MOSFET Scaling 1970s Trace Short Channel Threshold Rolloff

# MOSFET Scaling (1970s): Step 12 — Tracing Short-Channel Threshold Rolloff

The first clearly non-ideal effect scaled devices ran into. As channel lengths pushed below roughly 2-3 micrometers in the mid-to-late 1970s, measured threshold voltage began to drop below the long-channel value predicted by the step 9 equation, falling further as channel length shrank further — a phenomenon called threshold-voltage rolloff. Step 12 covers the mechanism and why it meant the long-channel V_t model used through most of this series needed a correction once channel length got short enough.

## 1. Why Source and Drain Depletion Regions Start to Matter

The long-channel V_t equation (step 9) assumes the gate alone controls all the depletion charge under the channel. In a short device, the source and drain junctions' own depletion regions extend laterally and start sharing responsibility for depleting part of the channel region that the gate would otherwise have had to deplete alone — meaning the gate needs less voltage to finish the job, and V_t drops. This is the same lateral-depletion encroachment mechanism introduced in step 8's junction-depth discussion, now viewed from the threshold-voltage side rather than the leakage side.

## 2. Real Diagram: Threshold Rolloff vs. Channel Length

Threshold Voltage Rolloff vs. Channel Length measured Vt falls below the long-channel prediction once L drops below roughly 2-3 µm channel length (long → short, right to left) threshold voltage (normalized) long-channel model (constant) measured Vt, short-channel devices the gap between the two curves is the rolloff this step attributes to source/drain depletion sharing

## 3. Why Rolloff Is a Design Problem, Not Just a Modeling Nuisance

If V_t drops as channel length shrinks, and real fabricated channel lengths always have some lot-to-lot and die-to-die variation around the nominal target, then the shortest devices in a distribution end up with the lowest V_t — and if V_t drops far enough, those devices stop behaving like a controlled switch and start leaking current even in the nominal 'off' state. Rolloff therefore couples channel-length control tightness directly to threshold-voltage control tightness in a way that did not exist at longer channel lengths, raising the stakes on the lithography and etch control discussed in step 5.

## 4. The Partial Mitigation: Heavier, More Localized Channel Doping

Increasing substrate doping specifically in the channel region (sometimes via a separate, deeper implant distinct from the shallow Vt-adjust implant of step 10) increases the gate's share of depletion control relative to the junctions', partially counteracting rolloff. This is a genuine partial fix, not a full one — it buys back threshold stability at the cost of yet more doping, which step 13 shows has its own scaling consequence through mobility degradation and velocity saturation.

## 5. Companion Diagram: Stage Sequence and Trend Summary

MOSFET Scaling (1970s): Step 12 stage sequence and schematic outcome trend for this step Why Source and Drai… 1 Real Diagram: Thres… 2 Why Rolloff Is a De… 3 The Partial Mitigat… 4 sequence progression outcome trend (schematic) companion diagram: stage sequence and schematic trend for this step

## Tracing Short-Channel Threshold Rolloff's Place in the Process Lineage

This is step 12 of 20 in the MOSFET Scaling (1970s) series, tracing planar NMOS scaling from early 1970s field pressure through the fixed-voltage-scaling break that opens the adjacent HMOS (1978) series.

Take mosfet scaling 1970s trace short channel threshold rolloff 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