Alloy Junction 1952 Inspect the Alloyed Assembly

# Inspect the Alloyed Assembly: Reading a Closed Structure From the Outside

Every inspection in this series since Step 4 has had something to look at directly — a bare crystal, a bare slice, a bare disk before it ever touched heat. Step 29 is different in a way none of Phase 3's steps were: the junctions Step 27 located, the doping profile Step 26 established, and the regrowth interface Step 25 brought to rest are all now sealed inside a single, continuous alloyed structure, with no cut face anywhere to look at them directly. Unlike the grown-junction and point-contact sister processes, where a bar is sliced straight through the junction-bearing region for direct inspection, this assembly stays whole. Step 29's real content is therefore not "what does the junction look like" — it is what can still be learned, non-destructively, from the one thing that *is* visible: the outside of a now-solid button sitting on an otherwise ordinary germanium wafer.

## 1. The Frozen Fillet Remembers How It Wetted

Step 21 showed that whether the liquid indium actually wetted the germanium face, rather than beading up on it, was governed by the equilibrium contact angle, $\theta$, set by the three relevant interfacial energies. That angle did not disappear when Step 25's cooling froze the residual button in place — it is preserved, geometrically, as the fillet angle where the solidified button's edge meets the wafer surface, and it is one of the only internal-process facts this assembly still exposes externally:

$$ \cos\theta \;=\; \frac{\gamma_{SV} - \gamma_{SL}}{\gamma_{LV}} $$

A fillet angle close to the value Step 21 predicted is a strong, purely external indicator that wetting proceeded normally and the furnace atmosphere Step 19 established held up for the full cycle. A fillet that instead sits high and beaded, closer to a non-wetting geometry, is visible evidence — without cutting anything open — that something upstream (a contaminated surface Step 9 should have prevented, or an atmosphere lapse Step 19 was meant to guard against) likely compromised the wetted contact area the junctions depend on, long before any electrical test in Phase 4's later steps could confirm it directly.

## 2. Real Diagram: What the Fillet Angle Is Telling You

Reading the Frozen Fillet, Not the Hidden Junction Two outcomes from the same disk, visible only at the surface Good wetting, low fillet angle shallow, spread edge matches Step 21's predicted theta Poor wetting, beaded angle steep, nearly round bead suggests a wetting failure upstream

## 3. Button Geometry as an Indirect Proxy for a Depth You Cannot See

The second question Step 29 can answer without cutting the assembly open is how much germanium actually dissolved at each face — the real values $x_E$ and $x_C$ that Steps 20 through 24 tracked in theory, but which no visual inspection of a sealed structure can measure directly. A workable external proxy exists in simple mass conservation: whatever germanium dissolved into the liquid had to come from somewhere, and its volume shows up as a *reduction* in the combined disk-plus-dissolved-region volume relative to the original pellet, offset by the germanium that joined the melt. Comparing the measured residual button's volume, $V_{\text{button}}$, against the original pellet volume, $V_{\text{pellet}}$, Step 13 recorded before firing gives an estimate of how much germanium left the solid and entered the liquid:

$$ V_{\text{Ge, dissolved}} \;\approx\; \left(V_{\text{button}} - V_{\text{pellet}}\right)\frac{\rho_{\text{In}}}{\rho_{\text{Ge}}} $$

This is only an estimate — it folds together real dissolution with whatever shrinkage Step 28's freezing itself introduced — but it gives inspectors a non-destructive way to flag an assembly whose dissolution ran unusually shallow or unusually deep relative to Step 11's thermodynamic ceiling, before committing it to any of Phase 4's later, more invasive electrical tests. An assembly whose button volume implies a dissolved depth well outside the expected range is exactly the kind of unit this step exists to catch early, rather than discovering the same problem only after lead attachment or final electrical testing has already consumed more process time on a device that was already compromised.

## Real Diagram: Estimated Depth Against the Known Ceiling

Button-Derived Depth Estimate vs. Step 6's Ceiling Each point is one inspected assembly's non-destructive estimate assembly, sampled in sequence estimated dissolved depth ceiling, Step 6/11 flagged: exceeds ceiling

## Inspect the Alloyed Assembly's Place in the Process Lineage

Inspect the Alloyed Assembly follows Step 28, Solidify the Residual Alloy Buttons, which gave this step the one external surface it has to work with; it precedes the post-alloy etching steps that open Phase 4's second category, which will remove surface material this inspection step can only observe, not correct. It is the first step of Phase 4 and the point where this series' working method changes for good: every step from here through Step 36 must infer internal reality from external evidence, because — unlike the sister grown-junction and point-contact processes, which sacrifice a cut bar to direct inspection — this assembly remains whole, and the fillet angle and button geometry examined here are the only two windows this step has into everything Phase 3 actually accomplished inside it.

Take alloy junction 1952 inspect the alloyed assembly 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