Diffused Base 1954 Perform Base Drive in

# Perform Base Drive-In: Why Temperature, Not Time, Is This Step's Real Precision Problem

This step runs the sealed anneal that drives the base dopant's fixed dose inward to the depth that will become the base-collector junction, and the overview article's own depth formula, $x_j \propto \sqrt{D t_2}$ , hides a sensitivity this step's operators have to respect in practice: $$D$$ depends on temperature exponentially, while depth only depends on time through a square root. A small timing error in $$t_2$$ barely moves the resulting junction depth, softened twice over by the square root. A small temperature error moves $$D$$ itself exponentially, which means this step's real precision challenge is holding furnace temperature steady, not timing the anneal to the second.

## 1. An Exponential Term Dominates a Square-Root Term Every Time They Compete

D(T) = D_0 \exp\!\left(-\frac{E_a}{k_B T}\right), \qquad x_j \propto \sqrt{D(T)\,t_2}

Because diffusivity $$D(T)$$ follows an Arrhenius relationship with activation energy $$E_a$$ , a temperature deviation of even a few degrees at typical drive-in temperatures changes $$D$$ by a percentage far larger than the same fractional deviation in $$t_2$$ would change the time term — and because $$x_j$$ depends on the product $$D t_2$$ only through a square root, the exponential sensitivity in $$D$$ simply dominates. This step can tolerate a timer that drifts by a noticeable margin far more easily than it can tolerate a furnace zone that runs a few degrees hot.

## 2. Real Diagram: The Same Percentage Error, Two Very Different Consequences

## 3. Furnace Zone Uniformity Reproduces This Same Sensitivity Across the Wafer, Not Just Across Runs

x_j(z) \propto \sqrt{D(T(z))\, t_2}

If this step's furnace maintains slightly different temperatures $$T(z)$$ at different positions along the tube, every wafer at a hotter zone sees a measurably deeper junction than a wafer at a cooler one, run under the identical same nominal recipe and identical $$t_2$$ — the same exponential sensitivity that makes run-to-run temperature control critical also makes zone-to-zone uniformity within a single run critical, for exactly the same underlying reason. This step cannot treat temperature uniformity as a secondary concern behind achieving the correct average temperature; a furnace that averages correctly but varies across its length still produces a batch of devices with base widths that vary device to device for a reason this step alone controls.

## Perform Base Drive-In's Place in the Process Lineage

Performing base drive-in is step fifteen of Bell Labs' twenty-six-step diffused-base manufacturing sequence — immediately after the wafer was sealed against further dopant supply, and before the resulting base-collector junction depth is ever verified. It is the step where the overview article's clean depth formula meets a practical reality this process's operators have to respect: temperature control, not timing precision, is what actually governs how closely the finished junction lands where the recipe intended, both run to run and across a single furnace's own zone-to-zone uniformity. Step sixteen, verifying the resulting base-collector junction depth, is the first direct measurement of whether this step's temperature discipline actually held.

Diffused Base 1954 Perform Base Drive in

Take diffused base 1954 perform base drive in further