Photolithography is an integral part of the manufacture of integrated circuits. It allows one to etch certain parts of a substrate (such as a doped silicon wafer) while leaving others intact. Here is a brief procedure for a photolithography step:

  1. Photoresist (positive or negative) is spun onto the wafer to be processed at a set rotational velocity to insure an even coating across the surface of said wafer.
  2. The wafer is placed in a pre-bake oven to harden the photoresist.

  3. Next, the wafer is secured into a mask aligner and a test pattern is used to aid in aligning the wafer so that successive masks overlap perfectly.

  4. A UV light is used to expose the image from the mask onto the photoresist-covered wafer for a pre-determined amount of time. Depending on the type of photoresist used, this light causes the photoresist to either harden or become more soluble when immersed in a developing solution.

  5. The wafer is placed in developer, causing the exposed photoresist to either harden or strip away.

  6. A buffered oxide etch is then used to etch exposed areas of the wafer. This can be used to open up areas which will then be doped by diffusion, creating n-wells or p-wells. Likewise, these areas could undergo a metallization process to create contacts. HF, or hydrofluoric acid, which is present in a buffered oxide solution, will also do a number to your bones about 6-8 hours after contact. The time used for this etching need to be controlled precisely to prevent errors such as overcut/undercut.

  7. Finally, a warm stripper solution is used to remove the remnants of the photoresist.

Kill all Webster circular definitions!

Pho`to*li*thog"ra*phy (?), n.

The art or process of producing photolithographs.

<-- The process by which the image of a pattern is transferred photographically to a sensitive surface, and the surface subsequently etched; used for printing or in the production of integrated circuits. -->

 

© Webster 1913.

Log in or register to write something here or to contact authors.