Double Diffused Silicon Mesa 1956 Base Predeposition Boron

# Base Predeposition: Laying Down the Boron That Will Become a Graded Base, Not Yet a Flat One

## 1. Why Predeposition Is Deliberately Separated From Drive-In

This step introduces boron, a P-type dopant, into the clean silicon surface from the previous step — but it does so as a shallow, high-concentration layer confined almost entirely to the surface itself, not yet as the deep graded profile the finished base will need. That separation is intentional. Predeposition and drive-in are kept as two distinct steps specifically because they control different things: predeposition fixes the total quantity of dopant introduced (the dose), while the later drive-in step controls how deep and how gradually that fixed dose is redistributed into the crystal. Attempting to do both at once would leave no independent way to tune the base's final grading — and grading, not depth alone, is what this device's drift field depends on. The boron is introduced from a vapor or solid source at a temperature high enough to let it diffuse a short distance into the surface, producing a concentration profile that is approximately constant very near the surface and falls off steeply within a fraction of a micron:

$$N(x,t) = N_s \,\text{erfc}\!\left(\frac{x}{2\sqrt{Dt}}\right)$$

where $N_s$ is the surface concentration set by the boron source's solid solubility limit at the predeposition temperature, $D$ is boron's diffusion coefficient at that temperature, and $t$ the short predeposition time — short enough that the dose is fixed before any meaningful depth is reached.

Predeposition Fixes a Dose; It Does Not Yet Shape a Profile that shaping is deliberately deferred to the next step, drive-in BORON CONCENTRATION, SURFACE TO DEPTH depth below surface, x Nₛ — surface concentration, set by solid solubility at this temperature falls to near zero within a shallow depth — this is dose, confined, not yet graded deep total dose Q = ∫ N(x) dx is fixed here; the shape of N(x) below the surface is fixed later, by drive-in separating these two controls is what lets the next step engineer a graded, not flat, base profile on purpose

## 2. Real Diagram: The Predeposition Furnace Setup

Physically, predeposition for this era's process is carried out in a resistance-heated diffusion furnace, with the boron source introduced either as a solid boron nitride wafer placed near the silicon, or as a boron-bearing gas flowed over the wafer surface, at a temperature typically in the 900–1000°C range — hot enough to let boron diffuse into the surface at a controlled, repeatable rate without yet driving it deep into the bulk.

Resistance-Heated Furnace, Boron Source, Controlled Short Exposure temperature and time are chosen jointly to fix Nₛ and keep the dose shallow furnace tube, 900–1000°C silicon wafer, cleaned surface up boron source (BN solid or gas-phase) boron atoms enter the exposed silicon surface time held short: dose fixed before depth grows large this furnace cycle's only job is to set the dose Q; depth and grading belong to the next step getting Nₛ repeatable here is what lets drive-in produce a predictable drift field later

## 3. Why This Step Is the First Real Departure From the 1954 Diffused-Base Process

The 1954 diffused-base germanium process this project has already documented also used a predeposition-then-drive-in sequence for its base diffusion, so the furnace mechanics here are not new. What is new is the intent behind the numbers chosen at this step: the 1954 process picked a predeposition dose and temperature aimed at producing a base that, after drive-in, would end up nearly uniformly doped — grading was an accepted side effect, not a target. Here, the boron dose and predeposition temperature are chosen with the opposite intent: enough total dose, placed shallowly enough, that the drive-in step to follow can stretch it into a deliberately steep concentration gradient rather than smoothing it into near-uniformity. The furnace step looks the same; the number written on the furnace log — the dose it is calibrated to deliver — is chosen for a different downstream purpose entirely.

Step 3 does not create the base or its field; it only banks the exact quantity of boron the next step will need in order to shape a gradient steep enough to matter.

Take double diffused silicon mesa 1956 base predeposition boron 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