a structure that merely ought to work is not the same thing as a structure proven to work, and this is the step where that proof becomes a measured number rather than a design intention.

$$R_{\text{collector}} = \frac{\rho_{\text{epi}}\, t_{\text{epi}}}{A} + \frac{\rho_{\text{sub}}\, t_{\text{sub}}}{A}$$

where $\rho_{\text{epi}}$ and $\rho_{\text{sub}}$ are the resistivities of the epitaxial layer and substrate, $t_{\text{epi}}$ and $t_{\text{sub}}$ their thicknesses, and $A$ the device's cross-sectional area — because the substrate is heavily doped, $\rho_{\text{sub}}$ is small, and because the substrate can be made thick relative to the thin epitaxial layer at no cost to the structure, the second term stays small in absolute ohms even though $\rho_{\text{sub}}$ is far below $\rho_{\text{epi}}$, which is this series' entire structural argument, written as an equation for the first time.

A Thick Layer Contributing Almost Nothing collector resistance, 1959 single substrate versus this series' two-layer structure TOTAL COLLECTOR RESISTANCE, SUBDIVIDED BY SOURCE Rcollector 1959, single substrate R ≈ high, no low-ρ option without hurting breakdown substrate term, small epitaxial term, small but present this series, two-layer a thick substrate, almost no resistance contributed Rcollector = ρepitepi/A + ρsubtsub/A most of the physical thickness contributes almost none of the resistance

## 2. Real Diagram: Current Crosses a Thin High-Resistivity Layer, Then a Thick Low-Resistivity One

The cross-section below traces the collector current's actual path — down through the base, across the thin epitaxial layer, and into the thick, heavily-doped substrate to the backside contact — with each segment's resistance contribution annotated directly on the path it belongs to.

The Current's Actual Path Through Two Resistivities a thin high-ρ segment, then a thick low-ρ segment, to the backside contact emitter / base, current enters here ρepitepi/A — thin, higher ρ, small term ρsubtsub/A — thick, far lower ρ, still a small term backside contact Rcollector sums both segments along this single current path

## 3. The Same Comparative Discipline 1959 Used Against the 1958 Mesa Baseline

The 1959 series' own Step 8 proved its reliability improvement not by argument but by measurement — building a mesa transistor and a planar transistor to the same specification, running the identical test on both under the identical conditions, and reading the resulting ratio as the quantified payoff for the structural change that series introduced. This step performs that exact comparative discipline again, on a different metric entirely: a 1959-style single-substrate planar transistor and this series' two-layer epitaxial transistor, built to the same breakdown-voltage specification, measured for collector resistance under identical conditions. Where 1959's comparison validated a change aimed at reliability, this comparison validates a change aimed at resistance — the number Step 1 identified as the price a single substrate was paying for its breakdown voltage, and the number every step since has been working, structurally, to bring down. If the epitaxial layer Step 2 grew came out thicker than specified, or if autodoping left its resistivity higher than Step 5's equation predicts, this measurement is exactly where that earlier drift becomes visible as a smaller-than-promised improvement.

Step 6 does not merely confirm the two-layer structure was built as designed; it puts the first measured number on the benefit this entire series exists to deliver, using the same build-both-and-compare method 1959 already trusted for its own headline claim.

Take epitaxial process 1961 measure collector resistance baseline 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