# Implantation, activation and diffusion

Canonical: https://semiaifoundry.com/learn/atlas/implantation-anneal-diffusion/
Revision: 2026-10-07
Rights: https://semiaifoundry.com/ai-atlas/licenses.html

Fabrication / FEOL and MOL · Atlas v1.3.0

# Implantation, activation and diffusion

Introduces dopants and uses heat to repair damage and establish electrical activity.

[Explore in Atlas](https://semiaifoundry.com/ai-atlas/#implantation-anneal-diffusion)Nanometer device cross-sections; reference flow with alternatives

Learning objectiveDistinguish implantation dose, activated dopants and thermal redistribution.

## The mechanism

Ion implantation places dopant atoms into a material using an energetic ion beam and selected masking. Their distribution is not a sharp rectangular block: scattering, crystal direction and layer structure affect where ions stop. Implantation also disturbs the lattice. A subsequent anneal can repair damage and activate dopants, but heat also redistributes species and affects other materials already present. Chemical dopant dose, electrically active concentration and final junction depth are therefore different measurements. This lesson follows an implanted region through thermal processing and explains why a process flow must manage a cumulative thermal budget rather than treating every heat step independently.

1.  Define the intended doped region and protect surfaces that should not receive the implant.
2.  Deliver a controlled ion population while accounting for stopping, scattering and crystal channeling.
3.  Measure or model the as-implanted distribution and lattice damage.
4.  Apply a compatible anneal to promote repair and electrical activation.
5.  Verify the active electrical result after the relevant thermal history, not merely the commanded beam dose.

## Science and assumptions

### Diffusion length scale

$$
L\approx\sqrt{2Dt}
$$

The one-dimensional diffusion length scale is approximately the square root of twice diffusion coefficient times time.

Symbols and units

L

One-dimensional diffusion length scale

cm

D

Constant effective diffusion coefficient

cm²/s

t

Diffusion time

s

D is an effective diffusion coefficient in cm²/s, t time in seconds and L length in cm.

Assumptions One-dimensional constant-D diffusion scale. Real dopant diffusion can depend on concentration, defects and time-varying temperature.

## Engineering example

Synthetic comparison: quadrupling anneal time at unchanged effective D doubles the diffusion length scale. This says nothing by itself about whether activation or a product junction target improves.

## Distinctions and tradeoffs

Implanted dopant dose is not identical to electrically active carrier concentration.

More thermal exposure can improve activation or damage recovery while broadening a junction and stressing other interfaces.

Controlled junctions and low-resistance regions help logic transistors switch with predictable leakage and drive current.

## Specifications and interfaces

Implant dose

Ions/cm²

An areal delivered population, not active carriers/cm³.

Electrical result

Active profile and resistance

Depends on activation, redistribution and the completed material stack.

### Interfaces

Masked semiconductor Input

Target region, crystal orientation and incoming layers

Annealed doped region Output

Electrical activity, residual damage and redistributed profile

## Check your understanding

### Why can two wafers receiving the same implant dose have different sheet resistance?

1.  Activation, mobility and the final profile may differ
2.  Dose fixes every subsequent thermal effect
3.  All dopants remain at the top surface

Answer and reasoning

Activation, mobility and the final profile may differ

The delivered chemical population is only one input to the final conductive region.

## Evidence and scope

No implant energy, dopant dose, beam angle or anneal recipe is prescribed. Diffusion models are limited approximations.

Use the manufacturing investigation to trace compatible inputs, assembly and operating constraints; no current standalone calculator models this lesson’s full mechanism.

## Sources

-   [MIT 6.152J — Micro/Nano Processing Technology lecture notes](https://ocw.mit.edu/courses/6-152j-micro-nano-processing-technology-fall-2005/pages/lecture-notes/)
-   [ASML — How microchips are made](https://www.asml.com/en/technology/all-about-microchips/how-microchips-are-made)

[Published lesson data](https://semiaifoundry.com/ai-atlas/lesson-data/implantation-anneal-diffusion-39065174fb143722.json) · [Atlas releases and licensing](https://semiaifoundry.com/ai-atlas/licenses.html)

## Continue exploring

### Foundations

-   [Donor and acceptor doping](https://semiaifoundry.com/learn/atlas/doping/)
-   [P–n junction and depletion region](https://semiaifoundry.com/learn/atlas/pn-junction/)

### Related lessons

-   [Source/drain, spacers, contacts and MOL](https://semiaifoundry.com/learn/atlas/source-drain-contact-integration/)