Alloy Junction 1952 Perform Post Alloy Etching

# Perform Post-Alloy Etching: Breaking a Parasitic Bridge, Not Cleaning a Surface

It would be easy to mistake this step for Step 8's cleanup etch happening a second time, later in the process — remove some residue, leave a tidy surface, move on. That is not what post-alloy etching actually does, and the real history of this exact process is specific about why it exists: this etch was RCA's way of restoring high reverse junction resistance, not a cosmetic step. The reason the junctions Step 27 located can show degraded reverse characteristics right out of the furnace is that the alloying cycle routinely leaves a thin, electrically continuous film — splashed or crept indium, or an indium-germanium skin — bridging the surface exactly where the P-N junction's depletion region reaches that surface. That bridge sits in parallel with the junction itself, and a parallel conductive path at the single most leakage-sensitive point on the whole device can dominate the measured junction behavior regardless of how clean the bulk junction underneath actually is.

## 1. A Parallel Path Beats a Clean Bulk Junction

Where the junction Step 27 located intersects the outer surface, two conduction paths exist side by side: the real, bulk P-N junction's own reverse resistance, $R_j$, and whatever stray conductive film happens to bridge across that same surface point, $R_{\text{bridge}}$. Because these two paths sit in parallel, the resistance anyone actually measures at the terminals is not $R_j$ alone:

$$ \frac{1}{R_{\text{measured}}} \;=\; \frac{1}{R_{j}} + \frac{1}{R_{\text{bridge}}} $$

If $R_{\text{bridge}}$ is low — and a continuous metallic film at the surface is exactly that, low enough to swamp even a genuinely well-formed junction — then $R_{\text{measured}}$ is dominated by the bridge almost entirely, no matter how good $R_j$ actually is. This is the real reason post-alloy etching was worth doing at all: it does not improve the junction itself in any way. It removes $R_{\text{bridge}}$ from the circuit by physically breaking the film's continuity, which lets $R_{\text{measured}}$ finally reflect $R_j$ — the number every step from 25 through 27 actually worked to establish, and the number this single surface film was otherwise free to hide completely.

## 2. Real Diagram: One Junction, Two Paths to Ground

The Surface Bridge Sits in Parallel With the Junction Before etching (left) and after (right), same bulk junction underneath Before: bridge intact continuous film shorts across the junction's surface edge After: bridge broken gap etched through the film; only R_j remains

## 3. Only the Periphery Needs Clearing, Not the Whole Face

Because the leakage path in Section 1 only matters where it actually bridges the junction's exposed edge, this etch does not need to be a blanket removal across the entire surface the way Step 8's damage-removal etch was — it needs to clear a narrow annular region right at the junction's periphery, where $R_{\text{bridge}}$ can actually form a parallel path to $R_j$. That distinction matters because it reopens, in a much more favorable form, the same base-width budget concern that would otherwise make any post-alloy surface etch risky: removing material only at the narrow ring where the junction meets the surface, rather than uniformly across both whole faces, spends only a small, localized fraction of the margin Step 27 established, following the same reaction-rate-limited removal law as any other immersion etch,

$$ \delta(t) \;=\; k\,t, $$

but applied over a far smaller area and, in practice, a much shorter exposure than a full-face cleanup would require — exactly long enough to interrupt $R_{\text{bridge}}$'s continuity at the periphery without meaningfully eating into $W_B$ anywhere else on the device.

## Real Diagram: How Little It Takes to Hide a Good Junction

Measured Resistance Collapses Long Before the Bridge Is Gone R_j held fixed; R_bridge swept from high (etched) to low (intact film) R_bridge, decreasing right to left R_measured R_j, true bulk value before etching after etching

## Perform Post-Alloy Etching's Place in the Process Lineage

Perform Post-Alloy Etching follows Step 29, Inspect the Alloyed Assembly, which is where a visibly poor fillet or an out-of-range button-volume estimate first flagged which units are most likely to carry a bridging film worth specifically targeting here; it precedes Step 31, Protect the Base Attachment During Etching, which addresses a risk this step's own chemistry creates — contaminating or attacking the base tab Step 18 and Step 24 already secured. It is the third step of Phase 4 and the point where this series' running base-width-budget concern (Step 6's ceiling, Step 27's settled junction depths) resurfaces one more time, now in a far more forgiving form: a narrow, targeted etch at the junction periphery rather than a full-face removal, because restoring the junction's true reverse resistance only ever required breaking one specific parallel path, not cleaning everything in sight.

Take alloy junction 1952 perform post alloy etching 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