Double Diffused Silicon Mesa 1956 Base Ohmic Contact

# Base Ohmic Contact Metallization: Attaching a Wire to a Region That Is Not Flat Electrically

## 1. Why a Contact to a Graded Region Needs More Care Than a Contact to a Uniform One

This step deposits metal onto the base region to form the electrical contact a lead will later be attached to, and the placement and quality of that metal-silicon interface matters more here than it would on a uniformly doped base, because the base this process built is not electrically uniform — it is graded, with concentration falling steeply from the emitter side toward the collector side. An ohmic contact needs to sit on a region of high enough dopant concentration that the metal-silicon junction behaves resistively rather than like a second, unwanted rectifying diode. Because this base's concentration falls with depth and with lateral position relative to where the drive-in dose was concentrated, the contact has to land specifically where the base remains heavily doped enough to guarantee that behavior:

$$R_c \propto \exp\!\left(\frac{2\sqrt{\phi_B \varepsilon_s m^{*}}}{\hbar\sqrt{N_B(x_c)}}\right)$$

where $R_c$ is the contact resistance, $\phi_B$ the metal-semiconductor barrier height, and $N_B(x_c)$ the base doping concentration at the specific contact location $x_c$ — a quantity that, on this device, depends on exactly where the contact lands on a surface that is not doped the same way everywhere. A contact placed carelessly on a lightly doped region of the base can add series resistance the rest of the device's careful field engineering cannot compensate for.

A Contact Resistance That Depends on Exactly Where It Lands the base is not uniformly doped, so this contact's quality depends on placement, not just metal choice base Nᵈ(x), falling toward collector side contact placed on heavily doped region low Rᴮ, ohmic as intended contact misplaced over lightly doped region high Rᴮ, undermines the whole device Rᴮ depends on Nᵈ(xᴮ) — the same gradient that creates the drift field also creates a placement constraint for this step

## 2. Real Diagram: Metal Deposited Through the Pattern the Mask Steps Already Defined

The metal for this contact is deposited by evaporation through a stencil or pattern that restricts it to the intended base contact area on the mesa top, away from both the emitter region and the mesa's passivated sidewall, landing only on the portion of exposed base surface the earlier steps left available for this purpose.

Metal Lands Only on the Base, Nowhere Else on the Mesa Top emitter contact comes next, as its own separate step, deliberately not combined with this one mesa top, from Steps 1–11 base contact lands on base ring, around emitter emitter region, untouched this step placement respects every boundary the mask, etch, and passivation steps already established

## 3. Why This Step's Placement Tolerance Is Tighter Than the 1954 Diffused-Base Process's Equivalent Step

The 1954 diffused-base germanium process this project has already documented also metallized a base contact onto a diffused base region, so the deposition mechanics described here are not new. The tolerance for where that contact can land is tighter in this process specifically because of the graded base: in a uniformly doped base, contact resistance barely depends on lateral placement, since $N_B$ is roughly the same everywhere; here, $N_B(x_c)$ varies meaningfully across the mesa top, and a contact nudged toward the collector-facing edge of the base ring lands on a region doped noticeably lighter than one nudged toward the emitter-facing edge. The alignment precision this step demands is therefore a direct, specific consequence of the drift field Steps 3 through 5 built — a requirement the 1954 process's otherwise similar contact step never had to satisfy.

Step 12 does not create new electrical behavior in the device; it decides how much of the base's carefully graded conductivity actually reaches the outside world through the lead this contact will eventually carry.

Take double diffused silicon mesa 1956 base ohmic contact 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