Contact Mask Aligner

Mask aligners predate projection scanners and still cover jobs scanners are overkill for — MEMS, power devices, compound semiconductors, thick-resist layers — by pressing a patterned mask close to or directly against the wafer instead of projecting a reduced image through a lens. The three main variants — contact, proximity, and double-sided — differ in exactly how close that mask gets to the wafer and whether alignment happens from one side or both.

Mask and wafer loaded
Coarse pre-alignment before fine registration
Aligner type decisionResolution, mask wear, and wafer-side access drive the choice
Contact aligner
Mask touches resist — best resolution, most mask wear
Proximity aligner
Small gap (10-50 µm) — less resolution, no mask contact
Double-sided aligner
Front and back masks aligned to each other through the wafer
Exposure
Flood UV exposure through the mask, full wafer at once

Contact mode gets the sharpest pattern because diffraction has nowhere to happen. With the mask physically touching the resist, there's essentially no gap for the exposing light to diffract across before it reaches the film, which is why contact aligners can resolve features a proximity aligner of the same wavelength simply can't. The price is paid in mask life: every contact exposure risks picking up resist residue or scratching the mask, so contact masks wear out and need replacement far sooner than a proximity mask that never touches anything.

Mask Aligner Types — Gap Sets the Tradeoff contact, proximity, and double-sided differ in mask-to-wafer distance and access Contact mask touches resist sharpest pattern fastest mask wear Proximity 10-50 µm air gap no mask contact diffraction softens edges Double-sided front + back mask aligned through wafer e.g. MEMS through-features Resolution: contact > proximity · Mask life: proximity > contact · Access: double-sided is the only one that patterns both faces all three flood-expose the full wafer at once — no stepping, no scanning

Double-sided aligners exist because some devices need features that line up through the wafer, not just on top of it. MEMS pressure sensors, through-wafer vias, and backside-release structures all need a backside pattern registered precisely to a frontside pattern that was already defined — a job neither a contact nor a proximity aligner can do alone, since both only ever see one face of the wafer at a time. A double-sided aligner solves this with infrared or optical alignment systems that can see through the wafer itself to register both masks to a common reference.

Mask aligners never went away even after projection scanners took over leading-edge logic, because contact, proximity, and double-sided exposure each solve a specific access or resolution problem that a scanner's reduction-lens architecture isn't built to handle — and MEMS, power, and compound-semiconductor fabs still run them in volume today.

Take contact mask aligner 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