Point Contact Transistor Grind Etch Second Contact Area
# Grind and Etch the Second Contact Area: The Face That Could Not Afford to Wait
The contact face was ground and etched back in Step 3 and Step 4, then left alone, because nothing happens to it again for many steps and it loses nothing by sitting finished and verified in the meantime. The base face cannot be treated the same way. It is about to receive a metal deposit, and a freshly exposed germanium surface is far more receptive to that deposit than one that has sat exposed to ordinary ambient conditions for any length of time. Step 5 grinds and etches this second contact area now, immediately before metallization, specifically so the gap between finishing this face and depositing metal on it is as short as the process can make it.
A freshly etched surface and a long-exposed surface are not the same starting point for a metal deposit, even if both measure as clean. Step 4's rinse and dry left the contact face free of residual etchant and contamination at the moment it was checked, but "clean at the moment of checking" is not the same guarantee as "clean at the moment metal arrives," and the contact face is allowed that gap only because nothing touches it again until much later. The base face does not have that luxury: any delay between finishing its etch and depositing metal on it gives ambient moisture, dust, or a renewed oxide layer time to form, and that layer sits directly at the metal-to-germanium interface the base electrode depends on for a dependable, low-resistance contact. Grinding and etching this face now, right before metallization, is how the process keeps that interval as short as it can be made.
This face's grinding and etching risks are the same ones Steps 3 and 4 already identified, carried over to a different standard and a tighter schedule. Grinding still drives a subsurface damage layer whose depth scales with grit size, and the following etch still has to clear that specific depth without overshooting it, exactly as on the contact face. What changes is the target the roughness has to meet — far looser here, because a broad metal electrode averages over imperfection the way a point contact cannot — and the consequence of taking too long to get there, since every added minute between finishing this face and metallizing it works against the freshness this step exists to protect. A process that grinds and etches this face to the contact face's tight roughness standard would be solving the wrong problem; the problem here is time, not precision.
| Step | Process operation | Input | Output | Specification | Constraint |
|---|---|---|---|---|---|
| 5.1 | Set the finished flatness standard for the base face from the planned metal contact method | Base contact method planned for metallization | Base face grind specification | Flatness adequate for a broad, low-resistance metal contact over the whole face | Over-specifying this face wastes grinding time the base contact's averaging effect never collects on |
| 5.2 | Choose grit progression and final pass sized to the etch allowance that will follow, scheduled immediately before metallization | Base face grind specification from 5.1, planned metallization timing | Grinding plan, time-constrained | Final grit sized so its expected subsurface damage depth is within the following etch's allowance, with grinding scheduled to begin only when metallization is imminent | Grinding this face too far ahead of metallization defeats the purpose of timing it this way at all |
| 5.3 | Grind the base face to its specification | As-cut body, grinding plan from 5.2 | Base face ground flat | Measured flatness within the specification from 5.1 | A base face ground off-specification compromises contact uniformity before metallization ever applies metal to it |
| 5.4 | Etch the base face for the time required to clear its recorded damage depth | Ground base face from 5.3 | Etched base face | Etch time sufficient to clear the damage depth from 5.2, without exceeding it | Over-etching this face wastes the same germanium thickness budget Step 2 established, for a target this face already met |
| 5.5 | Rinse, dry, and move immediately to metallization, minimizing the interval between this step and the next | Etched base face from 5.4 | Clean base face, ready for metallization | No residual contamination; elapsed time before metallization minimized and documented | A long, undocumented delay between this step and metallization can introduce exactly the oxide layer this step's timing was meant to prevent |
| 5.6 | Record this face's grind and etch parameters, timing, and verification against the body's provenance | Clean base face from 5.5 | Documented, base-face-ground-and-etched body | Specification, timing, and verification recorded against the record from Step 2.6 | Without this record, a later base-contact resistance anomaly cannot be distinguished from a contact face or metallization issue |
Step 5 does not repeat Steps 3 and 4 on a second face out of mere symmetry; it applies the same two operations under a completely different timing discipline, because this face's risk is exposure, not precision. The contact face could be finished early and trusted to wait, because nothing acts on it in the meantime. This face has to be finished late, on purpose, because the next operation in the sequence is the one thing most likely to undo a surface that sits unprotected for too long. Treating both faces as interchangeable stock-preparation steps would miss the one distinction that actually matters here: it is not how clean this face gets, but how quickly metal reaches it afterward.