Double Diffused Silicon Mesa 1956 Emitter Predeposition Phosphorus

# Emitter Predeposition: A Second Dopant, Opposite Polarity, Placed Inside a Field That Is Already Waiting For It

## 1. Why the Emitter Must Be Shallower and More Heavily Doped Than the Base It Sits Inside

This step introduces phosphorus, an N-type dopant, into the same surface the base was diffused into — but now the emitter has to be diffused shallower and doped more heavily than the base beneath it, or the drift field the previous two steps worked to create never gets a chance to do anything useful. The emitter's job is to inject a large density of minority carriers into the base at the emitter-base junction; the base's job, now that it carries a graded profile, is to sweep those carriers across to the collector by drift rather than by unassisted diffusion. For that handoff to work, the emitter-base junction has to sit near the high-concentration end of the base's gradient — the end the field already points toward — while remaining physically shallower than the full base width, leaving room between the two junctions for the field to act over:

where $N_E(0)$ is the emitter's surface concentration, $N_B(x_{jE})$ the base concentration at the emitter-base junction depth, and $N_B(x_{jC})$ the base concentration at the base-collector junction depth. Getting this doubly-ordered inequality right is why the emitter predeposition's dose and depth are chosen only after Step 5's measured base profile is in hand, not against a generic target.

## 2. Real Diagram: The Furnace Cycle, Now Introducing a Dopant of the Opposite Type

The predeposition furnace setup is mechanically similar to the base step, but the source is switched to a phosphorus-bearing gas or solid, and the temperature and time are deliberately lower and shorter than the base predeposition was, because this dose needs to stay confined to a much shallower depth than the base — the emitter must not be allowed to diffuse as deep as the base profile it is sitting inside of.

## 3. Why Getting the Opposite Dopant's Placement Wrong Here Breaks the Whole Device Differently Than in 1954

The 1954 diffused-base germanium process this project has already documented also diffused an emitter of opposite polarity into an existing base, so the predeposition mechanics are not unfamiliar. The difference is what a placement error costs: in that process, an emitter diffused slightly too deep mainly ate into the margin the base width needed to stay thin, degrading gain and frequency response gradually. Here, an emitter diffused too deep risks landing past the point in the base profile where the gradient is steep, injecting carriers into a region where the field this process spent two prior steps building has already weakened — silently discarding the drift advantage while still producing a transistor that passes a simple continuity check. The emitter's shallowness is not a generic good practice carried over from the earlier process; it is a specific requirement set by where, in this particular base's gradient, the field is actually doing its work.

Step 6 does not finish the junction structure; it banks a second, opposite-polarity dose at the one depth where the field already built into this base can make use of it.