Point Contact Transistor Apply Copper or Gold
# Apply Copper or Gold: The Base Electrode Solves the Same Resistance Problem by Doing the Opposite Thing
Step 5 ground and etched the base face immediately before this step, specifically so a metal film would bond to freshly exposed, chemically clean germanium rather than one that had sat exposed since early in the sequence. Step 6 is where that specification stops being a promise about a future contact and becomes an actual electrode. The point contacts still to come will be made as small and sharp as the process can manage, because their entire function depends on localizing current into a region a few thousandths of an inch wide. This electrode is built on the opposite principle: as large as the whole underside allows, because its only job is to collect the base current those point contacts generate and hand it off to the outside world adding as little resistance as possible along the way.
Contact resistance to the bulk falls as the electrode's effective radius grows, which is exactly why this electrode is made as large as the underside allows. The spreading resistance a current encounters converging toward or diverging from a contact on a semiconductor scales inversely with that contact's radius — a relationship that cuts both ways depending on which end of it a given contact sits on. A small contact accepts a larger spreading resistance in exchange for spatial localization, which is the entire reason the point contacts to come will be made small and sharp. A contact with no localization to offer gains nothing from being small and loses resistance margin for it, so the base electrode is pushed to the opposite extreme on purpose: as close to the full area of the face as practical, driving that same resistance term down toward a value small enough not to matter next to everything else happening in the circuit.
The metal film itself adds a second resistance the electrode's area alone cannot fix. However the deposit is applied, it carries its own sheet resistance set by the film's resistivity divided by its thickness, and that resistivity is not simply the metal's bulk value. A sufficiently thin evaporated or plated layer scatters electrons against its own surfaces and internal grain boundaries far more than the same metal does in bulk, which raises the effective resistivity above the number a materials handbook would quote. A large-area, thin film can therefore still carry an unwanted series resistance even after the geometric spreading-resistance problem has been solved by area alone — area answers one resistance term, thickness has to answer the other, and neither substitutes for the other.
Choosing between the two metals is a choice about which resistance risk the device can tolerate, not a search for the objectively better metal. Copper's lower bulk resistivity makes it the more attractive choice on the day of deposition, if the only question were the number this step's resistance target has to meet immediately. But copper oxidizes, and an oxidizing base electrode is a base electrode whose conductivity degrades across the working life of the device, in a part of the circuit every signal the transistor ever passes has to cross. Gold starts from a higher bulk resistivity but holds that number steady, because it resists the oxidation that would otherwise erode copper's advantage over time. Neither answer is wrong; each accepts a different kind of resistance risk, one immediate and one deferred.
| Step | Process operation | Input | Output | Specification | Constraint |
|---|---|---|---|---|---|
| 6.1 | Select copper or gold based on the device's expected working life and resistance margin | Clean, bare germanium base face from Step 5 | Metal selection | Copper where lowest immediate resistance is prioritized; gold where stable resistance over time is prioritized | Selecting on bulk resistivity alone, ignoring oxidation behavior, can look correct on day one and degrade afterward |
| 6.2 | Set target film thickness well above the regime where thin-film resistivity departs from bulk | Metal selection from 5.1 | Target film thickness | Thickness sufficient that the deposited film's effective resistivity approaches the metal's bulk value | A film thin enough to scatter electrons at its own surfaces and grain boundaries can carry a resistance the area-driven design never anticipated |
| 6.3 | Deposit the metal across as much of the underside as the device geometry allows | Target thickness from 6.2 | As-deposited metal film | Coverage extending over the full usable area of the base face, consistent with Step 5's flatness specification for that face | Depositing over only part of the available area reintroduces the small-contact spreading resistance this step exists to avoid |
| 6.4 | Verify film adhesion and continuity across the deposited area | As-deposited film from 5.3 | Adhesion-verified electrode | No voids, delamination, or discontinuity across the base face that would locally reduce the effective contact area | A discontinuous film quietly shrinks the effective radius the spreading-resistance target assumed was the whole face |
| 6.5 | Measure the finished electrode's sheet and contact resistance against the design target | Adhesion-verified electrode from 5.4 | Measured base electrode resistance | Combined geometric and film resistance within the margin the circuit design allows for base series resistance | An electrode within film-thickness and area specification can still miss the target if the two resistance terms compound unfavorably |
| 6.6 | Record metal choice, thickness, and measured resistance against this specific body | Measured electrode from 5.5 | Documented base electrode | Metal, thickness, and resistance recorded against the body's provenance record from Step 2.6 | An undocumented electrode makes a later discrepancy in base resistance impossible to trace to a specific deposition run |
Step 6 does not add a new kind of contact to this device; it finishes the one the earlier steps had already been building toward without naming it. Step 2 bounded the block's lateral extent partly to keep base series resistance manageable. Step 5 ground and etched this specific face, on its own deliberately tight schedule, to a standard explicitly described as adequate for "a broad, low-resistance metal contact," and cleaned it to bare germanium for no other reason than to let a metal film bond to it dependably. This step is simply the one where all three of those earlier decisions cash out as an actual electrode, built by the opposite strategy from the point contacts still to come — not small and sharp, but as broad as the body allows, because this contact's only job is to disappear electrically, not to be studied.